Background: The surgical management of large vestibular schwannomas (VS) yields a high risk for the facial and cochlear nerve functions. Gamma Knife radiosurgery (GKRS) allows optimal functional results in small- and medium-size VS, but cannot be used upfront in large VS because of the high rate of volume-related side effects.

Methods: To develop of a new treatment paradigm of combined approach with microsurgery and GKS, aiming at optimal functional outcome for the facial and cochlear nerves in patients with large VS (i.e. Koos grade IV). To perform planned subtotal resection followed by GKRS in a consecutive a series of patients with large VS. All patients were operated via a retrosigmoid approach in the lateral decubitus position with intraoperative neuromonitoring (IOM). Data pertaining to patient characteristics, surgical and dosimetric features and outcome were collected prospectively at time of treatment and during the follow-up course.

Results: A consecutive a series of 40 patients was treated between 2010 and January 2017. The mean presurgical tumor volume was 12 cm³ (1.47-34.9). All cases had normal facial nerve function (HB I) before surgery, except for one who was in HB IV. Postoperative status showed normal facial nerve function (House-Brackmann grade I) in all patients. In a subgroup of 22 patients in which cochlear nerve preservation was attempted at surgery (patients with residual hearing before surgery), 21 of them (95.4%) retained residual hearing. Among them, 16 patients had normal hearing (Gardner-Robertson class 1) before surgery, and 13 (81.2%) retained normal hearing after surgery. The mean duration between surgery and GKRS was 6.2 months (4-13.9, median 6 months). The mean tumor volume at the time of GKRS was 3.6 cm³ (0.5-12.8), which corresponds to a mean residual volume of 31.4% (range 3.6-50.2) of the pre-operative volume. There was a tendency towards larger postoperative residual volume in patients with attempt to cochlear nerve preservation. The mean marginal prescription dose for GKS was 11.9 Gy (range 11-12, median 12 Gy). Following GKRS, there were no new neurological deficits, with facial and hearing functions remaining identical to that after surgery. The mean follow-up after surgery was 31 months (range 3-72).

Conclusion: Our data suggest that the management of large VS with planned subtotal resection followed by GKRS may yield an excellent clinical outcome with respect to retaining facial and cochlear nerve functions. Our results with this approach are comparable to those obtained with GKRS alone in small- and medium-size VS.

Objective:Stereotactic radiosurgery (SRS) has emerged as a principal alternative to microsurgical resection for small and medium-sized cavernous sinus hemangiomas (CSH).  However, large CSH management remains a challenge for the neurosurgeon. Although preliminary studies have demonstrated that multi-fraction SRS for large CSHs is both safe and effective, the number of patients in the clinical series and the length of follow-up were limited. This study aimed to clarify the long-term outcomes of multi-fraction SRS for the treatment of large CSHs with minimum 3-year follow-up.

Methods: Between January 2008 and June 2012, forty patients with large CSHs underwent multi-fraction SRS. Thirty-seven patients underwent radiosurgery as primary treatment modality based on clinical and imaging criteria. Three patients had previous operation. The mean volume of the CSHs was 48.6 ± 12.1 cm3 (range, 11.3-140.1 cm3). Multi-fraction SRS was delivered in 2 to 4 fractions (median, 3 fractions). The median marginal dose was 21Gy (range, 17.8–22Gy) prescribed to a median 64% isodose line. Dose sculpting technique is used to focus high radiation dose within the tumors and the base of the tumors.

Results:  The median follow-up period was 53 months (range, 36–90 months). Tumor control was achieved in all patients (100%) during the follow-up period. At 12 months after multi-fraction SRS, MRI revealed a mean of 70% tumor volume reduction (range, 40%-99%).The last MRI showed a mean of 85% tumor volume reduction. At the last follow-up, one patient with marked tumor shrinkage died of stroke, unrelated to CSH. All patients demonstrated improvements in their neurological deficits post radiosurgery. The only observed side effects were mild headache during treatment in 4 patients with giant tumor and mild, asymptomatic brain edema in the follow-up MRI. One patient reported seizures which were controlled after antiepileptic drugs administration. There were no radiation induced neuropathies during the follow-up period.

Conclusion: Our extended experience confirms that multi-fraction SRS is a safe and an effective management strategy for large CSHs. Considering the risk involved in microsurgery, multi-fraction SRS may serve as the primary treatment option for patients with large CSHs.

Purpose:  To evaluate the treatment modalities utilized in the management of patients with acoustic neuroma (AN) and the potential disparities in the management of AN by race.

Materials and Methods:  The NCDB is a comprehensive national database that captures approximately 70% of  newly diagnosed cancer patients in the USA.  Patients diagnosed with AN were identified in the 2013 Participant Use File (PUF) encompassing years 2004-2013, using ICD-O-3 codes.  The utilization of observation, surgery, radiation therapy, and types of radiation therapy (stereotactic (SRS) vs standard-fractionated RT (SFRT)) were evaluated.  Race categories were identified as White, Black, American Indian (AI), Asian, or Hawaiian/Polynesian (HP).  Categorical data were summarized using descriptive statistics.    Chi-square and ANOVA univariate analysis (UVA) and multinomial logistic regression multivariate analysis (MVA) was performed for the types of treatments utilized.  This study was reviewed by our institutional IRB. 

Results: A total of 40,292 patients meeting the ICD-O-3 criteria for AN were identified in the 2013 PUF data file (90.7% white, 5.3% black, 0.3% AI, 3.6% Asian, and 0.2% HP).  Of these, 8367 patients did not undergo surgery or radiation therapy (RT), 22,926 patients underwent surgical resection and 9,727 patients underwent radiation therapy (with some patients receiving combined modality treatment).  Of the patients receiving RT, 89.1% underwent SRS (55.0% gamma knife, 8.9% Linac based and 25.2% SRS NOS) and 10.9% underwent standard fractionated radiation therapy.    Treatments utilized by race (for Observation, Surgery, SRS, SFRT, Combination treatment) were:  White: 20.8%, 55.1%, 20.0%,  2.3%, 1.8%; Black: 24.0%, 53.8%, 17.5%, 2.1%, 2.6%; American Indian: 11.4%, 61.0%, 21.0%, 1.9%, 4.8%; Asian: 18.6%, 60.1%, 15.6%, 2.5%, 3.3%;  Hawaiian/Polynesian: 20.8%, 55.2%, 19.7%, 2.3%, 1.9% (p < 0.001).  In general, black patients had the highest rate of observation, AI patients had the highest proportion of surgical resection and the highest rate of SRS treatment and Asian patients had the highest rate of combined surgery and SRS treatment.  Factors significantly associated with treatment utilization on UVA and MVA included age, tumor size, year of diagnosis, CDCC score, crowfly distance, and the geographic region of the facility location. Race was significant for the type of treatment utilized on UVA but not on MVA.

Conclusion:  The statistical analyses in this study showed differences in the types of treatments utilized in the management of AN based on patients age, race, and tumor size among other factors.  Further studies are needed to better evaluate potential disparities in treatments utilized for patients with AN.

Introduction. Stereotactic radiosurgery (SRS) is an important treatment option for patients with cavernous sinus meningiomas. The purpose of clinical study was retrospective analysis of the efficiency of linac SRS for cavernous sinus meningiomas with using lower prescription doses (≤12.5 Gy).

Materials and methods. Linear accelerator «Trilogy + BrainLab» stereotactic radiosurgery was performed in 30 patients (8 males and 22 females; mean age - 52 years) with cavernous sinus meningiomas. 23 patients (76.7%) received SRS alone, and 7 patients (23.3%) had undergone surgery before SRS.

Results. The marginal doses for the tumors were 11 Gy to 12.5 Gy (median, 12.1 Gy). Tumor volume ranged from 2.8 cc to 20.9 cc (median, 9.1 cc). Median tumor coverage 95.5%. Median follow-up of patients was 10 months (range, 2-39 months).
20 patients (66.7%) before SRS had visual disturbances and 4 patients (13.3%) had facial pain or numbness. Six (20%) patients demonstrated improvement in their neurological condition after SRS. Other 24 (80%) patients also had no worsening of their neurological status. None of the patients experienced optic neuropathy caused by radiation injury or any new neurological deficits after radiosurgery. Follow-up images showed a reduction in tumor size in 14 patients (46.7%), no further growth in 16 (53.3%) cases. The overall tumor control rates 100%. 
Conclusions. SRS − is an effective method of treatment for сavernous sinus meningiomas, providing high rate of tumor control and ensuring safety of patient's neurological status. SRS with using lower marginal doses (≤12.5 Gy), allows reducing frequency and severity of both transient and permanent neuropathy. Further study is needed to determine the tumor control and complication rates 5 years or more after low-dose linac radiosurgery.

Introduction.  This study is dedicated to actual scientific problem of improvement results of stereotactic radiosurgery (SRS) and neuroimaging monitoring of vestibular schwannomas.

Materials and methods. Linear accelerator «Trilogy + BrainLab» stereotactic radiosurgery was performed in 143 patients (53 males and 90 females; mean age - 49,6 years) with 145 vestibular schwannomas, 38 patients had undergone surgery before SRS. It was used original method of SRS. The maximum follow-up time 68 months.

Results. The marginal doses for the tumors were 10 Gy to 14,2 Gy (median, 12.5 Gy). Median tumor volume was 4,0 cc.

The  vestibular schwannomas response was seen in first 2 years of follow-up. It was proved, that invention of original method of SRS treatment Dyn Arc + IMRT significantly increased its efficacy (p=0,022).

The feasability of neuroimaging data application and dose regimen as predictors of response to radiosurgical treatment in patients with VS was also proved. New predictors of local control were investigated. It was established, that data of perfusion CT (gap(MTT) х gap(CBV)) are the strongest predictive factor influencing median tumor growth control ratio.

Conclusions. SRS with using original method of treatment is providing high rate of tumor control - 96,3% and ensuring safety of patient's. New mathematical model of response prognosis of vestibular schwannomas on SRS treatment was carried out. It allows to calculate the optimal time periods for postradiational neuroimaging monitoring during which decreases the frequency of imaging procedures during follow-up.

Object: The purpose of this study is to evaluate the efficacy of gamma knife radiosurgery (GKRS) in the management of remnant pituitary adenoma (PA) underwent surgical treatment. Materials and Methods: Data of 25 patients who were treated with GKRS were reviewed between June 2006 and December 2012. All included patients underwent surgical resection via trans-sphenoidal approach or craniotomy. Endocrinological, ophthalmological, and radiological responses were investigated. The duration of follow-up varied from 48 to 96 months. To analyze radiological response we did rating of tumor control by 6 grade (-1: enlarged, 0: no volume change, 1:1-25% of volume reduction, 2: 25-50%, 3: 50-90%, 4:90-100%). Results: Average marginal dose was 15.76Gy (range of 9-22Gy). Fifty percent isodose line was used. Nine patients (36%) among 25 patients showed that tumor was disappeared (grade 4). Four patients (16%) showed grade 3. Eleven patients (44%) showed between grade 0 and 2. One patient showed regrowth of tumor (grade -1). There was no significant difference of controllability between non-hormone-secreting pituitary adenoma and hormone-secreting pituitary adenoma. There was no aggravation of endocrniological or ophthalmological symptom during follow-up period. The limitations of this study were that number of patient was small and there was no control group. Conclusion: GKRS can achieve tumor control in remnant PA without adverse effect by relatively low dose selection (non-hormone-secreting PA, average 14.7Gy; hormone-secreting PA, average 16Gy) after surgery.

Object: The purpose of this study is to evaluate the long-term efficacy of gamma knife radiosurgery (GKRS) in the management of intracanalicular vestibular schwannoma (iVS).  Materials and Methods: Data of five patients who underwent GKRS for iVS were reviewed between February 2008 and December 2011. Radiological responses, auditory functions and adverse effect were investigated. Planned marginal dose was 12Gy (50% isodose line). The duration of follow-up was more than 5 years. Results: At last follow-up date two patients who visited hospital due to dizziness was improved and hearing functions were in normal range after GKRS. In other three patients who suffered from hearing impairment their hearing functions were not improved after GKRS. Two patients (40%) out of 5 patients were showed that tumors were shrunken and three patients (60%) showed stable status on final brain MRI. In aspect of adverse effects all patients complained different intensity of tinnitus after GKRS. The limitations of this study were that number of patient was small and there was no control group. Conclusion: GKRS can achieve good tumor control in iVS. However, clinicians have to consider a possibility of occurrence or aggravation of tinnitus.

Purpose: Evaluate the efficacy of linear accelerator radiosurgical treatment of vestibular schwannomas (VS) based on the observed changes in the tumor volume.

Methods: 121 patients with VS were treated with linear accelerator (LINAC)-based radiosurgery (SRS) in the period from 2006 to 2014. Average patient age was 57,5 years with an average volume of VS tumors of 2ml. Mean marginal dose of the irradiation comprised 12,3 Gy.

In the first 2 years after SRS patients underwent an MRI examination with a volumetric assessment of the VS (3-6 months, 1 year, and 2 years post SRS). Based on the observed volumetric changes patients were stratified in the following groups: volumetric "Progression of VS“, “Stabilization“, and „Partial regression“.

Results: After 3-6 months post SRS volumetric progression of VS (range 3,2 - 150%) was observed in 56,4% of cases (57 patients). Unchanged VS volume was observed in 39,6% of cases (40 patients), while a partial regressions ( 4 - 29% of the pre-SRS volume) - in 4% of patients.

One year follow-up post SRS: 9,4% (8 patients) had a VS volumetric progression (ranging 4,8 -  142,1% in comparison to the previous MRI). Stabilization of tumor growth was observed in 35,5% of cases (30 patients), while in 55,3% of all cases (47 patients) a partial VS regression (ranging  2,9 - 70% compared to the previous MRI results) was confirmed.

A 2-year follow-up after the SRS revealed a volumetric stabilization in 49,4% of cases (39 patients), partial regression of the tumor volume (ranging 9,1 - 80%) - in 46,8% (37 patients), and an increase in the tumor volume (ranging 10 - 71,4% of the previously observed volume) was documented in 3,8% of cases (3 patients).

Conclusions:  1) Most changes in VS volumes were observed in the first year after SRS. An increase in  VS volume accompanied by regressive structural changes was believed to be a reaction of the tumor tissue to irradiation and perceived as “pseudo-progression”.

2) In the 2-year follow-up VS volume stabilization or its partial regression were confirmed in 96,2% of patients, in contrast to a volume progression in 3,8% of cases. A volumetric increase following a 2-year period after SRS was perceived as a true tumor progression.

3) The efficacy of the treatment should not be assessed earlier than 2 years after the irradiation.

4) LINAC-based SRS is a highly effective method of treatment for patients with vestibular schwannomas.

Objective: Stereotactic radiosurgery (SRS) is the method of precise delivery of high doses of radiation for the benign cranial nerves tumors with limited radiation exposure to surrounding healthy tissue.  The CyberKnife M6™ delivers SRS, with extreme accuracy over a minimum number of treatments, reducing side effects and preserving patients' quality of life.

Materials and Methods: 18 patients had CyberKnife M6™ radiosurgery (SRS) treatment in SRC «Sigulda» from December 2015 till January 2017 for vestibulocochlear nerve schwannomas. Each patient receives from 1 to 3 fractions of radiosurgical treatment. Most patients (12 patients) had 1 fraction stereotactic radiosurgery with a single dose range from 12 Gy (8 patients) to 14 Gy (4 patients). Six patients with tumor recurrence after surgery had 3 fractions SRS with total dose 18 Gy. Treatment was with 130-180 non-complanar beams using 1-2-collimators (7,5-10mm), isodose curve was from 67 - 86%, coefficient conformity was 1.15-1.25, homogeneity factor was 1,1-1,3. All patients received a standardized neuro-ophtalmological examination, pure-tone audiometry. 8 patients received video head impulse test (vHIT) for vestibular function assessment before SRS. 3 patients in this group had medial reduction of vestibular function.   

Results: At this stage of research, we can evaluate the safety of the treatment and the absence of early complications after SRS radiosurgery. There were not acute post radiation complications, no severe postradiation neurological deficit and no local progression for 6 month. Any patient had postradiation n.facialis deficit. Moderate hearing decrease was detected in 8 patients after SRS. 4 patients in group which received vHIT before SRS, had moderate deterioration of vestibular function at the first follow-up 3 month after SRS.   

Conclusions: Stereotactic radiosurgery is a highly effective and safe treatment for patients with vestibulocochlear nerve schwannomas, which allows to avoid the risk of severe neurological complications and keep a high level of quality of life after treatment. 

Objective:

Microsurgical resection of cavernous sinus meningiomas (CSM) is challenging with a high percentage of recurrence due to incomplete resection (>10% reported recurrence rate after Simpson Grade II and >20% after Grade III resections) and/or permanent, partly severe cranial nerve deficits.  Stereotactic radiosurgery (SRS) has evolved as alternative first-line treatment for SCM. Here, we report about the long term clinical and radiological follow-up of an unique cohort of patients with CSM treated with LINAC based SRS.

Methods:

In this single center retrospective analysis we included all patients with SCM who underwent single fraction LINAC SRS between 1993 and 2012 and had a minimum follow-up of 3 months. We evaluated tumor control (no further intervention needed) by the Kaplan-Meier method. Additionally, patient data were analyzed in terms of clinical symptom control and incidence of complications or unexpected side effects rated by Common Terminology Criteria for Adverse Events (CTCAE; v4.03).

Results:

82 patients with 83 tumors (f/m =62/20, median age 53 ± 11, range 33-81 years) were identified. Mean tumor volume was 5.8 ± 3.5 cm³ (range, 0.6-16 cm³), the mean marginal dose was 12 ± 2 Gy (range, 7.0-18.75 Gy) at isodose levels of 64 ± 17% (range, 30-85%). Median follow-up (FU) was 57 months (range, 3-226 months).

Tumor control was 100% after 6 and 12 months, 97% after 5-years and 94% after 10 years. Symptoms prior to SRS remained stable in 88% (n=73), improved in 3.6% (n=3) and deteriorated in 2.4% (n=2) at last follow-up. Four patients (4.8%) report about adverse events CTCAE grade 1 (headache n=2, somnolence n=1, trigeminus disorder n=1).

Conclusion:

SRS for SCM provides reliable long term tumor control without considerable permanent side effects. Thus, SRS should be taken into account when counselling patients harbouring CSM.

Objective:

Tumor control and functional hearing preservation are the main goals to be reached in the treatment of acoustic neuroma (AN). Stereotactic radiosurgery (SRS) has evolved as alternative first-line treatment for AN. Here, we report the clinical and radiological long term follow-up of patients with AN treated with LINAC or Cyberknife® based SRS.

Methods:

In this single center retrospective analysis we consecutively included all patients with AN who underwent single fraction LINAC or Cyberknife® based radiosurgery between 1991 and 2015 with a minimum follow-up of 2 months. Patient data were analyzed in terms of tumor control (defined as no further intervention required), preservation of functional hearing and incidence of early and late treatment related complications (rated by using the Common Terminology Criteria for Adverse Events, CTCAE; v4.03).

Results:

335 patients (f:m = 176:159, median age = 58 years) were treated either with LINAC (n=270) or Cyberknife® (n=65) -based SRS. Median tumor volume was 1.1ml ± 2.6ml (0.1-23.7). The median radiation dose was 12 Gy (range 11-20) at a median isodose level of 72 % (31.9-86.21). Median follow-up was 30 months (2-224 months). According to the Koos classification we identified 53 (15.8%) Koos I, 180 (53.7%) Koos II, 42 (12.5%) Koos III and 60 (17.9%) Koos IV tumors. 70 patients had surgery prior to SRS.

The 2-, 5- and 10-year tumor control rate was 99%, 89% and 88%, respectively. The Kaplan-Meier estimated an objective hearing preservation rate of 89% after 12 months, 80% after 24 months and 57% after 50 months. Treatment failure was observed in 6.3% (n=21). Re-treatment included re-radiation in 43% (n=9/21) and microsurgery in 57% (n=12/21). De novo permanent adverse events (CTCAE grade I and II) were objectively observed in 8, 9% (n=30) cases.

Conclusion:

SRS for AN leads to very good long term tumor control with a considerable rate of hearing preservation and a low rate of permanent side effects. SRS can be proposed as save and effective first line treatment alternative to - or in combination with - microsurgical resection. 

Objective: Gamma knife radiosurgery has become an important tool in the management of pituitary adenomas. We have reported our treatment results, managed with this novel technology at Department of Neurosurgery, Ibn Sina Hospital, Kuwait.

Subjects and Methods: During the period from 2008-2016, we have managed 33 patients with pituitary adenomas with GK radiosurgery at our center. There were 22 male (66.6%) and 11 female (33.4%) patients and the mean age was 42.5 years. Nineteen patients (57.5%) presented with non-functioning and 14 (42.5%) with functional adenomas. Twenty-nine (88%) patients were managed with residual or recurrent tumors after surgery and 4 patients (12%) were managed as a primary mode. Gamma knife was accomplished with Leksell gamma unit 4-C at our center. Radiological, ophthalmological and hormonal responses were evaluated.

Results: The mean follow-up period was 39.8 months (range 3-87months). None of the patients experienced decline in visual acuity or field of vision related to the procedure. Tumor control was achieved in 91% patients. One patient had increase in size and underwent surgery. Hormonal evaluation showed no new endocrinopathy. Of the 8 patients with prolactinomas, 7 (77.7%) had normalization of hormone level; one patient had experienced increase in prolactin level. Five patients with acromegaly responded well. One patient with Cushing’s disease relapsed and underwent second session of gamma knife after 3 years. Of the one patient who presented with cranial nerve dysfunction, showed complete recovery within 12 months of therapy.

Conclusion: Gamma knife radiosurgery is an effective and safe mode of treatment for residual or recurrent pituitary adenomas. 

Introduction

The treatment planning for Gamma Knife (GK) stereotactic radiosurgery (SRS) that performs dose calculations based on tissue maximum ratio (TMR) algorithm has disadvantages in predicting dose in tissue heterogeneity. The latest version of the planning software is equipped with a convolution dose algorithm as an optional extra and the new algorithm is able to compensate for head inhomogeneity. However, the effect of this improved calculation method requires detailed validation in clinical cases. In this study, we compared absolute and relative dose distributions of treatment plans for acoustic schwannoma between TMR10  and the convolution calculation.

Methods

Twenty-nine clinically used plans created by TMR10 algorithm were recalculated convolution method. Differences between TMR and convolution were evaluated in terms

of absolute dose (beam-on time), dosimetric parameters including target coverage, selectivity, gradient index.

.

Results

The discrepancy in estimated absolute dose to the target ranged from 1 to 7 % between TMR and convolution. In addition, dosimetric parameters of the two methods achieved statistical significance. However, it was difficult to see the change of relative dose distribution by visual assessment  on a monitor.

Conclusions

Convolution algorithm are necessary to reduce the dosimetric uncertainty of each case in GK SRS.

Aim: The goal of our investigation was to evaluate efficacy and safety of SRS and HFSRT in patients suffering from growth hormone producing pituitary adenomas, treated with the use of linear accelerator at our institution between 2001 and 2011. Material and methods: A group of 96 patients treated with radiosurgery was reviewed to include 49 patients (26 women, 23 men) with acromegaly. The mean and the median age at the time of radiosurgery was 46 and 44 years, respectively (range 24 - 74). Almost all patients were treated surgically before SRS (94%). Anterior pituitary insufficiency was diagnosed in 17 cases. Diabetes insipidus occurred in 6 cases. The mean and the median total doses were 13.7 and 15 Gy, respectively (range 6 - 25Gy). The treatment was applied in 1 - 3 fractions. Serum IGF-1 and GH levels were obtained before radiotherapy and during the follow-up at 1 to 2-year intervals. Results: After the treatment, complete hormonal evaluation was performed in 30 (60%) cases. The median follow-up was 77 months (range: 9-166). Hormonal response was defined as a normalization (complete response) or more than 50% reduction (partial response) in serum IGF-1 levels. Normalization was noticed in 56% cases. However, progression after initial complete response was seen in 13% in the median time of 55 (24-120) months after the treatment. More than 50% reduction in IGF-1 levels was confirmed in another 17% patients. In 27% cases there was no hormonal response seen after irradiation and another course of radiotherapy was conducted in 7 cases. There were 8 cases of new hormonal deficits found in the current series. Radiological stabilization of the adenoma was seen in 32/34 patients (94%). In one case regression and in another one progression was diagnosed. Two-, five-, and ten-year local control was 96%.  There were no new visual deficits after SRS observed in our investigation. Conclusions: Stereotactic radiosurgery provides satisfactory radiological and hormonal control in patients suffering from acromegaly. Long term follow-up is essential in detecting new pituitary insufficiencies or disease progression.

Background:

The treatment of benign skull base tumors remains challenging. These tumors are often located in close relationship to critical structures and can be quite large when finally diagnosed. Thus a radical resection can be virtually impossible or would be associated with high morbidity. Therefore multimodal treatment concepts with partial tumor resection followed by radiosurgery are optimal. Adaptive Hybrid Surgery Analysis (AHSA) is a software developed by BrainLab that provides an automatic virtual real-time treatment plan during surgery for the feasibility of stereotactic radiosurgery.

Methods:

The AHSA software was applied in four patients who underwent partial resection of a benign skull base tumor. Tumor volumetry was obtained on preoperative and postoperative MR scans. Organs at risk were automatically delineated with the Atlas Mapping feature of the Elements software by BrainLab and adapted if needed.

Results:

Three patients underwent a suboccipital craniotomy for a partial resection of a tumor within the posterior fossa (vestibular schwannoma, ependymoma, petroclival meningioma) and in one patient a temporal approach for resection of a medial sphenoid wing meningioma was performed. The preoperative tumor volumes ranged between 8.52 and 25.2 cm³. The intraoperative volume of the residual tumor measured with the AHSA software was between 2.13-12.17 cm³ (25-47% of the preoperative tumor volume) differing between 8-20% from the real residual tumor volume measured on the postoperative MR scan. The preoperative estimated residual tumor volume differed only between 3-5% from the real residual volume measured on the postoperative MR. The intraoperative automatic AHSA plans of the remaining tumor volume suggested in all four patients that a safe radiosurgery would be feasible.

Conclusion:

The AHSA software is a smart tool that enables the neurosurgeon to preoperatively plan and estimate a multi-modal approach for complex skull base tumors and thus minimizing the morbidity risks for surgery as well as radiosurgery. Further the software provides a continuous intraoperative feedback of residual tumor volume and the resulting radiosurgery plan, which appraises if the remaining tumor could be feasibly and safely treated with radiosurgery. The accuracy of the intraoperative surface scanning of the residual tumor depends on the angle, location and depth of the remaining tumor together with the relation of the pointer and camera visibility respectively and therefore can differ quite a bit from the real residual tumor volume.

Objective. A comparison of dosimetric patametres of CyberKnife (CK) and Novalis plans for hypofractionated treatment of different intracranial pathologies.
Materials and Methods. 15 patients with single intracranial tumors (tumors have different histology, volume, and localization) previously treated on the CyberKnife were selected for comparison of dose distribution. Targets and critical structures were contoured by MultiPlan and transferred in iPlan with CT scan of the patient for recalculation. Mean doses in targets for each patient were equal in both planning systems. Homogeneity index (HI), coverage (CO), conformity indexes (CI_RTOG, nCI), gradient indexes (GI₅₀, GI₂₅), the dose in normal tissues (V₁₀, V₁₂) and critical structures (optic nerves, chiasm, brain stem) were obtained from DVH. To automate processing of DVH a software was developed in Borland C++ Builder 6.0.
Results. The most informative indicators for comparing plans were the GI50 gradient index, the conformity indexes CI_RTOG and nCI, as well as indicators of radiation doses on the brain V₁₀ and critical structures. While Novalis provided more homogeneous dose distributions within the target, CK irradiated less exposed volume in normal tissues. Doses in critical structures, located far from the target, are lower for Novalis. If the tumor is adjacent to critical organs, CK allows to reduce the dose in them more effectively, due to the high dose gradient outside the target. 

Background: The purpose of our report is to describe an innovative system used for mandibular immobilization during Gamma Knife surgery (GKS) procedures. It is based on an approach originally developed in Marseille in extracranial lesions, close to or involving the mandible, which may imply a certain degree of movement during the therapeutic image acquisitions and/or GKS treatment. Methods: The maxillofacial surgeon applied bone titanium self-tapping monocortical screws (4; 2 mm diameter, 10 mm length) between roots of the teeth in the fixed gingiva (upper and lower maxillae) the day before GKS (local anesthesia, 2 minutes time). Two rubber bands were sufficient for the desired tension required to undergo GKS. We further proceeded with application of the Leksell stereotactic G frame and carried out the usual GKS procedure. Results: The mean follow-up period was 2.3 years (range 0.6–3). Three patients have been treated with this approach: 2 cases with extracranial trigeminal schwannomas involving the mandibular branch, with decrease in tumor size on MR follow-up; 1 case with residual paracondylian mandibular arteriovenous malformation following partial embolization, completely obliterated at 7 months (digital subtraction angiography programmed 1 year after treatment). Conclusions: Jaw immobilization appears to be a quick, minimally invasive, safe and accurate adjunctive technique to enhance GKS targeting precision. Additionally, with the help of this procedure and/or the use of the new GKS Icon Model (Elekta Instruments AB, Sweden), we can expand the therapeutic radiosurgical management that can be offered to patients with lesions located at this level, by increasing targeting accuracy, which is particularly necessary within this location, due to possible jaw movement. 

Treatment of aggressive meningiomas may require a combination of surgery, radiotheraphy and radiosurgery to achieve long term local control (LC).  We report two patients with histologically defined malignant meningiomas infiltrating dural sinuses, which underwent several Gamma Knife treatment achieving long term radiosurgical LC. These patients after several years had systemic diffusion with thoracic and abdominal metastasis. Case 1: A frontal parasagittal meningioma was partially resected in a 48 yo woman (Simpson not 0 for sinus involving). The histological diagnosis was meningioma (G1 WHO). Two years later  she was reoperated for a removal of LR. The second histological diagnosis was  anaplastic meningioma (G3 WHO) with high cellularity,  frequent mitosis (16x10 HPF), high nuclear pleomorfism, necrosis areas and vimentin expression. During the following 9 years the patient underwent 5 GKRS procedures and a new surgery for LR (G3 WHO). In every occasion a dose of 15 Gy (50% isodose) was delivered to the growing tumor. At 62 yo she presented with 3 lesions al level of L5, of the right adrenal gland  and of right lung and underwent to a right lobectomy and mediastinum lymphadenectomy. The lung lesion was characterized  as an anaplastic meningioma metastasis (G3 WHO). No LR brain lesion at last follow-up, KPS 100. Case 2: In a  47 yo woman an anaplastic atypical meningioma (G2 WHO) involving the torcular was resected with Simpson not 0 for sinus involvement. 3 years later she had  SRT (32 Gy) and Tamoxifene for LR. During the following 5 years she undergo routinely brain MRI scans as follow up. During the following 5 years she underwent 5 GKRS for local recurrences  in contiguous fields (dose range 15-16 Gy, 50% isodose) important to note that the recurrence was not in field of previuos radiosurgical treatment. Octeotride was administered with no clear benefit.  At 61 yo multiple lesions in the lungs, liver and  kidney were found. A liver biopsy showed an  anaplastic meningioma (G3 WHO).  No LR brain lesions at last follow-up, KPS 100. Extracranial metastases of meningioma often reach the lungs (60%), abdomen and liver (34%). We present two long surival intracranial malignant meningiomas cases treated with multimodality treatment, also with several GKRS treatments, that gave metastatic sistemic diffusion. In our opinion the possibility to control the local recurrence of this small percentage of  malignant meningiomas allows the opportunity for these lesions to metastasize so we suggest radiological systemic control in those particular cases.

Background: The natural history of many small-to-moderate sized vestibular schwannomas is to remain stable over time, without any treatment. Numerous large studies have reported excellent control rates (>95%) with SRS, but have not accounted for whether or not the treated tumors were growing before SRS. The goal of this study was to determine the control rate of growing vestibular schwannomas to SRS.

Methods: This retrospective single-institution study included all sporadic vestibular schwannomas treated primarily with Gamma Knife radiosurgery between 2002 -2014. Patients with a history of NF2, prior surgical treatment, or less than 2 years of imaging follow-up were excluded. Clinical data including reason for treatment, dose prescription, and need for salvage treatment, were collected. Volumetric tumor analysis was performed on the initial, treatment planning, and latest follow-up MR T1 post contrast imaging, from which the pre-treatment and post-treatment percent volume change was calculated.  Tumors were then classified as  ‘growing’ or ‘stable’ based on the presence of a volume change exceeding published volumetric measurement errors (Snell 2006) based on tumor volume and MR slice thickness. Radiographic and clinical control rates were calculated for these two groups and compared using a Χ² test.

Results: 184 sporadic vestibular schwannomas were primarily treated with Gamma Knife SRS between 2002 and 2014. 97 cases were excluded due to insufficient follow-up information. The remaining 87 patients included 50 women, had a median age of 60.9 years, a median tumor volume of 0.78 cm², and a median follow-up time of 4.25 years.  Median dose prescription was 12 Gy to the 50% isodose line. Five patients required salvage treatment (4 surgical, 1 repeat Gamma Knife). The overall clinical (no salvage treatment required) and radiographic (no significant volume increase) control rates were 96% and 76%, respectively. Prior to SRS treatment, 34 patients had significant tumor growth (median increase 60% per year, median treatment volume 0.9cm³), 36 did not have significant growth (15% per year, 0.6 cm³), and 17 did not have available interval pre-treatment imaging. The clinical control rate was 97% for both growing and stable tumors (at median follow-up of 3.3 and 4.5 years, respectively). Radiographic control rates were 68% and 81%, respectively.  There was no significant difference between the control rates (clinical and radiographic) between growing and stable tumors.

Conclusion: In this small series, we did not see a significant difference in the clinical or radiographic control rates after SRS for growing versus stable vestibular schwannomas. 

Therapeutic options for benign intracranial tumors have expanded beyond surgery to include fractionated and single fraction radiation therapy. Stereotactic radiosurgery delivers a highly conformal dose to the intracranial target in a single or few fraction. Multicenter studies and single institutional analyses report high control rates and low toxicity after radiosurgery for intracranial neuromas and meningiomas, making it an appealing alternative to surgical intervention. Radiosurgery has been traditionally performed as Gamma Knife surgery but linac-based modalities have gained popularity and are now also widespread.

We present the results of linac-based stereotactic radiosurgery for 34 consecutive neuroma and meningioma patients after long follow up. All patients were given primary or adjunct radiosurgical treatment for a single intracranial tumor with a median volume of 5.95 cm³ and a maximum diameter of 35 mm. Prior intracranial radiotherapy was an exclusion criterion. Stereotactic radiosurgery was performed using the 6 MV beam of a non-dedicated  Elekta SL-18 linear accelerator converted for radiosurgery with the attachment of an isocentric subsystem (Phillips SRS200XK). Non-coplanar arc irradiation was delivered with circular collimators ranging in diameter from 10-30 mm. A stereotactic head-ring fixation was used. A treatment plan was achieved using 1-8 isocenters. Neuroma patients were treated with 11-12 Gy, while larger doses of 12-15.5 Gy were given to meningioma patients. Dose refers to the prescriptionisodose which covered 95-100% of the tumor. Combining a different number, span and weight of non-coplanar arcs, as well as weight and collimator size of each isocenter used, high conformality of the treatment dose to the borders of the tumor was established at the expense of planning time.

An extended follow-up is required after radiosurgery for benign intracranial tumors to assess not only tumor control but also the safety of the method. Our follow-up protocol consisted of imaging studies and clinical examination with cranial nerve function assessment at 6-months intervals for the first year and yearly thereafter. Clinical follow up was obtained from the patients or their referral doctors.

After a follow-up time ranging from 50 to 171 months (median 86), 56% of tumors decreased in size. Shrinkage of the tumor was observed in 58% and 61,5% of neuroma and meningioma patients, respectively. No patient developed new permanent facial or trigeminal neuropathy.

Overall, our study confirms the efficacy and safety of linac-radiosurgery after long follow up for  neuroma and meningioma patients.

Background:

 Radiosurgery is an increasingly preferred treatment option for localized prostate cancer, and stereotactic body radiation therapy (SBRT) a relatively established modality of therapeutic irradiation. The present study analyzes the toxicity and biochemical efficacy of SBRT in 6 consecutive prostate cancer patients treated with CyberKnife Robotic Radiosurgery System.

Methods:

Six patients were treated with SBRT at Cyber Knife SRS Department Jinnah Postgraduate Medical Center, Karachi, Pakistan. All patients included in this IRB-approved protocol-driven prospective study had biopsy proven prostate cancer. None of the patients developed low, intermediate or high risk post Cyber knife toxicities. The patients were treated with CyberKnife-SBRT (CK-SBRT), the prescription dose was 35 Gy in five fractions, corresponding to 92 Gy in 2-Gy fractions (α/β=1.5 Gy).

 Results:

The follow up PSA was repeated after every 3 months of radiosurgery. The mean PSA value for patients treated with Cyber Knife was 10.5ng/ml with a minimum value of 8ng/ml and maximum of 14 ng/ml. first follow up revealed a mean PSA value of 4.75 ng/ml with 6.2 ng/ml and 3.4 ng/ml maximum and minimum PSA values respectively. For next two follow up PSA evaluation the mean PSA values were, 3.31 ng/ml and 1.59 ng/ml respectively. None of the patients developed genitourinary and gastrointestinal toxicities.

Conclusions:

In this study CK-SBRT has provided promising outcomes in localized prostate cancer with good PSA response, minimal toxicity and patient inconvenience.

[Purpose]: To evaluate clinical outcomes of stereotactic body radiation therapy (SBRT) for lung oligometastases.

[Methods]: Between June 2009 and October 2016, 27 patients (16 males and 11 females) with 32 lung oligometastatic lesions were treated with SBRT at our institution. The median age was 73 (range, 48-85) years. The primary sites were the lungs (n = 17), colorectum (n = 3), uterus (n = 2), stomach (n = 2), liver (n = 1), thymus (n = 1), and buccal mucosa (n = 1). The histological types included adenocarcinoma (n = 16), squamous carcinoma (n = 8), large cell neuroendocrine carcinoma (n = 1), leiomyosarcoma (n = 1), and hepatocellular carcinoma (n = 1). The primary cancer sites were treated with surgery (n = 19), chemo-radiotherapy (n = 6) and SBRT (n = 2). The median interval between initial therapy and SBRT was 32 (range, 6–94) months. The median tumor diameter was 15 (range, 8–29) mm. The prescribed dose was defined at the isocenter. The total doses and fraction numbers were 48 Gy in 4 fractions for 27 lesions, 50 Gy in 5fractions for 5 lesions, respectively. Over all survival (OS), cause specific survival (CSS), local control rate (LCR), and relapse free survival (RFS) were calculated using Kaplan-Meier curves. Univariate and Multivariate analysis for RFS were performed using log-rank test and a Cox- proportional hazards model, respectively. Statistical significance was set at p<0.05.

[Results]: The median follow-up period was 26 months (range, 4 - 88 months), for all patients, and 37 months (range, 4 - 88 months) for survivors. During follow-up, 14 patients had recurrence (in-field failure, 2, primary site, 2, the other lung metastases, 6, and other distant metastases, 4). Of the 27 patients, 9 died of the primary disease, 3 died of other causes, and 15 survived. Of the 15 survivors, 8 are alive without recurrence and the other 7 are alive with cancer-bearing state. Of the 6 patients who presented with new lung metastases, 3 with oligometastasis underwent SBRT for the second time, of which one underwent SBRT for the third time. The 3-year OS, CSS, LCR, and RFS were 61.0%, 68.5%, 91.5%, and 37.4%, respectively. Univariate and multivariate analysis revealed that no factor significantly affected RFS. No adverse event of grade 3 or greater occurred.

 [Conclusion]: SBRT is an effective and safe treatment option for lung oligometastases.

Purpose: To improve the setup accuracy of thermoplastic mask, we developed a new monitoring method based on force sensing technology and evaluated its feasibility.

Methods: The thermoplastic mask setup monitoring system consists of a force sensing resistor sensor unit, a signal transport device, a control PC and an in-house software. The system is designed to monitor pressure variation between the mask and patient in real time. It also provides a warning to the user when there is a possibility of movement. A preliminary study was performed to evaluate the reliability of the sensor unit and developed monitoring system with a head phantom. Then, a simulation study with volunteers was conducted to evaluate the feasibility of the monitoring system. Note that the sensor unit can have multiple end-sensors and every end-sensor was confirmed to be within 2% reliability in pressure reading through a screening test.  

Results: To evaluate the reproducibility of the proposed monitoring system in practice, we simulated a mask setup with the head phantom. FRS sensors were attached on the face of the head phantom and pressure was monitored. For 3 repeated mask setups on the phantom, the variation of the pressure was less than 3% (only 1% larger than 2% potential uncertainty confirmed in the screening test). In the volunteer study, we intended to verify that the system could detect patient movements within the mask. Thus, volunteers were asked to turn their head or lift their chin. The system was able to detect movements effectively, confirming the clinical feasibility of the monitoring system developed.

Conclusion: Through the proposed setup monitoring method, it is possible to monitor patient motion inside a mask in real time, which has never been possible with most commonly used systems using non-radiographic technology such as infrared camera system and surface imaging system.

Purpose: Hard-plate based abdominal compression is known to be effective, but no explicit method exists to quantify abdominal compression force (ACF) and maintain the proper ACF through the whole procedure. In addition, even with compression, it is necessary to do 4D CT to manage residual motion but, 4D CT is often not possible due to reduced surrogating sensitivity. In this study, we developed and evaluated a system that both monitors ACF in real time and provides surrogating signal even under compression. The system can also provide visual-biofeedback.

Methods: The system developed consists of a compression plate, an ACF monitoring unit and a visual-biofeedback device. The ACF monitoring unit contains a thin air balloon in the size of compression plate and a gas pressure sensor. The unit is attached to the bottom of the plate thus, placed between the plate and the patient when compression is applied, and detects compression pressure. For reliability test, 3 volunteers were directed to take several different breathing patterns and the ACF variation was compared with the respiratory flow and external respiratory signal to assure that the system provides corresponding behavior. In addition, guiding waveform were generated based on free breathing, and then applied for evaluating the effectiveness of visual-biofeedback.

Results: We could monitor ACF variation in real time and confirmed that the data was correlated with both respiratory flow data and external respiratory signal. Even under abdominal compression, in addition, it was possible to make the subjects successfully follow the guide patterns using the visual biofeedback system.

Conclusion: The developed real time ACF monitoring system was found to be functional as intended and consistent. With the capability of both providing real time surrogating signal under compression and enabling visual-biofeedback, it is considered that the system would improve the quality of respiratory motion management in radiation therapy.

[Purpose/Objective]

Neck lymph node recurrences in irradiated field occur in 5 to 10%, which are important mode of failure for cancer patients. In spite of this, few studies have reported about salvage re-irradiation for lymph node recurrences. We retrospectively investigated the safety and efficacy of CyberKnife (CK) treatment for neck lymph node recurrences after radiotherapy.

[Material/methods]

Between 2008 and 2014, 16 patients with neck lymph node recurrences after radiotherapy were treated with CK at our institution. All of the recurrences were inoperable due to medical reasons. The treatment doses and fractions were decided depending on tumor- and surrounding critical structures’ volume. The dose to the 10 cm³ of skin and mucosa was limited to be < 14 Gy given as stereotactic radiosurgery (SRS) to avoid adverse effects. Local control (LC) and overall survival (OS) were calculated using SPSS software.

[Results]

The follow up period after CK ranged from 2 to 53 months (median, 11 months). The previous radiation dose ranged from 50 to 70 Gy (median, 65 Gy). Neck dissection had been performed for 6 patients (38%) as primary treatment. The target volume ranged 0.05 to 91 cm³ (median, 1.2 cm³). The prescribed isodoses ranged from 52 to 89% (median, 64%) for the target. The prescribed marginal dose was ranged 18 to 40 Gy in 1 to 8 fractions. Assuming SRS, the marginal dose ranged from 17 to 22 Gy (median, 20 Gy). SRS, the 3-fraction, 5-fraction, 6-fraction and 8-fraction radiotherapy was administered to 40, 5, 2, 1, and 1 lesions, respectively. The local recurrence was observed as 9 lesions in 6 patients. The 1-year LC rate was 81%. The 1-year OS was 84%. Fatal bleeding was observed in one patient who had huge (91 cm³) and widespread tumor invasion to carotid artery before CK. The other patients did not develope => Grade 3 toxicities. The 1-year LC with the target volume ≦ 1.3 cm³ was significantly higher than with the target volume >1.3 cm³ (p = 0.007). There was no significant difference in the 1-year LC due to the marginal dose for SRS (< 20 Gy versus ≧ 20 Gy).

[Conclusion]

CyberKnife for neck lymph node recurrences were performed safely and efficiently in most cases.  CK may be a candidate as a salvage treatment. We have to pay much attention to large and widely invaseve tumor before CK.

Purpose: Aim of the present study was to evaluate the efficacy and toxicity of Stereotactic body radiotherapy (SBRT) in patients with low or intermediate risk prostate cancer.

Materials and methods: Patients with the following characteristics were enrolled: initial PSA ≤ 20 ng/ml, Gleason Score ≤ 7, IPSS ≤ 7. The treatment schedule was 35 Gy in five fractions, delivered with Volumetric Modulated Arcs (VMAT) and flattening filter free beams (FFFs). Toxicity was recorded according to CTCAE criteria v4.0. Biochemical failure was calculated according to the Phoenix definition. The EPIC questionnaire was used to evaluate quality of life (QoL).

Results: Between March 2012 and April 2015, 100 patients were enrolled (58 low risk, 42 intermediate risk). The median age was 72 years (range 48 - 82). Acute toxicity was mild. Rectal grade 1 toxicity was found in 15 patients, whereas grade 2 toxicity in 6 patients. Thirty patients presented grade 1 urinary toxicity and 31 patients presented grade 2, mainly represented by urgency, dysuria and stranguria. Regarding late toxicity, grade 1 proctitis was recorded in 14  patients and grade 1 urinary in 44 patients; only 2 events of grade 2 urinary toxicity were observed. At a median follow-up of 33 months, three intermediate risk patients and one low risk patient experienced a biochemical failure. Health-related QoL revealed a slight worsening in all the domains during treatment, with a return to baseline 3 months after treatment.

Conclusions: Stereotactic body radiotherapy proved to be safe and effective in localized prostate cancer patients.

Objectives: Jugulotympanic parangangliomas can cause significant morbidity, despite the majority behaving in a non-malignant fashion. Radiotherapy is useful both in the salvage setting where surgery fails, and in the primary setting it provides a good alternative to surgery. We report on the outcomes of 27 cases of jugulotympanic paraganglioma treated with Linac-based stereotactic radiotherapy at a single institution.

Methods and materials: Between 1999-2015, a total of 27 patients were treated with Linac-based stereotactic radiotherapy. Small tumours were treated in a single fraction with a median dose of 16Gy prescribed to the 80% isodose. Intermediate sized tumours received 28Gy in 7 fractions (prescribed to the 80%) and large tumours or those in broad-contact with the brainstem received 50.4Gy in 28 fractions (prescribed to 90%).

Results: Median follow-up was 53 months (range 8-138). Eight out of 27 patients had surgery prior to radiotherapy and the median time to progression (after surgery) was 5 years.

Fifteen (56%) tumours were treated in a single fraction. Mean percentage volume coverage 98.8% and the mean conformity index was 1.55, defined as the ratio between the prescription isodose volume and the target volume.

Of the 27 patients, 26 (96%) achieved radiological control with 5 tumours reducing in size, 3 patients documented to have a temporary enlargement of the tumour before subsequently reducing and the remaining 18 cases did not change in size throughout the follow up period. Clinical control (defined as stability or improvement in symptoms) was achieved in 25 (93%) patients. The total number of cranial neuropathies present prior to radiotherapy was 58 and after radiotherapy this reduced to 49. There were 3 cases of new onset hearing impairment attributable to radiotherapy.  

Conclusion: Linac-based stereotactic radiotherapy represents a safe and effective treatment for Jugulotympanic paragangliomas, both as a primary and secondary treatment modality.

Purpose/Objectives: Locoregional recurrence is a major cause of death in patients with squamous cell carcinoma of the head and neck (HNSCC). At the moment, there are no clear recommendations and standards regarding the timing, total doses and dose tolerance of normal tissues to re-exposure. Based on limited studies on the re-irradiation with high total doses, we evaluated the tolerability of high-dose re-irradiation with SBRT.

Materials/Methods: 6 patients with histologically confirmed locoregional recurrence of HNSCC, received reirradiation in 2015-2016. Median time after primary radiotherapy course was 36 months (9-108). The treatment volumes and total doses were formed as follows: GTV (local recurrent tumor or nodal recurrence delineated on CT, MRI and ¹⁸F-FDG PET-CT) + CTV (0.3-0.5 cm, excluding unaffected bones and OARs) + PTV (0.2 cm or 1 mm near OARs) was treated to the total dose of 35-39 Gy (equivalent to 66-70 Gy of conventional fractionation, α/β=3). Single doses to these volumes were 7-13 Gy. Radiation treatment course consisted of 3-5 fractions. Varian Eclipse v.10 (VMAT) and Accuray MultiPlan 4.6 were used for treatment planning; patients were treated on Varian TrueBeam STx and Accuray CyberKnife 4G. According to the literature, in a year after primary irradiation almost complete recovery of normal tissue tolerances is observed. Tolerances of the eye, lens, optic nerves and chiasm, brain stem, spinal cord, parotid gland, blood vessels, intact mucosa of the mouth and pharynx were not exceeded. Patient positioning accuracy was controlled by kV-imaging and cone beam CT on linac and kV-imaging system on CyberKnife.

Results: All 6 patients received full course of radiation therapy. Radiation toxicity manifested with grade 2 oral and pharyngeal mucositis (depends on tumor localization) and grade 2 radiation epidermitis in 2-3 weeks after irradiation. After one month, almost complete relief of radiation mucositis and dermatitis was observed. In 5 patients partial response and stabilization of irradiated lesion were achieved according to MRI at one, three and six months after irradiation. Three patients performed ¹⁸F-FDG PET-CT with full metabolic response. At present time 4 patients are alive. One patient died in 4 months after RT due to comorbidities. Cause of death of another patient was fatal bleeding from irradiated tumor, located in hypopharynx, in 6 days after RT completion.

Conclusion: Using SBRT for curative reirradiation of recurrent HNSCC is available with maintaining satisfactory tolerability. Local control is quite promising, but accurate patient selection needed according to comorbidities and fatal complications risk.

From March 2013 to December 2016, 39 patients with lung tumors received SBRT at our institution. The aim of this study was to analyze clinical and dosimetric data, including pretreatment quality assurance (QA) measurements, for patients treated with volumetric modulated arc therapy technique (VMAT) and with a follow-up greater than 3 months.

Nineteen patients were evaluated (14 male, 5 female) with median age of 76 (59-83) and median performance status, according to Karnofski scale, of 80 (70-90). Eighteen were primary lung tumors (cT1a-cT2a, 3 histologically proven) and 1 was a metastasis. Five lesions were centrally located and 14 were peripherally located. The prescribed dose ranged from 40 to 60 Gy in 3-8 fractions. The BED₍₁₀₎ ranged from 80₍₁₀₎ to 112.5₍₁₀₎ Gy.

The target volumes (PTV) were delineated on the average CT series of a 4D-CT acquisition as 5mm isotropic expansion of the ITV. The median PTV volume was 17.9 cm³ (9.1-66.3 cm³). All plans were elaborated with TPS eclipse v.10.0 using 2-5 coplanar and non-coplanar arcs and delivered with a Varian Trilogy.

The median dose received by 99%, 95%, 50% and 2% of PTV was respectively 102.7%, 105.8%, 114.9% and 121.3% of the prescribed dose. Dose constraints for critical structures, according to the AAPM-TG101 report, were respected in all plans.

QA measurements were analyzed with EPIQA v2.2.3 comparing planar dose distributions at the isoplane calculated in water phantom and acquired with the EPID during the delivery of each arc. Dose distributions were locally compared in terms of gamma analysis with 3mm and 3% (Gamma_{3mm, 3%}) as distance-to-agreement and dose difference. The Gamma Index Passing Rate was calculated for all arcs (A-GIPR, n=58) and then averaged for all arcs in the 19 plans (P-GIPR). Average A-GIPR and P-GIPR were respectively 96.4% (89.2%-99.4%) and 96.4% (91.2%-98.6%). With more restrictive criteria, (Gamma_{2mm, 2%}, Gamma_{1mm, 1%}) the average P-GIPR decreased to 87.9% and 54.8% indicating the need of different criteria for acceptance.

With a median follow-up of 13.6 months (3.3-30.8), no patients experienced G3 lung toxicities (according to RTOG score). Two acute G2 and 4 late G2 lung toxicities were observed. Complete response or stable disease were obtained respectively in 10 and 3 patients. Three patients experienced locoregional outfield disease progression and 1 died. Three patients experienced distant disease progression and 2 died.

SBRT with VMAT technique offers a safe and effective approach for treatment of T1-2a lung tumors or small metastatic lesions. 

Objectives:  Stereotactic body radiotherapy (SBRT) for organ-confined prostate cancer has increased in acceptance over the past few years, as multiple studies continue to emerge demonstrating excellent biochemical control and low toxicity. We examined clinical outcomes of extreme SBRT using Cyberknife.

Methods and Materials:  From July 2007 to December 2016 a retrospective analysis was carried out on 218 consecutive patients with a median age of 74 years  (range  52 – 86) years, mean prostate volume of  75.6 cc (range 37.03 -163.16)  and  clinically localized prostate cancer.  Pre-treatment PSA ranged from 1.75 to 51.13 ng/ml (median 7.19 ng/ml). Cyberknife was used to deliver fiducials based image guided hypofractionated Stereotactic Radiotherapy. According to  D'amico risk classification the majority of patients 116 (53%)  were low risk , 58 pts (27%) were  intermediate risk and 44 pts (20%) were high risk. Among the entire study cohort 19 of 44 high risk patients received androgen deprivation therapy (ADT), ADT was not administered to any low – intermediate risk  patients.  A prescribed dose of 38 Gy in four fraction was delivered to the PTV, which was defined as the prostate (plus seminal vesicles in high risk patients). Real-time intrafractional motion tracking was used.

Results: Acute urinary symptoms (frequency, dysuria, urgency, hesitancy and nicturia) were common in 44 % of patients experiencing grade I-II RTOG acute urinary  toxicity. No patients experienced RTOG grade 3 acute urinary toxicity while 7 patients (3%) experienced RTOG grade 3 late urinary toxicity, in 3 of them as consequences of repeated urological instrumentation, including cistoscopy and urethral dilatation.  No RTOG grade 3 acute and late rectal toxicity was observed. The actuarial median follow up is 58 months (range 12 – 114 months). The five years actuarial psa relapse free survival rate according to D'Amico risk classification is 94.3% (CI: 92.5%-96.1%) with 99.1% for low risk, 94.6% for intermediate–risk  and 83.7.% for high risk respectively. Overall 11 (5%) of 218 patients failed biochemical failure, occurring in 1 low-risk patient, 4 intermediate- risk patients and 6 high-risk patients. 23 patients (10%) died during follow up for unrelated causes, only one patient (0,4%) died for prostate cancer (bone metastases).

Conclusions: Cyberknife SBRT produces excellent biochemical control rates with mild toxicity and minimal impact on quality of life. Median PSA levels compare favourably with other radiation modalities and strongly suggest durability of our results. 

Objectives:  Recently The International Society of Urological Pathology (ISUP) proposed a new "Grade Group" system to better reflect prostate cancer biological behavior.  In patients with clinically localized prostate cancer treated with Cyberknife Stereotactic Body Radiotherapy (SBRT) we compared the D'Amico Risk classification with the New " Grade Group" System in attempt to better stratify patients.

Methods and Materials:  From July 2007 through December 2016 a retrospective analysis was carried out on 218 consecutive patients with clinically localized prostate cancer.  Pre-treatment PSA ranged from 1.75 to 51.13 ng/ml (median 7.19 ng/ml). According to the D'amico risk classification 116 patients (53%) were low risk , 58 patients (27%) were  intermediate risk and 44 patients (20%) were high risk. Following the new grading System patients were classify: Group 1 (GS 3+3) 145 pts (67%) , Group 2 (GS 3+4) 29 pts 12%, Group 3 (GS 4+3) 17 pts (8%), Group 4 (GS 4+4) 19 pts (9%), Group 5 (GS 9-10)  8 pts (4%). Cyberknife was used to deliver fiducials based image guided Hypofractionated Stereotactic Radiotherapy, all patients were treated with 38 Gy in four fractions given daily.

Results: With a median follow up of 58 months (range 12 – 114 months) the five years actuarial psa relapse free survival rate according to D'Amico risk classification is 94.3% (CI: 92.5%-96.1%) with 99.1% for low risk, 94.6% for intermediate risk  and 83.7.% for high risk. Overall 11 (5%) of 218 patients failed biochemical, occurring in 1 low-risk patient, 4 intermediate risk patients and 6 high risk patients. 23 patients (10%) died during follow up for unrelated causes, only one patient (0,4%)for prostate cancer. Kaplan-Meyer five years actuarial PSA relapse free survival rate according to the new " Grade Group" System observed are: group 1: 98.6%, group 2: 89,1%, group 3: 84.7%, group 4: 84.8%, group 5: 87.5%.

Conclusions:  The comparison on our patients cohorts shows that a slightly higher number of patients (67% vs 53%) can be classified in a low risk group class with similar five years PSA relapse free survival, a restricted number (13% vs 20%) belongs to high risk class with no difference on PSA relapse free survival, while a significant difference on PSA relapse free survival was registered in intermediate group (85-89 % - vs 95%). The new " Grade Group" System classification differs significantly from the D'Amico risk classification and seems to better stratify patients in predicting prostate cancer recurrence.

Objectives: Clinical radiobiological data indicates that delivery of larger  daily doses of radiation may improve the therapeutic ratio for prostate cancer.  Treatment, however, often impact the quality of life due to side-effects and treatment-related toxicities.  We report prospectively collected toxicity data from a cohort  of localized prostate cancer patients treated with Cyberknife Stereotactic Body Radiation Therapy  (SBRT).

Methods and Materials:  Between July  2007 to December 2016 a retrospective analysis was carried out on 115 consecutive patients with a median age of 75 years (range  60 – 86), mean prostate volume of  67.7 cc (range 37.03 -163.16),   and  clinically localized prostate cancer treated with Cyberknife stereotactic radiosurgery. The majority of patients 58 (50%)  were low risk , 35 pts (30%) were  intermediate risk and 20 pts (20%) were  high risk  using the NCCN criteria . Pre-treatment PSAs ranged from 1.75 to 51 ng/ml (median 7.6 ng/ml).  42% of patients had moderate to severe lower urinary tract symptom prior to treatment (baseline AUA > 8). The course of radiotherapy consisted of 38 Gy over four fraction given daily to the GTV, which was defined as the prostate (plus seminal vesicles in high risk patients). Real-time intrafractional motion tracking was used. RTOG toxicity grades were assigned for genitourinary (GU) and gastrointestinal (GI)

Results: In total 13 patients died during follow up for unrelated causes. Data to assess GI and GU Toxicity were available for 102 patients with a median follow up of 80 months (range 61 -114). Acute urinary symptoms (frequency, dysuria, urgency, hesitancy and nicturia) were common with 54 % of patients experiencing grade I-II RTOG acute urinary  toxicity. No patients experienced RTOG grade 3 acute urinary toxicity, in 6 patients (5%) we recorded RTOG grade 3 late urinary toxicity, in two of them urethral dilatation was required  for bulbar urethral stricture.  In one patient a bladder fistula was observed. No RTOG grade 3 acute and late rectal toxicity was observed. The median time from CK radiotherapy completion to the occurrence of late grade 3 GU toxicity was 42 months (range 18-76)

Conclusions: Significant long term toxicities are minimal when Cyberknife stereotactic Hypofractionated radiotherapy is performed as monotherapy: this probably reflects the ability for current technology to minimize adverse effects of therapy. Continued accrual and follow up will be necessary to confirm long-term results.

Purpose:   Although surgical resection remains the standard of care for patients with early stage non-small cell lung cancer (NSCLC), Stereotactic body radiation therapy (SBRT) utilizing high doses per fraction is being increasingly employed in the setting of early stage non small cell  lung cancer (NSCLC). We report prospectively collected data from a cohort of early NSCLC patients treated with  Cyberknife (CK) Stereotactic Body Radiation Therapy (SBRT).

 Matherials and methods:  Between February 2005 to October 2013, a total of  90 consecutive patients with a median age of 76 years (range 42 - 90) and pathologically proven NSCLC (squamous cell carcinoma, adenocarcinoma, large-cell carcinoma, bronchoalveolar cell carcinoma, or NSCLC not otherwise specified) diagnosed as 41 stage IA and 56 stage IB were referred to our Radiotherapy Department for Cyberknife stereotactic radiotherapy treatment (SBRT).   In all patients selected tracking modalities depending on tumor size, location and extent of respiratory movement were performed. 27 pts with lower lobe lesions were treated with fiducial markers (1-3), in the others 70 fiducial less X sight option was used. The median tumor volume was 30 cc (range, 4-143). Depending on tumor size and location different curative dose regimens were used, but in all pts a high BED (> 100 Gy) was delivered to the tumor bed. The prescribed dose was  25-60 Gy in 1-5 fractions.  Median follow-up was 30 months (range, 6 - 102).

Results:  The Kaplan-Meier LC rate at 1,2,3 and 5 years was respectively 98%, 92%, 88,5% and 81%. Reported acute side effects have generally been mild including esophagitis, fatigue, chest wall tenderness and cough. For the evaluation of toxicity we used the RTOG/EORTC scale. We detected a late radiation-induced local fibrosis using CT imaging in most pts, in 3 we observed grade 3 pneumonitis, in 1 patient chest wall pain and rib fracture and in another one G3 radiation induced mielopathy at an interval of 30 months following CK/SRS treatment.

Conclusions: CK/SBRT for limited-stage primary lung cancer has been shown to be a promising treatment with high LC rates and very low toxicity. Prospective studies should be performed to determine whether this therapy may be an alternative option for surgery.

We investigated geometric and dosimetric accuracy of SRS treatments planned on a standard external beam treatment planning system (TPS, Eclipse, Varian Medical Systems) and delivered with an MLC-equipped accelerator (Edge, Varian Medical Systems).  Nine field sizes from 1.6x5 to 30x30 mm were planned for delivery to a plastic block phantom using 10 MV flattening filter free (FFF) rotating beams.  The fields were shaped by an MLC with 2.5 mm wide leaves.  For each field size, the dose distribution generated by a 360° gantry arc was computed and compared to film measurements.  The phantom had markings for precise positioning at kV isocenter.  It was split along the coronal plane for insertion of radiochromic film.  After exposure, films were scanned and evaluated using in-house software.  Averaged over all field sizes and 10 deliveries per field, planned peak doses agreed with measurements within 2.4±2.0%.   Planned and measured beam profiles along the x- and y- directions agreed better than 0.2 mm.  The centers of delivered doses missed target center on average by no more than 0.18 ± 0.07 mm, with maximum error 0.33 mm.  For multiple-arc SRS at various couch angles, the effect of table excursions was investigated by tracking the position of a pointer, originally positioned at CBCT isocenter, as the couch was rotated.  Deviations up to 0.5 mm from its setup position were noted.  These were entered into the TPS as isocenter shifts in a plan consisting of 9 arcs of 160° gantry sweeps delivered at 6 different couch angles with 2.4x5 mm fields.  The plan showed that couch excursions shifted the dose cloud by only 0.13 mm and had negligible effect on the shape of dose profiles.  For end-to-end tests, a cylindrical phantom was constructed having a 3 mm long, 3.15 mm diameter air-filled cavity for simulation of a small brain target.  Following CT simulation (head protocol, 1 mm slices) the 9-arc regimen was planned and delivered under CBCT guidance.  Based on film in the coronal and sagittal planes, the average targeting error was 0.33±0.06 mm with 0.41 mm maximum.  The measured dose agreed with the TPS within 1.6 ±2.0%.  Accelerator and imaging system proved stable and reliable throughout the 6 month time span of the tests.  In conclusion, even for the small field sizes used in SRS the Eclipse TPS and Edge accelerator with MLC can provide a high degree of dosimetric and spatial accuracy.

Objectives: Modern dose calculation algorithms (DCA), solving the linear Boltzmann equation (which is statistically solved with Monte Carlo (MC) methods), such as Acuros XB, are presently being introduced into routine practice. When using MC calculation model, the absorbed dose is calculated to the medium (Dmed) divided in voxels, instead to water (Dw) as with most previous generation algorithms. The purpose is to attract attention to the required dosimetric parameters, radiobiological criteria and dose tolerance values for target and organs at risks (OARs) to safely implement the Dmed mode into routine treatment planning. 

Methods: Ten lung cancer patient cases were used. The doses were calculated using Dw algorithms as reference, and recalculated with Dmed either using the same number of monitor units (MUs) obtained with Dw, or the same prescription dose (Dpr), keeping the same beam arrangements. The dosimetric parameters were compared. The 2D gamma (γ) was used to calculate γ-average, γ-passing rates (γrates), and γ-maps. The equivalent uniform dose (EUD) model was used for TCP and NTCP calculations. The bootstrap simulation method with no parametric tests was used to estimate the significant difference, with p < 0.05.

Results and discussion: To precisely calculate Dmed, taking into account the secondary electron transport in lungs, requires significantly more MUs, compared to Dw models, inducing an over irradiation of the patient. The dosimetric differences, sometimes exceeding 5%, between Dw and Dmed depend on the selected and final DCA. This dosimetric shift should be taken into consideration regarding Dpr, if the 2%/2 mm criteria, at least, would not be satisfied with γrates. However, caution should also be paid, when comparing dose distribution from a reference ‘true’ MC simulation with any DCA using γ-value and γrates due to the statistical noise of the dose distributions calculated by MC. The radiobiological analysis showed that EUD(Dmed) < EUD(Dw) for target, using same MUs, leads to an overestimated TCP with Dw; but comparable EUD are produced for OARs. When using the same Dpr, EUD(Dmed) > EUD(Dw) leading to more TCP/NTCP.

Conclusion: Although it would be justifiable to suggest an adjustment of Dpr, when changing DCA from Dw to Dmed, most departments would certainly prefer to keep on with the same Dpr, thus taking advantage of the small increase of TCP. Then a reasonable goal would be to increase the constraints for OARs to limit the NTCP shift.

Objectives: To compare Flattening Filter (FF) and Flattening Filter Free (FFF) photon beams SBRT treatments for lung cancer; and determine potential benefits and drawbacks of using unflattened beams.

Methods: Since July 2016, 11 patients were treated,by Elekta VERSA HD 6-10MV FFF photons beams: 10 peripheral and 3 central lesions (max diameters 1-3 cm). Treatment plans were developed both for FFF and FF by Monaco TPS with Montecarlo. The prescription dose at 95% of PTV (GTV+0,5cm all directions) was 26 Gy/1Fx (peripheral lesions) and 60Gy/8Fx (central lesions).

Results: Median follow-up was 3,5 months: no thoracic/mediastinal disorders; minimal pulmonary toxicity G1 at imaging; 8 no-captating PET and 5 TC lesions reduction at follow-up.

Isodoses/DVH for both techniques were compared: a lower Sigma index for FFF indicates an improved dose homogeneity at PTV; conformity index is comparable for both techniques; peripheral doses are smaller and at organs at risk mean doses are 2-4% lower in FFF modality, mostly using 10 MV energy.

FFF plans (higher dose rate 2200MU/min) displayed a higher MonitorUnit average number, but a reduction in treatment delivery/immobilization time, with clinical benefits in intrafraction organ motion management, patient comfort and dose delivery accuracy.

FFF reduces scatter, leakage, off‐axis variations in beam energy and photoneutron fluence per monitor unit produced by the high energies, in comparison to FF.

Conclusion: FFF photon beams modality seems currently advantageous and shows no drawbacks to deliver clinically acceptable SBRT/SBRS treatments for lung cancer.

By using FFF modality with small field sizes, high dose rate and high dose fraction, as well as in SBRT/SBRS, some recommendations would have to be considered.

Beam data acquisition, scanning techniques and beam modeling are vitally important. Because of the proximity to critical organs of PTV, the penumbra determination of small fields is clinically important. The dosimetry of FFF beams is made more complex by the P‐ion determination: corrections to beam percentage depth doses and profiles need.

The FFF X‐rays have a lower “effective energy” compared to FF X‐rays. Therefore, it is important to evaluate the skin (entrance) doses for internal organs. For VERSA HD, FFF skin doses are marginally higher than FF skin doses, but likely not clinically significant.

Moreover, several scientific articles address radiobiological concerns using FFF photons beams, regarding to high dose rate use and high dose per fraction delivery.

Scientific protocols and a longer follow-up are claimed to evaluate these topics and potential benefits on the treatments outcome.

Purpose: To report our clinical experience on brain and spine stereotactic radiosurgery (SRS) using InCise™2 Multileaf Collimator (MLC) on CyberKnife M6™ System.

Method: The InCise™2 MLC has been successfully implemented in routine clinical use in our institute for more than a year. To assess the benefits of using the MLC over cone based collimator, we performed plan comparison on 14 selected patients (7 brain cases and 7 spine cases) who were treated with MLC, and two potential MLC candidates (1 brain and 1 spine) who were ultimately treated with Iris. For these selected cases, the planning treatment volume (PTV) ranged from 20 cc to 258 cc (average 75 cc). Treatment doses ranged from 20 Gy to 40 Gy in 1 to 5 fractions. All the treatments were planned using the same constraints and objectives with both MLC and Iris collimator. Target coverage, critical structure sparing, prescription isodose line, conformity index (CI), monitor units (MU), treatment time, as well as the dose gradient (defined by the ratio of V50% to PTV) were compared between the two methods.

Results:

MLC plans were found to be clinically equivalent or better compared to Iris plans in the fourteen patients treated with MLC. MLC plans were prescribed to (79.9±2.8)% with (94.7±3.9)% target coverage, and CI of 1.28±0.1. The Iris plans were prescribed to (80.6±2.8)% with (95.7±2.9)% target coverage, and CI of 1.27±0.1. All of the 14 plans met constraints on the critical structure sparing. Less MUs and treatment time were required for all the MLC plans. The average reduction in MUs and treatment time were 36% (9% - 66%) and 35% (21%- 53%) respectively. All the MLC plans had slight smaller dose gradient compared to that of Iris plans (3.03 for MLC plans versus 3.46 for the Iris plans), which means better dose drop off around the target for MLC plans. For the two patients with superior Iris plans, their MLC plans were not able to meet the critical structure constraints with the same PTV coverage. Both patients were in the setting of repeat irradiation with PTVs in contact with critical structures and with stricter dose constraints.

Conclusion: The Incise™2 MLC is beneficial for treatment with larger PTVs (>20 cc). Compared to Iris collimator, MLC allows an average reduction of 35% in MUs and treatment time, with an improved dose gradient.

Purpose: Few definitive treatment options exist for elderly, medically inoperable patients diagnosed with early stage breast cancer. Historical data suggests very poor local control with hormone therapy alone. In this study, we examine the dosimetric feasibility of definitive robotic stereotactic body radiation therapy (SBRT) for the treatment of early stage breast cancer.

Methods and materials: Patients with biopsy-proven early stage breast cancer were identified who underwent pre-operative CT scans. Using these scans, a GTV was identified and contoured to include all visible tumor. A CTV was generated by applying a 5 mm expansion, appreciating anatomic boundaries of spread. The PTV was a 2 mm expansion from CTV for robotic SBRT plans, and standard RTOG expansions were utilized to generate comparison PTVs for 3D-CRT plans. The appropriate PTV was treated to a total dose of 50 Gy in five fractions utilizing both robotic SBRT and 3D-CRT for dosimetric comparison. Five-fraction dose volume criteria were specified as per NSABP B-39/RTOG 0413 and TG-101.

Results: The mean GTV and CTV volume were 4.92 cc and 16.84 cc, respectively. The mean PTV for SBRT and EBRT plans were 24.35 cc and 65.18 cc, respectively (p < 0.01). GTV and PTV coverage was excellent for robotic SBRT plans, with mean V50Gy values of 99.93% and 98.41% respectively. Although GTV coverage was equivalent in 3D-CRT plans, PTV coverage was notably worse (V50Gy = 91.73%, p = 0.02). The mean ipsilateral breast V40Gy was significantly lower in SBRT plans than 3D-CRT plans (5.34% vs. 13.14%, p < 0.01). The mean maximum dose to the chest wall was 44.08 Gy in SBRT plans, compared to 50.17 Gy in 3D-CRT plans (p = 0.02). The mean maximum dose to skin was 36.39 Gy in SBRT plans, compared to 41.66 Gy in 3D-CRT plans (p = 0.05). No significant differences were noted in ipsilateral lung, contralateral lung, heart, or thyroid dosimetry. Cardiac sparing with robotic SBRT was excellent, with a mean heart V2Gy for patients with left sided tumors of 5.12 cc.

Conclusions: Based on this preliminary data, definitive robotic SBRT appears to be a dosimetrically feasible treatment for early stage breast cancer. Significant normal tissue sparing, even with the high doses necessary for ablation, seem to be achievable. Prospective clinical trials will be crucial to explore the safety and efficacy of definitive SBRT in this patient population.

OBJECTIVE:  Role of CyberKnife Radiosurgery in pituitary mass and improvement in Growth hormone levels.

STUDY DESIGN: A prospective study.

PLACE AND DURATION OF STUDY: This study was conducted in the Department of Radiology and Cyber Knife Stereotactic Radiosurgery, JPMC Karachi from January 2013 to August 2015.

METHODOLOGY: A prospective study was conducted during the year of January 2013 to August 2015 in the department of CyberKnife, JPMC, Karachi in patients with pituitary adenoma diagnosed through MRI having clinical symptoms of amenorrhea, visual disturbances and headaches. Tumor size and Growth hormone levels were assessed before going through surgery and after Cyberknife treatment, reduction in tumor. A total of 50 patients with pituitary adenoma underwent Cyberknife radiosurgery procedure. The patient group comprised of 50 diagnosed cases of pituitary adenoma of men (56%) and women (44%). Included patients were either those with surgically inaccessible or post-surgical residual lesions. The median tumor volume was 10.14cm³ with a mean volume of 24.70 cm³. Maxim prescription dose was 25 Gy with prescription isodose line ranging from 60 to 84. The mean and median follow-up periods were 3 months and 6 months respectively.

RESULTS: Out of the 50 patients, 21(42%) patients with pituitary adenoma showed Reduction in size of tumor and marked decrease in growth hormone levels 38(76%) after Cyberknife robotic radiosurgery. 29(58%) patients showed stable size radiologically with reduced hormone level but 12(24%) patients showed same hormone level on follow ups. A pre and post cyber knife Radiosurgery visual acuity test was performed for each individual. No visual complication was observed on post cyber knife Radiosurgery.

CONCLUSION:-. Cyberknife is a cutting edge technology in the field of radiation oncology. After cyberknife Radiosurgery, the normal functioning of pituitary gland was found less affected with maximum dose delivery to the tumor. Our results proved cyber knife Radiosurgery as the safest treatment for those patients with post operative residual pituitary adenomas and those with inoperable tumors without any post radiosurgical complication.

Introduction.

To our knowledge this is the first report of functional radiosurgery with a fully automatized Rotating Gamma Ray Unit (InfiniÒ) by Masep (Shenzhen, China)in the Americas. One of the singularity of this machine is that uses  30 cobalt sources that rotate around the isocenter at one revolution per minute, the treatment requires an invasive frame in a similar fashion that its better known predecessor, the GammaknifeÒ.

Material and methods.

From April 2014 to October 2016, 37 patients of functional disorders. We used the Infini, gamma ray with the treatment planning system SuperPlanÒ, using image data sets acquired by Magnetic Resonance Imaging on T1 or T2 sequences as needed. We report four thalamotomies, three for tremor and one for intractable pain, two hipofisectomies for pain in terminal cancer patients, two bilateral capsulotomies for psychological disorders, one case of epilepsy and 28 cases for trigeminal neuralgia. In general single 4mm isocenters were chosen and on occasion tandem 4mm. Doses ranged from 78 Gy to 96 Gy in trigeminal neuralgia and 124Gy to 150gy to the 100% isodose line for the rest of the disorders.

Results.

Using the Dorsal Root Entry Zone (DREZ) as target for the first half of the trigeminal patients we achieved an 86% of excellent to good results according to Barrow Scale. In the cases of pain in terminal cancer patients the pain relief varied from 80 to 90% in both cases and results were seen according to Visual Analog Scale starting at 48 to 72 hours post treatment. In tremor from diverse pathologies the Fahn-Tolosa-Marin-tremor rating scale improved in 60 and 70%, one is pending as well as the medial thalamotomy for pain as treatments are recent. Finally, in one case of Obsesive Cumpulsive Disorder an 80% improvement is noticed two years after treatment and on and aggressive and autism patient that is the case remaining, minimal to no improvement has been noticed one year out after treatment.

Conclusions.

We report our initial experience in functional radiosurgery using InfiniÒ and the clinical results are in accordance of what has been published with similar technology.

We for see a wider application of radiosurgical treatments in countries like ours were other technologies used in surgery are economically less feasible.  

Objective: To evaluate the outcomes of high-dose gamma knife radiosurgery (GKRS) for the treatment of medically refractory trigeminal neuralgia (TN), and to clarify the effectiveness of early GKRS in the treatment of medically refractory TN.

 Material and Methods: This retrospective review covers 108 consecutive patients presenting with medically refractory, idiopathic TN between 2006 and 2014. The median follow-up period was 43 months. All patients underwent GKRS treatment targeting the root entry zone with median maximum dose of 90 Gy and isodose line of 20%. Telephone questionnaires were used for the collection of pre-operative and post-operative data. Outcomes pertaining to pain and facial numbness were scored using the Barrow Neurological Institute (BNI) pain and facial numbness scales, respectively. BNI pain scores of I to IIIb were defined as pain relief.

 Results: Following a median latency period of four weeks, we observed complete or adequate pain relief (BNI scores of I to IIIB) in 86 of the 108 patients (90%). Twenty-two patients (26%) experienced pain recurrence at a median period of 17 months. Fifty-nine patients (55%) reported new onset facial numbness. The treatment failure rates were highest among patients with a longer history of pain. Compared to patients with pain history of ≤5 years, those with pain history of >5 years experienced longer latency prior to pain relief (p=0.027, Kaplan-Meier). We did not observe a significant difference in the pain relief maintenance rate between patients with pain histories of ≤5 and >5 years (p=0.776). Univariate and multivariate analyses demonstrated that pain history of ≤5 years was a significant predictor of pain relief (p=0.049 and p=0.045, respectively).

 Conclusions: GKRS achieves a high rate of pain relief among patients with medically refractory, idiopathic TN. Pain history of ≤5 years is a reliable predictor of pain relief and appears to be associated with shorter latency to pain relief following GKRS. This means that early GKRS treatment for patients with medically refractory, idiopathic TN is recommended.

Background: Trigeminal neuralgia (TN) affects 7% of patients with multiple sclerosis (MS). In such patients, TN is difficult to manage either pharmacologically and surgically. Radiosurgical rhizotomy is an effective treatment option. The non-isocentric geometry of radiation beams of CyberKnife introduces new concepts in the treatment of TN. Its efficacy for MS-related TN has not yet been demonstrated.

Methods: Twenty-seven patients with refractory TN and MS were treated. A non-isocentric beams distribution was chosen; the maximal target dose was 72.5 Gy with a 4-6 mm of the trigeminal nerve included in the 80% isodose line (58 Gy). The maximal dose to the brainstem was

Results: Median follow-up was 37 (18-72) months. Significant pain relief was achieved in 23/27 (85%) patients within 45 days. Prescription isodose line (80%) accounting for a dose of 58 Gy incorporated an average of 4.85 mm (4-6 mm)  of the nerve and mean nerve volume of 26.4 mm3 (range 20-38 mm3). No  complication was observed. The rate of pain control decreased progressively after the first year and only 44% of patient retained pain control 4 years later.

Conclusions: Frameless radiosurgery provides a fully non-invasive technique for retroGasserian rhizotomy. Pain relief was satisfactory and, with our dose/volume constraints, no sensory complications were recorded. Nonetheless, long-term pain control was possible in less than half of the patients. This is a limitation that Cyberknife radiosurgery shares with other techniques in MS patients.

Introduction: Glossopharyngeal neuralgia (GPN) is a rare and disabling condition. In the manner of trigeminal neuralgia, Gamma Knife radiosurgery (GKRS) is more and more described as a therapeutic option. We propose to assess long-term safety and efficacy of GKRS in this indication.

Methods: From 2007 to 2015, 9 patients (4 males, 5 females) underwent a total of 10 GKRS procedures. All the patients presented GPN that were refractory to all medical treatment with a long-lasting pain story. One patient had previously undergone a surgical micro-vascular decompression. 5 patients presented a MRI-identified neurovascular conflict. Concerning GKRS procedure, the IXth nerve was localized on MRI and CT under stereotactic conditions. The target was located at the glossopharyngeal meatus of the jugular foramen. The dose administered to the nerve was 80 Gy in 3 patients and then 90 Gy for the others. Patients were followed at 6 and 12 months of the procedure and then repeatedly.

Results: 8 patients presented an improvement within a median onset of 7 weeks after GKRS (range 2-12 months). At the first follow-up, 6 patients were pain-free (Class 1-3) including 4 patients medication-free (Class 1). 2 patients were partially improved (Class 4) and one remained unchanged. The mean follow-up time was 46 months (range 10-90 months). At the last follow-up 6 patients remained pain-free (Class 1-3) including 4 patients medication-free (Class 1). No side effect was observed.

Conclusion: Because of its safety and efficacy, GKRS appears as a useful tool to treat essential glossopharyngeal neuralgia, including as a first-line treatment. 

Background / Purpose

The Parkinson`s Desease is the second most common neurodegenerative desease. Most frequent and essential clinical symptom is tremor. For treatment of this symptom alternative method to pharmacotherapy and radiofrequency ablation is radiosurgery.

The aim of this pilot study is to show the method and results of the stereotactic radiosurgery planning using CyberKnife System in patients with Parkinson’s Disease.

Materials & Methods

Heads of 9 patients had immobilised using thermoplastic masks. Precise symmetrical positioning of the head with the saggital axis and the orbital line perpendicular to the table was a key point of the treatment planning. Target and organs at risk was defined using MR (T1-weighted with contrast medium and fractional anisothropy) and CT (with and without contrast medium) fusion. Complex irradiated area included thalamic nucleus (Ventralis oralis Anterior and Ventralis oralis Posterior) on the opposite side to the limb suffering the tremor. The direct location of the target area was determined based  the guidelines defined in the stereotactic atlas (usually 12 millimeters to the side of the midcommisural point).

Radiosurgery treatment planning was performed using MultiPlan CyberKnife System. To collect base set of the images for dose calculation, CT without contrast medium was used. Treatment plans were performed according to approved protocol of the clinical trial.

Results

Three of 9 patients received total dose of 70 Gy as a maximal dose in the center part of the target. Total dose graduały increased by 5 Gy for the following next two groups, each of three patients. Therefore, total doses of 70 Gy, 75 Gy and 80 Gy were delivered.

Maximum dose to brainstem was 7,8 Gy ± 0,6 Gy, optic chiasm 1,9 Gy ± 0,8 Gy, lenses 0,2 Gy ± 0,1 Gy, optic nerves 1,1 Gy ± 0,9 Gy and internal capsule 22,2 Gy ± 7,9 Gy.

The 50% isodose had less than 8mm in the horizontal plane. Irradiation takes 47 ± 5 minutes and the number of radiation beams was 127 ± 11.

Conslusions

At the first glance, there is no reasonable arguments to ignore Stereotactic Radiosurgery using CyberKnife System as a treatment alternative to traditional treatment modalities of the Parkinson’s Disease.

Objetive:

Trigeminal neuralgia (TN) is usually idiopathic. However, there are rare cases (aprox 10%), in which TN is due to intracranial lesions. The most frequent are cavernous sinus meningiomas and trigeminal schwannomas. The purpose of the study was to assess the role of Gamma Knife surgery (GKS) in secondary TN caused byspace-occupying lesions.

Methods:

From July 2010 till November 2016, 20 patients had GKS for secondary TN. Eleven cases (55%) were meningiomas, 6 (30%) trigeminal schwannomas, 2 (10%) brain metastases and 1 (5%) arterio-venous malformation (AVM). The primary outcome was tumor control. The secondary outcomes were the alleviation of pain and the eventual secondary effects. Covariates were the age, duration of symptoms, duration till alleviation etc. The tumor was the primary target in all cases, with the exception of one, in which both the tumor and the nerve were targeted, with a free pain interval between treatments. The former was excluded from the present analysis

Results:

The mean age in this series was 61.9 years (range 39-79). The mean follow-up period was 2.6 years (range 1.3-5). The mean duration of TN was 11.8 months (range 0.5-160). Follow-up was available for 19 patients (95%). Pain alleviation appeared after a mean time of 4.2 months (1-11) in 17 patients (89.5%). Six (31.6%) patients completely stopped medication in a mean time of 7.2 months (range 1-12) and three (15.8%) decreased it at half of the initial doses. No patient developed new hypoesthesia or other cranial nerve complication. One patient presented at 6 months with a worsening of the symptomatology; after a corticotherapy, which lasted for one week, the patient regained the previous state. The marginal doses for meningiomas and trigeminal schwannomas were 12 Gy (12-14), for metastasis 20 (20-20) and for AVM 24 Gy. The mean maximal diameter was 20.17 (range 9-36). The mean target volume was 1.94 cc (range 0.12-8.10). The mean prescription isodose volume was 2.77 cc (0.19-11.90). The mean number of isocenters was 14.7 (4-27). At last follow-up, tumor decreased in 13 (65%) patients, was stable in 6 (30%) and increased in one (5%), the latest at 6 months.

Conclusions:

Gamma Knife surgery is safe and effective in treating intracranial lesions presenting with secondary TN secondary to intracranial lesions. The initial pain freedom response (BNI I-IIIA) was close to 90%, while having no secondary effect. Pain alleviation is achieved even in the absence of a volume variation of the lesions. 

 Introduction

For the treatment of essential trigeminal neuralgia(TN), Gamma Knife radiosurgery(GKRS) is widely adopted now. But there are few reports about GKRS for symptomatic TN caused by a tumor, arteriovenous malformation, etc. Generally, pain of symptomatic TN is controlled by GKRS of targeting the lesion earlier than that of essential TN. We also have experienced some cases of symptomatic TN. Here, we summarize the results of our cases and reviewed the strategy of GKRS for symptomatic TN.

Methods

From January 2006 to December 2015, 24 patients suffering from symptomatic TN were treated by GKRS at our institution. Of them, 20 patients that could be followed up for a minimum of 6 months were retrospectively examined.

Results

In 15 of the 20 cases, pain could be controlled by GKRS targeting only the lesion. But in some cases, pain could not be controlled in spite of the lesion control by GKRS. In these cases, the pain could be controlled by an additional GKRS directly targeting the fifth cranial nerve, using for treating essential TN. And, in cases of emergency or in cases where GKRS was not suitable for treatment of the lesion, from the first, adding or using the same GKRS for treating essential TN was effective.

Conclusions

We reviewed the strategy of performing GKRS for symptomatic TN from our experienced cases. In some symptomatic TN cases, pain could not be controlled only through lesion control by GKRS. Although GKRS for symptomatic TN should be directed to the lesion rather than the trigeminal root originally, in such cases, it was indicated that adding or using the same GKRS for treating essential TN was effective and safe for the treatment of symptomatic TN. Thus, we recommend that GKRS should be directed to the trigeminal root in the following cases: 1. when GKRS targeted at the lesion has not been effective. 2. if there is an emergency. 3. if the lesion is deemed not amenable to GKRS.

Functional disorders of the brain, such as movement disorders, dystonia and neuropathic pain, may respond poorly to medical therapy. A noninvasive alternative to deep brain stimulation (DBS) is radiosurgical ablation with high radiation doses. Although considerable positive experience has been recently reported, nowadays functional radiosurgery suffers from some limitations: MRI anatomical targeting is difficult and debatable and one therefore must rely mainly on stereotactic coordinates. Furthermore, lesion sizes may vary, and shielding adjacent radiosensitive neural structures may be difficult. The target volumes are delineated by a skilled multidisciplinary team of expert radiotherapists and neurosurgeons, hopefully with the support of  semi-automatic atlas-based methods that limits the identification errors. Despite the usefulness to assess the axonal organization of the white matter tracts for the saving of the main neural pathways, the MR diffusion tensor imaging (DTI) in functional radiosurgery has not been proven and its use has not been reported yet.

In  this methodological and feasibility study we used the envelope of DTI tractography fiber bundles in the treatment planning of essential tremor with Cyberknife in four patients. For this purpose, after a standard DTI preprocessing we reconstructed  the corticospinal tracts (CST) using a two regions of interests (ROI) approach, based on anatomic knowledge about fiber bundles with one ROI in the anterior mid pons and  the other in the white matter close to the sensorimotor cortex. Then the CST, coregistered and fused on a 3D T1 MRI sequence,  where exported in DICOM format and loaded into the treatment planning system (TPS), contoured and considered as organs at risk.

The dose distributions obtained with and without considering the CST as organ at risk during the inverse planning optimization procedure showed strong differences in the dose to the white matter tracts (up to 100%), while target coverage, average dose to the brain, and treatment delivery time remained unchanged. Such dose differences depend on the subjects anatomical variability and on the relative position between target and CST.

In conclusion we implemented a method that go beyond the visual inspection of fibers before planning, integrating them into a Cyberknife environment. This may be useful for the evaluation and control of radiation dose to the white matter tracts, particularly for the functional radiosurgery where high doses are involved (around 100 Gy) and to improve the knowledge about the pyramidal tract dose tolerance.

Objectives: Psychiatric surgery evolved from a highly controversial field to a sophisticated approach to treat severe, disabling and otherwise intractable mental disorders. In this context, Gamma ventral capsulotomy (GVC), is a promising treatment option for refractory obsessive-compulsive disorder (OCD). Most studies suggest that double shot GVC is efficacious, despite a few complications (eg., brain cysts). Recently, single shot GVC has been proposed as a less deleterious technique, showing good efficacy and favorable adverse events profile. The aim of this study was to replicate these findings on a cohort of five refractory OCD patients.

Methods: Five patients with intractable OCD were selected by OCD-experienced psychiatrists and received single shot GVC. Radiosurgery was performed using Gamma Knife Perfexion from 2014 to 2016. Targets were located at the most ventral borders of the anterior limb of the internal capsule, using a maximum dose of 150 Gy and 4 mm collimators. Patients were assessed for symptom severity (Yale-Brown Obsessive-Compulsive Scale – Y-BOCS), global functioning (Clinical Global Impressions Improvement scale- CGI), anxiety (Beck Anxiety Inventory and Hamilton Anxiety Rating Scale), depressive symptoms (Beck Depression Inventory and Hamilton Depression Rating Scale), and neuropsychological and personality changes, at baseline. Follow-up evaluations were done by psychiatrists periodically, checking for adverse events and treatment response during the follow-up. Response criteria was defined as a minimum of 35% reduction in Y-BOCS scores and an “improved” or “much improved” score on the CGI.

Results: No severe adverse events were reported after the procedure. Four patients (A, B, C, D) have already completed 12 months of follow up and one patient (E) has been followed up for 3 months. Among the five patients that have been submitted to the single shot GVC, none fulfilled our response criteria: the median decrease in Y-BOCS score was 4,48% for the first 4 patients with >12 months of follow-up. No patients presented “improved” or “much improved” scores on the CGI after the surgery. There were no significant changes on depression or anxiety symptoms.

Conclusions: Although a previous study suggested positive findings with single shot GVC, our results do not indicate so far a good efficacy profile of such small volume lesions. Future studies should address the relationships between target location, white matter fiber tracking along the internal capsule and clinical response. The effects of an additional dorsal shot lesion should also be investigated.

Objective: To evaluate the safety and efficacy of unilateral Gamma Knife (GK) thalamotomy for treatment of severe tremor with a retrospective assessment in our Institute.

Methods: From October 2014 till September 2016, 24 patients (mean age, 78 years; 13 men; mean duration of symptoms, 25.8 years) with severe drug-resistant tremor (21 essential, 1 dystonian, 1 post-thalamic hemorrhage, 1 parkinsonian) were treated with unilateral GK (all on left side) targeting the ventral intermediate nucleus (VIM) with single shot through a 4-mm collimator. Beam blocking technique was used in 4 patients. Neurosurgeon initially selected the target in the same way as with usual thermo-coagulation, then discussed the modification to the final target with GK surgeon. The maximum dose was 130 Gy. Neurologic and neuropsychological assessments were performed by an independent neurosurgeon before treatment and at 12 months thereafter. MRI follow-up was achieved at 1, 3, 6, and 12 months. Upper limb severity of the tremor in 14 patients with less than 12 month follow-up was statistically analyzed by the modified TETRAS (the essential tremor rating assessment scale).

Results: The upper limb every tremor sub-score, writing sub-score and circle drawing sub-score (0-4) was significantly improved (p=0.001, 0.001 and 0005 respectively according to Wilcoxon test). Tremor score was improved by 1 to 12 months (mean period, 5.5 months) after GK, while 2 patients had no effect after GK. The only side effect was a hemiparesis associated with excessive edema around the thalamotomy in one patient.

Conclusion: The well-discussed target decision with the strategy for GKT is a safe and efficient treatment for severe medically refractory tremor. 

Several studies have reported the efficacy and safety of hypofractionated stereotactic radiosurgery (hSRS) in the treatment of benign perioptic tumors. This study went further and evaluated the feasibility of hSRS in the treatment of those causing compressive cranial neuropathies (CCNs) among perioptic tumors with special consideration of functional improvement. Twenty-six patients with CCNs (CN II=19; CN III/IV/VI=9; CN V=3) caused by perioptic tumors underwent hSRS between 2011 and 2015. hSRS was delivered in 5 fractions with a median marginal dose of 27.8 Gy (≈14 Gy in a single fraction, assuming an α/β of 3) to a tumor volume of 8.2±8.3 cm3. All tumors except 1 shrank after treatment, with a mean volume decrease of 35% (range, 4%-84%) during the mean follow-up period of 20 months. In 19 patients (38 eyes) with compressive optic neuropathy, vision improved in 55.3% of eyes (n = 21), was unchanged in 36.8% (n = 14), and worsened in 7.9% (n = 3) (2.6% after excluding 2 eyes deteriorated due to transient tumor swelling). A higher conformity index (p = 0.034) and volume of the optic apparatus receiving > 23.0 Gy (p =0 .019) were associated with greater tumor shrinkage. A greater decrease in tumor volume (p = 0.035) was associated with a better improvement in vision. Ophthalmoplegia and facial hypesthesia improved in 6 of 9 (66.7%) and 3 of 3 (100%) patients, respectively. There was no newly developed neurological deficit. Decompressive SRS for benign perioptic tumors causing CCN is feasible using hypofractionation, representing a useful alternative to microsurgical resection.

Purpose: We investigated the dosimetric differences between robotic-arm based Cyberknife (CK) and frame based Gamma Knife (GK) in treatment planning of hypofractionated stereotactic radiosurgery (hSRS) for perioptic tumors.

Methods: We selected 18 perioptic tumors previously treated using CK between 2011 and 2015. The CK plan was generated using 4 or 5 dose-limiting auto-shells, one at prescription dose (PD) level, another at intermediate dose level (50% of PD) for steeper dose fall-off, and the others at low-dose levels, with an optimized shell-dilation size based on our experience. The GK plan was also produced using the original contour set. Thus, both data set of dosimetric parameters was generated and compared. hSRS was delivered in five fractions with a median marginal dose of 27.8 Gy (≈14 Gy in a single fraction, assuming an α/β of 3) to a median tumor volume of 3.1 cm³. All treatment planning goals was maximizing the minimum dose to perioptic tumor, while satisfying the dose-volume constraints criteria (D_max < 25 Gy and V_20Gy < 0.2 cc) of the optic apparatus as closely as possible.

Results: There were no differences between CK and GK plans in the D99.9% of tumor (mean 24.05 and 23.44, respectively; P=0.054) and conformity index (mean 1.24 and 1.27, respectively; P=0.825). However, CK plans produced significantly higher values of tumor coverage than those produced by GK plans (P<0.001). The OA doses were well within the tolerated dose-volume limits in all plans.

Conclusions: These results indicate that CK and GK have provided similar dosimetric quality, except for tumor coverage, maintained dose-volume limit of the OA.

Purpose: To analyze the outcomes of patients treated with the LINAC frameless stereotactic system for the resection cavity of brain metastasis (BM). Methods and materials: We retrospectively reviewed patients with no prior whole brain radiotherapy (WBRT) who underwent hypofractionated stereotactic radiotherapy (HFSRT) to the surgical bed of brain metastases at a dose of 23.1Gy (3 x 7.7Gy) prescribed to the 70% isodose line. The planning target volume was the resection cavity with the addition of a 2-mm margin. Results: Between February 2012 and February 2015, 86 patients with 92 cavities were treated with post-operative HFSRT. The lung had the most primary tumors (51.2%), and 58 patients (63%) had a single metastasis. The median follow-up time was 12.9 months. The actuarial local control rates at the resection cavity were 94% at 6 months, 90% at 1 year and 88% at 2 years. On multivariate analysis, a higher PTV was predictive of local failure (LF) (HR=1.04, p=0.031), and patients with PTV>22mL had 3.94 times higher risk of developing local recurrence than patients with PTV≤22 mL (HR=3.94, p=0.026). Distant brain failure (DBF) occurred in 57% of cases, and leptomeningeal disease (LMD) developed in 13% of cases. Subtotal resection of BM was significantly predictive of developing LMD (HR=7.4, p=0.0006). The median overall survival (OS) was 14.5 months. On multivariate analysis for OS, KPS>70 % (HR=0.53, p=0.044), a radiological suspicious of radionecrosis (HR=0.088, p<0.001), the time interval between the primary tumor diagnosis and manifestation of BM (HR=0.99, p=0.04), the presence of a single metastasis at treatment (HR=0.057, p=0.008) and a decreasing time between planning MRI and HFSRT (HR=0.92, p=0.0002) were favorable prognostic factors for OS. Salvage therapies included WBRT (28%), stereotactic radiotherapy for LF (n=4) or DBF (n=27), and surgery (n=5 for LF; n=2 for DBF). Fifteen patients (16%) developed brain radionecrosis at a median time of 14.4 months. Conclusion: HFSRT to the tumor bed without WBRT is an effective treatment for resected metastases. A PTV>22 mL was associated with a higher risk of local failure and subtotal resection of BM with an increasing risk of developing LMD. Although most patients develop distant intracranial failure, the stereotactic radiotherapy approach allowed the avoidance or delay of WBRT in most cases.

Purpose: To assess the risks of local and distant failure and overall survival time in patients treated with hypofractionated stereotactic radiotherapy (HFSRT) to the postoperative cavity for brain metastases compared with patients treated with adjuvant whole-brain radiation therapy (WBRT).

Methods and Materials: Between July 2005 and February 2015, 196 non-randomized patients with 202 resected brain metastases were treated with post-operative WBRT or HFSRT at a single institution. The propensity score was included as a covariate to compare the interval to local failure, distant brain failure and overall survival time. The matching covariates consisted of the age, Karnofsky performance status, primary disease, number of brain metastases, extracranial disease status and presence or absence of extracranial metastases.

Results: In total, 110 patients (54.5 %) received adjuvant WBRT, and 92 HFSRT procedures (45.5 %) were delivered after surgery. A Cox model adjusted on the propensity score showed that the brain distant failure was significantly associated with treatment modality. Compared with WBRT, the patients who received HFSRT had a significantly greater risk of developing distant brain recurrence (HR = 3.37 [2.13-5.33], p < 0.001). No difference was observed in local failure (HR = 1.16, p = 0.77) between the groups. In the propensity-matched cohort, the effect of treatment on survival was not significant (p = 0.14), but it depended on the time. Within the first 20 months, the patients treated with WBRT had a 2 times higher risk of death than did patients treated with HFSRT (HR = 2.17 [1.42-3.32], p = 0.0003).

Conclusion: Compared with the standard WBRT after the resection of brain metastases, stereotactic radiotherapy to the surgical bed produced a comparable local control rate to that of WBRT with a survival advantage in the first 20 months.

BACKGROUND

Optic nerve sheath meningioma（ONSM）is a rare benign tumor. Despite of the histologically benign tumor, it can cause visual disturbance and lead to blindness. The management of ONSM remains controversial especially in patients with bilateral lesion. We report a case of 27 years old woman with bilateral ONSM who was underwent fractionated stereotactic radiotherapy.

CASE PRESENTATION

A 27-year-old woman felt blurred vision of right eye. Her symptom gradually deteriorated and then she was referred to neurosurgical department. She underwent   MRI and it demonstrated abnormal mass involving bilateral optic nerves. The lesion showed iso intensity on T2 weighted image and homogeneously enhanced with gadolinium. The tumor progressed and her visual function worsened in a couple of months. She had an operation and bilateral unroofing of optic canal and biopsy of the tumor were performed. Despite of the operation, visual acuity did not improve and gradually deteriorated in both side instead. So she was introduced to our hospital for radiation therapy. When she was referred to our hospital, her visual acuity was deteriorated to hand motion on the right side and 24/40 vision on the left side. She underwent fractionated stereotactic radiotherapy of 50.4Gy in 28 fractions. After radiation treatment, ophthalmologic examination showed improvement in a visual acuity of the left eye and MRI demonstrated slight reduction of tumor volume.

DISCUSSION

Although ONSM is benign tumor, this slow growing tumor can lead to visual loss. Bilateral ONSM can cause total blindness. The treatment for patients with ONSM remains controversial because of the variability of the natural course of this disease. A small fraction of patients might have a stable course and/or present a slight improvement of the symptom during a follow-up period.

Historically, surgery was recommended to confirm the diagnosis and to decompress the optic nerve. However, surgery may deteriorate visual acuity instead. Recently, it has been reported that radiation therapy can control the growth of this tumor and maintain or improve visual function. The optimal timing of radiotherapy is uncertain. Some group reported that a longer time from the onset to radiotherapy might relate to poor outcome in visual acuity after treatment and so it is important to start radiotherapy immediately after appropriate diagnosis in order to avoid visual loss especially in patients with bilateral ONSM.

CONCLUSION　

We experienced a case of bilateral ONSM.

Fractionated stereotactic radiotherapy immediately after diagnosis could control the growth of tumor and improve visual function.

Subependymal giant cell astrocytomas (SEGA) are benign gliomas that occur in patients with the tuberous sclerosis complex (TSC). Surgical removal is the preferred treatment and expert opinion is strongly against the use of radiotherapy, insomuch that it is not even mentioned as a treatment option in reviews or guidelines. Recently, success has been reported with the mTor inhibitor everolimus in reducing tumor volume, but regrowth has been observed after cessation of the drug.  We report about a case with inoperable bilateral SEGA that was successfully treated with fractionated stereotactic radiotherapy (30 X 2Gy) with a follow-up of over 7 years before being treated with everolimus for renal involvement of TSC and currently a follow-up of 10 y. Successive volumetries on MR demonstrated a, 80% volume reduction after radiotherapy only, further enhanced by everolimus until only minimal residue is visible. Review of the literature shows there is no evidence for contra-indicating fractionated stereotactic radiotherapy in SEGA and that real-life surgery is still associated with recurrence, mobidity and mortality. We believe that stereotactic radiotherapy alone or sequentially combined with everolimus deserves to be studied as an alternative to removal in difficult surgical cases that do not require immediate volume reduction.

Purpose: Stereotactic radiosurgery (SRS) for skull base perioptic tumors is challenging with concerns about the risk of radiation-induced optic neuropathy. For these cases, hypofractionation of SRS with an optimal dosimetric planning in terms of dose-volume constraints to the optic apparatus (OA) as well as higher dose to tumor may potentially reduce the risk of radiation toxicity with acceptable tumor control. Here we investigated dosimetric outcomes of the Cyberknife (CK) and the Gamma Knife (GK) for perioptic tumors.

Methods: We selected 18 perioptic tumors previously treated using CK between 2011 and 2015. All tumors harbored no margin from OA with 10 of them compressing or 8 abutting onto it on MRI. Tumor type included 11 meningiomas, 4 hemangiomas, 2 pituitary adenomas, and 1 craniopharyngioma. For dosimetric planning, thin-sliced CT and MR images were obtained and fused for delineation and contouring of both tumor and OA. Identical contour set was used to produce both CK and GK plans for each case. For 5 fraction treatment, the goal of dosimetric planning was to limit dose to OA with dose-volume constraints of D_max < 25 Gy and V_20Gy < 0.2 cc, while maintaining tumor coverage as high as possible with median prescription dose of 27.8 Gy. A paired data set of dosimetric parameters based on each modality was obtained and analyzed.

Results: Although all plans of both CK and GK complied with the dose-volume constraints to OA, tumor coverage (mean 94.4% in CK vs 88.5% in GK; P<0.001) and minimum dose to tumor (mean 22.9 Gy vs 19.1 Gy; P<0.001) were significantly higher in CK than in GK. There were no differences in the conformity index (mean 1.24 vs 1.27; P=0.825). The gradient index (mean 3.39 vs 2.89; P=0.004) was significantly lower in GK than in CK.

Conclusions: These results suggest that CK inverse planning system appears to be more suitable for an optimal dosimetric planning for perioptic tumors than GK planning system.

OBJECTIVE. Stereotactic radiosurgery (SRS) in single session (SSR) has proved its efficacy in treating anterior cranial base lesions inferior to 15-20 mL.  However SSR seems potentially hazardous  in cases of anterior optic pathways (AOP) close-fitting or larger (>20 mL) lesions, especially for risks of radiotoxicity.  Following this, many Centers have adopted hypo-fractionation protocols. Objective of this study is to compare SSR versus Hypofracionated Stereotactic Radiosurgery (HSR) performed with Gamma Knife (GK) or Cyber Knife (CK).

Method.  HSR group consists of 100 patients (40 GK; 60 CK – minimum follow-up 24 months). Twenty-six males; 74 females. Mean age 56.05 yrs (range 35-85). Mean follow-up was 52.3 months (range 25.3-71.34). Mean planning target volume (PTV) was 5.69 mL (range 0.71 -19.17 mL).

GK-HSR was performed in 3 fractions (Mean PD/fx. 6.4 Gy; mean PI 50.5%) and CK-HSR respectively in 2 (26%), 3(60%), 5(10%) and 4 (4%) fractions (21 pts treated with PD of 18 Gy; 14 with a PD

RESULTS. HSR group: Seventy-four (74%) patients were stable (if no deficit pre-GK reported) or improved; 10 (10%) were unchanged while 16 (16%) worsened. In 48/100 we radiologically observed lesional a shrinkage; 38/100 were unchanged, 11/100 showed a “delayed control” (initial pseudo-progression than stable) and 3/100 increased. SSR group: 48 (42.4%) patients were stable or improved; 45 (39.8%) were unchanged, 20 (17.8%) worsened (7/20 for tumor progression, 13/20 for ARE). Seventy-eight cases (69.1%) showed a volume shrinkage at neuroradiological follow-up, 23 (20.3%) unchanged and 10/113 (8.8%) progressed.  In 2 (1.8%) cases, tumor was no longer recognizable at MRI.

Conclusion. Hypofractionated stereotactic radiosurgery appears to be a safe and effective even in patients with AOP close-fitting meningiomas. Pre-operative visual conservation seems to be associated with a more favourable clinical course.

Objective  : To analyze role of Sterotactic Radiosurgery for Brainstem Gloimas

Method  :Sixty nine patients with Glial tumors were treated consecutively between 2008 and 2016 with Leksell Gamma knife Model B and Perfexion. Out of these 16 tumors were involving brainstems. They were selected using clinical criteria without histopathology, with MR Spectroscopy. Functional imaging FDG PET / Radionucleide Spectroscopy was used for diagnosis and planning where ever possible. Dose prescription was limited by extent involvement of brainstem and volume. There were 3 females and only one was operated before treatment. Age of patient was from 8 to 67 years [mean 39.9yrs]. Five patients had disabling neurological defects and were bed ridden at time of treatment. Two patients were treated twice. Tumor volume ranged from 1.2 to 22.3cc [Mean 11.8cc]. Prescription dose was 8-25Gy [mean 13.9 Gy]. Patients were followed with contrast MRI scan every 3 months

Results: Patients survived from 1-13 years [mean 4.5 years] after Radiosurgery. Five patients developed side effect with significant neurological deficit with in 6 months. Longest surviving patient treated second time after 10 years died of coronary disease.  Eight patient showed initial improvement in their clinical status as compared to pre SRS with in 6 months. Various Clinical details and comparison with various published series will be compared.

Conclusion: Gamma Knife Radiosurgery has role in controlling inoperable brainstem glial lesions.

Introduction: Following gamma knife (GK) therapy for intracranial AVM’s, obliteration of the nidus occurs over several years. During this period there is a risk of edema and hemorrhage. We hypothesized that decreasing radiation dose to the draining vein(s) may prevent early draining vein obliteration leading to a decrease in edema and hemorrhage rates in the post GK period.

Materials and methods: This retro-prospective study over 5 years (Jan 2009 to Feb 2014) included patients with intracranial AVM who underwent gamma knife therapy (Leksell Perfexion^®, Elekta, Stockholm) at our center. Approval from the institute’s ethics committee was taken. All patients who underwent draining vein shielding (DVS) by senior author (DA) were included in the test group and patients who did not undergo DVA were put in the control group.  Patients with less than 6 months of follow up were excluded. All patients were followed up 6 monthly clinically as well as radiologically with CT head/MRI brain to see for edema. DSA was done at 2 years for all patients and repeated at 5 years for those with incomplete obliteration of nidus.

Results: 185 patients were included in this study of which 96 were in the control group and 89 in the test group. The mean age, sex distribution, co-morbidities and adjuvant treatment were comparable in both groups. The lobar distribution of the AVM, angio-architecture and radiation dose were comparable between the two groups. Due to shielding, the test group patients received significantly less radiation to the draining vein (0.0001). On follow up, significantly less number of patients in the test group had new neurological deficits (p = 0.001).

Importantly, significantly more number of patients in the control group had post-radiosurgery intracranial hemorrhage (p = 0.03) and brain edema (p = 0.002). Both the group had comparable AVM obliteration rates following radiosurgery.

Conclusion: Shielding of draining vein is a potent new strategy in minimizing edema and hemorrhage as well as clinical deterioration following gamma knife therapy for intracranial AVM’s.

Background:  A recent prospective observational study revealed that the survival pattern of SRS treated patients with 2-4 BMs are comparable to those with 5-10 BMs. The survival pattern of patients with >10 BM remain poorly characterized.

Methods:  A retrospective analysis of data from 5976 patients with brain metastasis treated with SRS were analyzed. Multivariate Cox proportional hazard modeling was performed, and overall survival was estimated by the Kaplan-Meier method.

Results:  The primary cancer types of the study cohort were: breast (n=763), gastrointestional (n=699), lung (n=3840), melanoma (n=341), and renal cell carcinoma (n=333). We recapitulated the findings of the previous prospective observational study. Median overall survival for patients with solitary BM (7.1 months) was superior to those with 2-4 BMs (6.4 mo) and 5-10 BMs (6.3 mo). The median survival of patients with 2-4 BMs did not significantly differ from those with 5-10 BMs (6.3 mo). The median survival of patients with >10 BMs (5.5 months), however, were lower than those with either 2-4 or 5-10 BMs (p<.001). The difference in survival between these two patient populations remained significant in a multi-variate model that accounted for age, Karnofsky Performance Score, systemic disease status, tumor histology, and cumulative intracranial tumor volume. Cox-proportional hazard analysis revealed that the hazard of death after SRS was increased by 15.8% in patients with >10 BMs relative to those with 2-10 BMs.

Conclusions:  The survival profile of patients with >10 BMs were less favorable relative to those with < 10 BMs.

Objectives: To compare the plan quality for brain metastatic lesions treated on Gamma Knife Perfexion (GK) with Volumetric Modulated Arc Therapy (VMAT) planned on a non-radiosurgery  linear accelerator.

Methods: 5 patients with a single and multiple brain metastatic lesions planned and treated on GK, these cases were re-planned on Eclipse with a single isocenter, 5 to 6 non-coplaner partial arcs using VMAT technique on 2100CD Varian linear accelerator which equipped with Millennium Multi Leaves Collimator 120 (MLC), and maximum doserate of 600MU/min. The MLC is 5 mm width over the 20 cm central region and 10 mm width over the two 10 cm peripheral regions. The size of the target ranges from 0.1 cc to 5.9 cc, and the prescribed dose ranges from 18 to 22Gy according to the target volume.  Plans were normalized so that 99% - 100% of the target volume receive 100% of the prescribed dose.

Dosimetric parameters used to analyze the plans were: conformity index (CI) (100% isodose volume/Target volume), gradient index (GI) (50% isodose volume /100% isodose volume),12, 9 and 3 Gy isodose volume (V12, V9 and V3), dose to the nearby organs at risk and the beam on time.

Results: The VMAT plans showed a lower CI than the GK (1.28 ± 0.16 vs 1.43 ± 0.25), but it has a higher GI (4.94 ± 1.8 vs 2.75 ±0.25). The V12 and V9 were comparable between the two modalities, for the low dose region, GK showed a smaller volume (V3=105 cc vs 179 cc), it has been noticed that  the low dose region depends on the position and numbers of the treated lesions. For the nearby organs at risk (brain stem and optic nerve) the results were comparable. The mean beam on time for VMAT was 14 min while for GK at the time of  treatment was 115 min.

Conclusion: For the range of target volumes used in this comparison, VMAT technique for multiple brain metastasis planned on a non-radiosurgery  linear accelerator showed an equivalent plan quality to GK, with better CI, shorter beam on time and comparable V12 and V9 and nearby organ at risk, so it could be an optional treatment modality, bearing in mind the quality of the linear accelerator in terms of accuracy and precision that has to meet the recommended tolerances for radiosurgery treatment delivery.

Disclosure: No significant relationship

In this study, we evaluated the dosimetric plan quality of single isocenter VMAT (Volumetric Modulated Arc Therapy) versus multi-isocenter DCAT (Dynamic Conformal Arc Therapy) for treatment of multiple brain metastases (≥4). Twenty patients treated with DCAT technique were replanned with VMAT technique. Plan quality has been evaluated by target conformity, quality of coverage, healthy brain tissue dose, dose gradient index, dose to organs at risk and delivery treatment time. The results showed that no significant difference was found for the conformity index (CI) and quality of coverage index. However, the healthy brain volume exposed to 12 Gy and 21 Gy was respectively 64% and 36% lower with DCAT and the dose gradient R_12Gy-21Gy went from 1.74 cm with DCAT to 2.19 cm with VMAT. Moreover, organs at risk near PTVs (< 5mm) are better spared by VMAT than DCAT. Finally, single isocenter VMAT required a 25 minutes treatment time regardless the number of metastases against 25 minutes for each lesion with DCAT technique. DCAT and VMAT are both efficient techniques to treat multiple cranial metastases and the choice of treatment technique should be done with respect to target coverage and OARs (Organs At Risk) protection as well as patient age and health condition.

Purpose

The previous treatment of brain metastasis from small cell lung cancer(SCLC) was Whole Brain Radiotherapy(WBRT). WBRT could be performed only once, so this method could not be used twice when metastatic brain tumors recurred after WBRT. Gamma Knife Radiosurgery(GKRS) is a new treatment method in recurred metastasis after WBRT. The effect of GKRS after WBRT for brain metastasis from SCLC was analyzed retrospectively.

Material and method

Thirty onepatients of brain metastasis from SCLC were analyzed in our study. When brain metastasis recurred after WBRT, GKRS was performed in all cases. The survival rate of GKRS plus WBRT was analyzed.

The tumor control rates, local recurrence rates and distant new recurrence rates were analyzed after GKRS with WBRT.

We also analyzed the cause of death - brain lesion or systemic disease.

Results

Thirty one patients could be follow-up except one patient. Median survival rate was 26(6-65) months. All patients were dead during follow-up periods. Kaplan Meier’s survival curve reveals that 1 year survival rate was 93.3%, 2 year was 66.7%, 3 year was 6.7%. Radiologic follow-up was possible in 23 patients. Mean follow-up period was 27(12-63)months. Only one patient had local recurrence after GKRS, whereas 5 patients had new metastasis. Both local recurrence and new metastasis had occurred in 1 patient. When local or new metastasis occurred, reGKRS could be performed. The cause of death from brain lesion was in 7 patients, from systemic disease was in 18 patients and both reasons were in 5 patients.

Conclusions

Patients who underwent GKRS after WBRT had significantly improved survival rates compared to the patients who had WBRT alone. The cause of death after WBRT plus GKRS was mainly due to the systemic disease.

Purpose: We investigated the impact of optimization in dose-limiting auto-shell function on the dosimetric quality of Cyberknife (CK) plans in treating brain metastases (BMs).

Methods: We selected 19 BMs previously treated using CK between 2014 and 2015. The original CK plans (CK_original) had been produced using one to 3 dose-limiting auto-shells, one at the prescription dose (PD) level for dose conformity and others at low-dose levels (10–30% of PD) for dose spillage. In each case, a modified CK plan (CK_modified) was generated using 5 dose-limiting auto-shells, one at PD level, another at intermediate dose level (50% of PD) for steeper dose fall-off, and the others at low-dose levels, with an optimized shell-dilation size based on our experience. A Gamma Knife (GK) plan was also produced using the original contour set. Thus, a triplet data set of dosimetric parameters was generated and compared.

Results: There were no differences among CK_original, CK_modified, and GK plans in the conformity index (mean 1.22, 1.18, and 1.24, respectively; P=0.079) and tumor coverage (mean 99.5%, 99.5%, and 99.4%, respectively; P=0.177), whereas CK_modified plans produced significantly smaller normal tissue volumes receiving 50% of PD than those produced by CK_original plans (P<0.001), with no statistical differences in those volumes compared with GK plans (P=0.345)

Conclusions: These results indicate that significantly steeper dose fall-off can be further achieved in the CK system by optimizing the auto-shell function, while maintaining high conformity of dose to tumor.

Purpose : The purpose of this study was to evaluate control rate and reducing volume of gamma knife surgery of single or multiple metastases in each pathology of lung cancer.

Material & Methods: Between November 2012 and July 2016, 144 patients (681 lesions) with lung cancer metastases were treated with gamma knife surgery. Among them, We studied 93 patients　(431 lesions) who were able to follow up MRI for more than 3 months.

Results: The male to female ratio in the SCLC and NSCLC was 8 to 1, 2 to 1 and the mean age of the patients was 64.11, 63.24 years. The average follow-up time after gamma knife surgery was 8.40 months (range 3-42 months). The mean volume reduction rate of tumor in SCLC and NSCLC was 76% and 48.5%. The tumor control rate in SCLC (89 tumors) & NSCLC (342 tumors) was 95.5% and 95.6%. 26 patients had single metastasis. The rest of the patients had multiple brain metastases : every 38 patient had 2~5 lesions (110 tumors) and 29 patients had more than 5 lesions (295 tumors).

Conclusion: Gamma knife surgery was effective for tumor control and reducing volume of metastases in SCLC and NSCLC.

OBJECTIVE

Breast cancer is well known as one of the common primary tumors of brain metastasis. Although the incidence of brain metastases in breast cancer patients has increased, the management of brain metastasis is still a challenge in clinical practice.  Breast cancer is classified by presence of some hormone and gene receptors (e.g., estrogen, progesterone, HER-2). Subtype classification is useful for prognosis prediction and treatment planning. We investigated the prognosis of the patients with in breast cancer brain metastasis managed by stereotactic radiosurgery (SRS) according to breast cancer subtypes.

PATENTS and METHODS

From 2004 to 2015, total 42 breast cancer brain metastasis patients were managed by gamma knife radiosurgery (GKRS) in our hospital. The median age of the patients was 51.02 (range, 25-83) at the time of initial diagnosis of brain metastasis. The prognosis was analyzed according to their breast cancer subtypes; Luminal A, Luminal B, HER2 and triple negative (TN) groups.

The total number of metastatic lesions was 136, and the indications for GKRS were the number of metastasis below 10 and maximal diameter 3.0 cm. According to breast cancer subtypes, the number of Luminal-A type was 17, and Luminal-B 53, HER-2 44, and TN 22.

RESULT

The tumor control was achieved in 87 out of 136 tumors, and the control rate was 64.0 % with the median progression free survival (PFS) of 370 days (range, 53-1901 days). Luminal-A type as breast cancer subtype group’s median PFS was 595 days (range, 171-648). Luminal-B, HER-2 and TN subtype groups’ PFS were 682 (range, 53-800), 915 (range, 67-1901) and 189 days (range, 73-520), respectively. Statistically, the tumor control of TN group was relatively lower than those of other three types (p=0.001). Median overall survival (OS) after GKRS was 529 days (range, 65-2302). Among those groups, Luminal-A type as breast cancer subtype group’s median OS was 295 days (range, 151-807). Luminal-B, HER-2 and TN subtype groups’ median OS were 320 (range, 65-1104), 573 (range, 161-2302) and 338 days (range, 91-990), respectively.

CONCLUSION

Breast cancer TN subtype group showed most poor prognosis in overall survival, and their brain metastases showed shortest PFS after SRS compared with other subtype groups. Therefore, the patient confirmed pathologically TN subtype breast cancer should be followed up more closely after SRS.

Gastric cancer brain metastasis is rare and it is known to be relatively radio-resistant. We report a gastric cancer patient with brain metastases who underwent stereotactic radiosurgery (SRS) and systemic pembrolizumab (Keytruda®) treatment, and showed dramatic tumor control.

A 39-year-old man underwent 2nd SRS for a locally recurred brain metastasis and newly developed several metastases. One year ago, he had undergone SRS for a single metastsasis in another gamma knife center, and the target tumor volume was 5.4cc. At the 2nd SRS, the locally recurred tumor volume was 14.6 cc and a margin dose of 14 Gy was prescribed to the 50% isodose line at the tumor border. The tumor volumes of the five new metastases were 0.3, 1.4, 1.5, 10.8, and 4.0 cc, and the prescribed marginal dose was 22, 22, 20, 16, and 20 Gy, respectively. After that, for the locally recurred advanced gastric cancer, the patient underwent intravenous pembrolizumab (Keytruda®) treatment at the department of oncology. The dose was 100mg, and 5 times injections were performed with two-week interval.          

Three months after the 2nd SRS, those new metastases were controlled dramaticallyn the follow-up MRI. Three small tumors disappeared, and the other two larger tumors were regressed dramatically, and their volume were 1.1 and 0.2 cc, respectively. Although the volume of repeated treated tumor was similar, the methionine positron emission tomography (PET) showed only small hot spots.

Considering both local progression of the previously treated with only SRS and dramatic control of the tumors managed by both SRS and consecutive systemic pembrolizumab (Keytruda®) treatment, SRS and systemic pembrolizumab (Keytruda®) might have a synergic effect in the treatment for metastatic brain tumors from gastric cancer.

Purpose

The project objective was to safely introduce frameless fractionated stereotactic radiotherapy (SRT) into our clinic using existing equipment.

Background

We have traditionally treated brain metastasis with whole brain radiotherapy (WBRT) in our clinic. Recently published American Society for Radiation Oncology evidence-based guidelines on the management of newly diagnosed brain metastases state that patients presenting with multiple brain metastases (all less than 3–4 cm) have various treatment options, including stereotactic radiosurgery (SRS) alone, WBRT with SRS boost, or WBRT alone, with no mention of SRT(1). However, the use of SRT has spread considerably in the past 10 years. SRT has been shown to be more effective than WBRT in the management of low volume brain metastases specifically in patients with tumours arising from the gastro-intestinal tract (2).

SRS and SRT treatments have largely occurred in departments with dedicated SRS systems. New Zealand has only one centre with dedicated SRS equipment. This often requires patients to travel long distances for treatment.   

Our department has decided to take advantage of the advances in imaging technology (on board imaging), patient support assembly systems (tighter couch top rotation, 6 Degrees of Freedom couch (6DoF)) and enhanced beam delivery systems (Flattening Filter Free (FFF)) to introduce SRT. The introduction of SRT offers more options to our patients and potentially more convenience as they do not have to travel long distances for treatment.  

Methods

Firstly, we validated our existing beam model for small field dosimetry. We evaluated our immobilisation system to determine margins. Planning technique studies were retrospectively done to decide the best technique between RapidArc, Conformal arc and static fields (for both co-planar and non-coplanar configurations). Winston Lutz tests were done to confirm coincidence of imaging and radiation isocentres. End to end testing was done with a Rando head phantom. Extensive testing was done using pin point chamber and GaF chromic film measurements. Pre-treatment QA procedures were established using a CIRS head phantom.

Results

The beam model was accurate down to a 1x1cm². The margins were determined to be 2mm, with a margin review task for instances where there was rotation exceeding 2% in patient set up.

Conclusion

Our experience has shown it is possible to implement an SRT brain technique with existing resources without huge capital expenditure in dedicated SRT/SRS systems. The next focus will be to explore doing single fraction treatments when the 6DoF feature has been commissioned and fully functional.

Objectives: Cerebral metastases located close to eloquent brain are one of the biggest challenges in neuro-oncology. The use of adaptive hypofractionated gamma knife radiosurgery might prove beneficial to patients that are not suitable candidates for both surgical intervention and single fraction radiosurgery. By dynamically adapting peripheral prescription doses and tumor bed dose distributions to ongoing tumor volume changes over the course of the treatment, this technique applied in next-to emergency settings aims to achieve rapid results in terms of tumor volume reduction whilst minimizing the risk of radiation induced side effects. The aim of our study is to report the short term results (at 1 and 4 weeks after treatment completion) in patients treated at our institution, with focus on tumor ablation, salvage of organs at risk and toxicity.

Methods: 23 patients with 26 critically located brain metastases, treated over 7 days by means of three (3) separate Gamma Knife radiosurgeries (GKRS)  between November 2013 and November 2016, were retrospectively analyzed in terms of tumor volume decrement at 1 week (between GKRS 1 and 3) and at 4 weeks after GKRS 3 (first follow up MRI). Karnofsky score, RPA, perilesional edema, cortisone treatment and salvage of organs at risk were taken into consideration.

Results: Mean peripheral doses at GKRS 1, 2 and 3 were 7,8 Gy, 8Gy and 8,5Gy (range 6-9 Gy) respectively. Mean tumor volume depletion at one week was 11%. Five patients died prior to first follow-up. In the surviving group (18) mean tumor volume reduction was 61% at 5 weeks. No treatment related side-effects were observed.

Conclusion: In this study, adaptive hypofractionated gamma-knife radiosurgery proved effective in achieving rapid tumor volume reduction in critically located cerebral metastasis with limited toxicity.

Objective: To evaluate hippocampal doses and target volume coverage with and without hippocampal-sparing when treating multiple brain metastases using various stereotactic radiosurgery (SRS) platforms.

Materials and Methods: We selected 10 consecutive patients with a total of 14 separate treatments who had been treated in our department for 3-10 brain metastases and added hippocampal avoidance contours. All 14 treatments were planned with GammaPlan® for Gamma Knife®, EclipseTM for single isocenter volumetric modulated arc therapy (VMAT), TomoTherapy® Treatment Planning SystemTM for TomoTherapy®, and MultiPlan® for CyberKnife®. Initial planning was performed with the goal of PTV coverage of V100>95% without hippocampal avoidance. If the maximum hippocamal point dose (Dmax) was <6.6 Gy in a single fraction and <40% of the hippocampi received ≤4.5 Gy, no second plan was performed. If either constraint wasn’t met, replanning was performed with these constraints.

Results: There was a median of 6 metastases per plan, with an average total tumor volume of 7.32 mL per plan. The median hippocampal Dmax (in Gy) without sparing averaged 1.65, 9.81, 4.38, and 5.46, respectively (p<0.0001). Of 14 plans, three Gamma Knife and CyberKnife plans required replanning, while 13 VMAT and eight TomoTherapy plans required replanning. The hippocampal constraints were not achievable in one plan on any platform when the tumor was bordering the hippocampus. The mean volume of brain receiving 12 Gy (in mL), which has been associated with symptomatic radionecrosis, was 23.57 with Gamma Knife, 76.77 with VMAT, 40.86 with CyberKnife, and 104.06 with TomoTherapy (p = 0.01). The overall average conformity indices for all plans ranged from 0.36 to 0.52.

Conclusions: Even with SRS, the hippocampi can receive a considerable dose; however, if the hippocampi are outlined as organs of risk, sparing of these structures is feasible in nearly all situations with all four platforms, without detriment to target coverage, and should be considered in all patients undergoing SRS for multiple brain metastases.

Background: To assess clinical outcome using linac-based fractionated stereotactic radiotherapy (FSRT) for the large brain metastasis (LBM) unsuitable for surgical resection.

Material and methods: Between January 2009 and October 2016, we have treated 26 patients who had a LBM by using linac-based FSRT. LBM was defined as larger than 25mm with maximum diameter of axial, coronal or sagittal image of gadolinium enhanced magnetic resonance images. This study included the patients who were unsuitable for general anesthesia and did not want to have an operation. The median age of these patients was 72.8 years (range 47-87 years). Fifteen patients were male, eleven patients were male. LBM originated from the lung (n=21, 80.7%), the ovary (n=2, 7.7%), the gastrointestinal tract (n=2, 7.7%), and the breast (n=1, 3.8%). Pretreatment KPS was 61.1 (range 50-90). No patient was in RPA class 1, ten patients (38.5%) were in class 2, and sixteen patients (61.5%) were in class 3.

Results: The median follow-up was 9 months (range 1-52 months). Tumor volume ranged from 6.23 to 26.5cm³ (median 13.7cm³). All patients were treated with FSRT using linear accelerator with micro-multileaf collimator having a high spatial resolution of 3mm (Novalis, BrainLAB, Germany). 35 Gy in five fractions was basically prescribed in twenty patients, 40 Gy in ten fractions in two patients, 42 Gy in six fractions in one, 36 Gy in ten fractions in one, and 30 Gy in three fractions in one. The median survival time was 7.0 months (95% confidence interval: 2.660-11.340). Fifteen patients died of systemic disease and two patients died of intracranial brain metastasis. Grade 2 radiation necrosis was observed in two patients (7.7%), Grade 3 in three patients (11.5%) according to the Common Terminology Criteria for Adverse Events version 4.0. Among them, three patients underwent the administration of bevacizumab for radiation necrosis. One patient underwent the additional surgical removal for local progression.

Conclusion: Linac-based FSRT was effective and safe non-invasive treatment for large brain metastasis unsuitable for surgical resection with limited toxicity.

Purpose

Stereotactic radiosurgery (SRS) is the preferred primary treatment option for patients with a limited number of asymptomatic brain metastases. In case of a central nervous system (CNS) oligo-progression after initial SRS the optimal salvage treatment is not well defined.

The aim of our retrospective analysis was to investigate the feasibility and toxicity of repeated courses of SRS to defer or avoid Whole-Brain Radiation Therapy (WBRT).

Material and Methods

From 2014 until 2016, 23 patients with a total of 114 brain metastases have been treated with multiple courses of SRS at our institution. Treatment was delivered as single fraction (18 or 20Gy) or hypo-fractionated (6 fractions with 5Gy) radiosurgery prescribed to the 80%-isodose line. Regular follow-up included clinical examination and contrast-enhanced cMRI at 3-months intervals were performed.

Results

Median Diagnosis-specific Graded Prognostic Assessment (ds-GPA) at time of first SRS course and repeated SRS courses were 2.5 (range, 1-4) and 2 (range: 0-3.5) respectively. A median number of 1 lesion (range: 1-13) per course and a median of 2 courses (range: 2-4) per patient were administered. The median interval between SRS courses was 4.5 months (range: 0.8-18.5 months). Median tumor volume was 0.65cm³ (range: 0.1-32.5cm³).

With a median follow-up of 12 months (range: 4-25.5 months) a local control rate of 90% (95%CI: NA) was observed at 1 year independent of treatment course. Median time to out-of-field-brain-failure was 3 months (95%CI 2-4 months). WBRT as a salvage treatment was required in only 5 patients (21.7%) after median interval of 6.6 months (range: 5.5-10 months). Of the 18 patients who received repeated SRS only, 11 were still alive at time of last follow-up. 5 of the 10 deceased patients had died of neurological death, 3 of them had received WBRT previously. Median overall survival (OS) was 22 months (95%CI: 15-NA). No grade 4 or 5 acute or late toxicity occurred. Acute and late toxicity grade 3 was observed in only 3, or 2 patients respectively.

Conclusion

In selected patients with a relapse of brain metastases amenable to SRS repeat courses of SRS can safely be administered with excellent local control and low toxicity. Despite high rates of distant cranial recurrence, the majority could be salvaged by repeat SRS. With an encouraging OS in our cohort, potentially neurotoxic WBRT could be avoided in significant proportion of patients. 

Aim: The aim of this study was to evaluate the performance of stereotactic radiotherapy (SRT) with either CyberKnife (Accuray, Sunnyvale, CA) or Vero™ (BrainLab Mitsubishi Feldkirchen, Germany) in term of toxicity  and local control (LC) of cranial o extracranial metastasis in oligometastatic renal cell carcinoma (RCC)  patients (pts).

Materials and methods: Between January 2012 and September 2015, 26 patients (overall 33 metastatic lesions) with metastases of RCC were treated. SRT was performed on a singular new site lesion or on residual disease during a systemic therapy. Disease control was evaluated with serial imaging. Toxicity was recorded according to the Common Toxicity Criteria version 4.0 and reported as early and late. The total radiotherapy doses ranged from 10 to 54 Gy,  given in 1 to 3 fractions. The biological equivalent doses (BED) and 2Gy-per-fraction equivalent dose (EQD2) were calculated using the α/β=10 for tumors. We stratified the patients into 3 groups (gr) according to the EQD2 delivered group 1  range 12-36 Gy; group 2  range 42-50 Gy; group 3  66-126 Gy.

Results: After a median follow-up of 14 months (range 1-36), we achieved the best result in gr 3 ( CR and SD in 77.7% and  22.3 % of pts) versus 60 % and 40 % in Gr 2 and 50 and 50 % in Gr 1. Only out-field progression of disease was observed in 8 cases after a median 5.1 months from SRT. No toxicity was registered. Eleven pts had more than 12 months follow-up: 11/11had complete response in the site of treatment and two of these had PD in other site (2 pts are of Gr 1).Clinical and radiological response was thus evaluated in all patients

Conclusion: SRT is a feasible approach in oligometastatic RCC patients with an excellent LC and a favorable  toxicity profile. SRT could play a role in the therapeutic strategy for these patients allowing them a delay in the start of a systemic therapy and its toxicity or a drug holiday after a long treatment period. Further studies are warranted to increase the quality of evidence of such composite approach .

Objective:

To report the outcome of patients with brain metastases (BM) treated with radiosurgery (SRS) or hypofractionated stereotactic radiotherapy (HFSRT) alone to the surgical cavity, with a particular aim to determine the benefit of regular MR follow-ups regarding outcome.

Methods:

We retrospectively analyzed patients with one to three BM, who underwent robotic and linac based SRS and HFSRT to the tumor bed after surgical resection, between January 2010 and November 2015 at our department. Additional BM were managed with SRS. Patients with prior whole brain radiotherapy (WBRT) were excluded. Overall survival (OS) was estimated by the Kaplan-Meier method and prognostic factors examined with log-rank test and Cox proportional hazards model. Primary endpoint was local control (LC), secondary endpoints were distant brain control (DBC), OS and benefit of MRI follow up on outcome.

Results:

A total of 75 Patients with 77 treated surgical cavities were enrolled. Median follow-up time was 15.8 months (range: 2-69 months). Single, 2 and 3 BM were seen in 65 (86%), 9 (12%) and 1 (2%) patient. Most frequent histologies were lung (37%) and melanoma (13%). Mean target – volume was 13cc (range 1.8-45.5cc). Most patients (n=60, [80%]) were treated with a single fraction (fx). Five, 6 and 10fx were given in 3 (4%), 3 (4%) and 9 (12%) patients (mean = 2.4fx). Mean single and total dose (SD and TD) were 15Gy (range 4-20Gy) and 21Gy (range 15-40Gy). Median OS was 19.4 months (95% confidence interval (CI): 12-26 months). Actuarial LC, DBC and OS at 1 year were 72% (95% CI: 60%-83%), 60% (95% CI: 48%-72%), and 66% (95% CI: 53%-76%) respectively. Median number of follow up MRI scans at six months was n=1 (range 0-2). No median value was reached for local failure. Median distant brain failure occurred at 16.2 months (95% CI: 4.5-28 months). Salvage treatment was needed in 45 patients (WBRT n=15, surgery n= 15, SRS n=15). The number of MRI follow-up scans at six months did not have an impact on OS (p= 0.56), but it had a significant association on DBC (p<0.035).

Conclusion:

Patients with one to three BM treated with SRS and HFSRT alone to the surgical cavity and regular follow-up with MRI scans have an excellent outcome. The impact of regular MRI scan follow-up on DBC deserves further prospective evaluation with an additional focus on its financial impact on cost-effectiveness. 

OBJECTIVE:

The second-generation multi-leaf collimator (MLC2) for robotic radiosurgery and robotic stereotactic body radiotherapy (SBRT) has been recently introduced. We compared plan qualities and dose-parameters, which are specifically relevant in clinical routine use of robotic liver SBRT, generated using the MLC2 and variable aperture collimator (VAC).

METHODS:

Planning scans and outlines of 10 patients with one to three liver lesions treated in our clinic with VAC to 25-30Gy in 3 to 5 fractions with robotic SBRT were retrieved. Treatment plans were generated for all lesions with MLC2 and VAC within Multiplan 5.3, using previously applied doses and fractionation. Doses were prescribed to the 67-70% isodose. Optimization objectives included: clinically applicable treatment plans with planning target volume (PTV) coverage ≥97%. Dose constraints to OAR were defined according to Grimm et al. 2011 (1).  For the MLC2 plans, the optimization script (input for optimization) was chosen as similar as possible as for VAC plans. The following parameters were analyzed: minimum, mean and maximum dose of PTV, liver, liver-GTV, kidneys, intestines, spinal cord, lung and heart, Nakamura conformality index (nCI), homogeneity index (HI), treatment time per fraction - including setup time of 15min and 60sec. imaging interval - number of beams and monitor units (MUs) per fraction. Statistical analysis was performed with STAT/SE 14.1, using the t- test to compare plan quality metrics.

RESULTS:

MLC2 plans did not differ significantly in conformality (p= 0.87) and homogeneity (p=0.33) compared to VAC plans. Conformality index was 1.17 ± 0.06 and 1.17 ± 0.05 and HI was 1.43 ± 0.06 and 1.43 ± 0.00 for the MLC 2 and VAC plans respectively. No statistically significant difference in mean dose to the PTV was found (p= 0.9543). OAR dosimetric parameters (Dmin, Dmean, Dmax, D75, D50, D25, D10, D1cc) did not differ significantly between plans (p>0.50). Mean MUs per fraction for the MLC2 and VAC plans were 6830 and 7018, respectively (p= 0.86). There was a statistically significant difference for the minimum dose to PTV in MLC2 plans (p<0.01), number of beams (mean 55 beams MLC2 vs. VAC 161 beams, p<0.0001), and delivery time including setup time of 15min and 60sec. imaging interval (median time MLC2 43min vs. IRIS 59min, p< 0.0001) regardless of single versus multiple lesions.

CONCLUSION:

MLC2 produces treatment plans of similar quality as the VAC system with much faster treatment times and significant fewer beams for the clinical use of robotic liver SBRT.

Metastatic pancreatic cancer is characterized by poor prognosis. Only appearance of FOLFIRINOX increased the median survival of 11 months. However, high toxicity limits its use in elderly patients with severe comorbidities. Meanwhile, in patients with metastatic disease, local treatment didn`t show positive effect on life expectancy.

We present clinical case of 72 years old woman with metastatic pancreatic adenocarcinoma T3N0M1. CT revealed tumor in head of pancreas 40x61x47 mm and two lesions in the liver. 4 cycles of Gemcitabine+Erlotinib were performed, but due to progression second-line chemotherapy with CAPOX was started. Despite partial response after 4 cycles, chemotherapy was stopped because of toxicity. Due to high risk of progression, it was decided to perform consolidating radiosurgical treatment.

SBRT was performed with the use of Cyberknife. In order to monitor displacements of lesions during respiration, 1 radiopaque marker was implanted in the tumor of the pancreas and 4 in the liver 14 days before treatment. Two separate irradiation plans (one for pancreatic lesion and one for liver lesions) were prepared. Dose to pancreatic and liver lesions was 35Gy in 5 fractions. Pancreatic GTV was 26,3 cm³, 3 mm margin was added for PTV. Dose for all critical structures was acceptable in both plans. Nevertheless, during evaluation of the sum of two plans it was revealed, that dose to stomach and duodenum was outside possible tolerable dose. In order to avoid complications but to save the treatment dose and not to lose antitumor effect, it was decided to make a four week brake between two treatments.

During first follow-up 3 months after treatment shrinkage of all 3 lesions was described on MRI. After 9 months only one liver lesion and primary pancreatic tumor, stable in size were determined. At present time patient is alive and in good condition, disease is stable 45 months since diagnosis and 37 months after SBRT. No any other cancer treatment was performed after SBRT. No Grade II-IV adverse events were observed. Grade I diarrhea developed 3 months after irradiation and is controlled with loperamide.

Metastatic pancreatic cancer still has a poor prognosis, but combination of systemic and local treatment can potentially increase survival. SBRT provides a high level of local control, does not require interruption of drug therapy and is not accompanied with surgical trauma and high risk of postoperative complications. Further research is needed to determine the role of SBRT in the treatment of oligometastatic disease.

Background:          Treatment of multiple brain metastases (BM), without inducing cognitive side effects, is becoming more important, as more patients live longer due to better treatment options. Whole brain radiation therapy (WBRT) has long been the mainstay of treatment for patients with multiple BM. Meanwhile, Gamma Knife radiosurgery (GKRS) is increasingly applied in these patients and is expected to have less adverse effects on cognitive functioning than WBRT. Cognitive functioning is a collective term for cognitive skills such as memory, attention and concentration. There are no published randomized trials yet directly comparing GKRS to WBRT in patients with multiple BM, including objective neuropsychological testing.

Methods:              CAR-Study B is a randomised trial comparing cognitive outcome after GKRS or WBRT in adult patients with 11-20 newly diagnosed BM (on a triple-dose contrast-enhanced MRI-scan), KPS ≥70 and life expectancy of at least 3 months. Before randomisation, patients complete a comprehensive neuropsychological test battery to assess cognitive functioning (HVLT-R, COWA, WAIS Digit Span and Digit Symbol, TMT A and B, Grooved Pegboard). In addition, patients fill-out 3 questionnaires concerning health related quality of life (FACT-BR), depression, anxiety (HADS), and fatigue (MVI). Randomisation is stratified by the cumulative tumour volume in the brain, systemic treatment, KPS, histology, baseline HVLT-R Total Recall score (a memory task) and age.

Follow-up for both groups is scheduled at 3, 6, 9, 12 and 15 months after treatment with cognitive testing, questionnaires and MRI-scans. The primary endpoint is the between-group difference in the percentage of patients with significant memory decline at 3 months. Significant decline is defined as a 5-point decrease from baseline in HVLT-R Total Recall score, based on the Reliable Change Index (RCI) with correction for practice effects.Secondary endpoints include cognitive functioning over time, health related quality of life, depression and anxiety, fatigue, overall survival, local control and development of new brain metastases.

A power analysis based on Bayesian stopping rules suggest that the trial will be halted  when 46 patients (23 in both groups) are enrolled.  An independent statistician will perform monitoring of this trial using Bayesian statistical methods.

Discussion:            This trial is currently recruiting. Results of this study are relevant in clinical decision-making and may be used to inform individual patients with BM more precisely about the cognitive effects they can expect from treatment. Knowledge gained from this trial may be used to change clinical practice (in the Netherlands).

Trial registration: Clinicaltrials.gov identifier: NCT02953717

Objectives

In the absence of trial data, SBRT is reported to safely prolong progression free survival in patients with prostate cancer oligometastases. 18-F choline is sensitive and specific for prostate cancer oligometastases outside the pelvis and a CT or MRI correlate assists in target delineation for SBRT. We report the outcomes of two patients with challenging patient and tumour factors. 

Methods

Patient 1: A choline PET/CT revealed a single painless oligometastasis in T2 in a 59 yr old patient with biochemical relapse (PSA 3.95 ug/l) 6 years after radical treatment of a Gleason 8 prostate cancer. Prior gastroplasty resulted in frequent hiccups. A head rest, wing board, mat and knee rest were used for immobilisation. 5 x 4 Gy with a 5 x 8 Gy integrated boost was prescribed analogous to the DOSIS study protocol and delivered with VMAT.

Patient 2: A 76 yr old patient with PSA 13.7ug/l 6 years after radical treatment of a Gleason 8 prostate cancer was found to have two 18F-PET positive lymph nodes (7.5 mm diam at L4 and 12.2 mm diam right common iliac) within 5 cm of each other. A head rest, wing board, mat, knee rest and chest/abdominal belt were used for immobilization. 3 x 15 Gy was delivered with VMAT within the NRG BR001study due to unestablished normal tissue constraints in this situation. Androgen deprivation therapy had already been started in both patients.

Results

Patient 1: SBRT was delivered with a mean intrafraction motion vector of 0.55 mm (0.45 -0.86 mm) despite hiccups during irradiation. 8 weeks after SBRT, the patient remained asymptomatic. A follow-up CT scan is planned 4 months after SBRT.

Patient 2: SBRT was delivered with a mean intrafraction motion vector of 0.53mm (0.01-0.99mm). Mean dose to the aorta was 6.2 Gy (min. 0.16-max. 49.3 Gy) and to the inferior vena cava was mean 11.5 Gy (min. 0.18 Gy – max. 47.6 Gy). Mean dose to the small bowel was 2 Gy (min. 0.07 Gy – max. 26.9 Gy). The patient remains on study follow-up.

Conclusions

SBRT for oligometastases is reproducible and is well tolerated but should be delivered in clinical trials where available. The high sensitivity and specificity of the new tracer Ga-68 PSMA may assist in the diagnosis of prostatic oligometastases. The recent decision to reimburse PSMA PETCT scans in Switzerland is likely to increase future referrals for SBRT for limited prostate cancer metastases.

Objective: Radiosurgery is an emerging therapeutic approach for the treatment of brain metastases (BMs). Different technological modalities have been used. We draw this phase III trial with the aim to evaluate incidence of symptomatic radionecrosis using gammaknife versus linac-based (EDGE) radiosurgery. Local control and overall survival  (OS) were assessed.

Materials and Methods: Patients with up to 4 BMs were enrolled. Inclusion criteria were: KPS ≥70, RPA class I-II, BMs maximum diameter ≤3 cm and/or with a total tumor volume <30 cm³. Prescribed  dose was 24 Gy or 20 Gy depending on lesion dimension. Clinical outcome was evaluated by neurological examination and MRI .

Results:  From October 2014 to December  2016, 101 patients of the expected 250, for 167 BMs were evaluated. Most common primary was NSCLC (55.4%). Symptomatic radionecrosis was observed in a total of 12 (11.9%) cases of the entire cohort evaluated. In the gammaknife ARM Grade II radionecrosis was recorded in 5 cases and Grade III in 2 cases.  In the EDGE ARM only grade II radionecrosis occurred in 5 cases treated. No local progression in site of SRS occurred. At a median time of 11 months, 41.5% patients had new distant brain metastases, and 58% had extracranial progression at a median time of 8 months. The 6 months and 1 year OS were 81% and 70%, comparable in both arms.

Conclusion: Gamma-knife and LINAC based SRS for BMs were comparable in terms of LC. In this preliminary evaluation the risk of symptomatic radionecrosis was greater in the gammaknife  arm.  

Purpose: To assess the safety and efficacy of Stereotactic Body Radiotherapy (SBRT) in patients with liver metastases from colorectal cancer (CRC)  unsuitable for surgery and radiofrequency ablation (RFA).

Matherials and Methods: Patients with 1–3 inoperable CRC liver metastases and maximum tumor diameter

Results: From February 2010 to January 2016, 90 patients for a total of 129 CRC liver metastases were analyzed. Median follow-up was 14 months (range 5–80). One, two and three years LC rates for lesions were 90%, 81% and 75%, respectively. At univariate analysis, no correlation between LC and lesions size was detected. Median OS was 29 months. Actuarial OS rate at 1,2 and 3 years was 88%, 67% and 33%, respectively.  Median PFS was 10 months, with a PFS rates of 81% at 1 year, 63% at 2 years and 40% at 3 years. No patients experienced radiation-induced liver disease or grade >3 toxicity.

Conclusions: SBRT is a non-invasive, safe and effective therapeutic option for patients with unresectable colorectal cancer liver metastases. Short treatment time, non-invasiveness, optimal tolerability of treatment joined to promising LC and OS rates represents the crucial advantages of this alternative ablative therapy for these selected oligometastatic patients.

Background: Cerebral metastasis of gynecological tumors are rare. Despite treatment, the majority of these patients do no survive >1 year due to extracranial disease. In case of brain metastases (BM) one can state that the survival is even less. Treatment options  are operation, WBRT, stereotactic radiotherapy (SRT) or stereotactic radiosurgery (SRS). The use of SRS for these BM is not well known within the medical world.

Objective: The goal of this study is to evaluate the efficacy of Gamma Knife Radiosurgery (GKRS) for gynecological BM.

Material: Between 10-6-2002 and 19-5-2015, 3000 procedures were performed at our center for BM. A retrospective analysis of this data was performed. We included all patients who received GKRS for BM of gynecological tumors. Patients with multiple primary tumors were excluded. Pre- and post-GKRS clinical characteristics were reviewed. Outcome measures were local tumor control (TC), development of new BM and/or leptomeningeal disease after initial SRS, overall intracranial progression free survival (PFS).

Results: We included 41 patients (median age 64, median Karnofski-index 80). Primairy tumors were endometrium (8), ovarian (29) and cervix carcinoma (4). A median dose of 22 Gy (range 18-25 Gy) was prescribed to the isodoseline covering 90-100% tumor volume. The mean tumor volume was 3.507 cm³ (range 0.0014-21.125 cm³). Sixteen patients (39%) had single BM and 25 (61%) had ≥2 lesions (median number per patient was 2, ranging 1-7). Eventually a total of 92 metastases (74 (80%) supratentorial and 18 (20%) infratentorial) were treated. Eleven (27%) patients had undergone intracranial treatment for the BM prior to SRS (7 resection followed by postoperative SRT, 2 biopsy and 2 WBRT). The median and mean FU time was 11 and 19 months, respectively. The TC was 100%, 92%, 88%, 76% and 67% at 2.5, 5, 7.5, 12.5 and 15 months, respectively. The PFS was 90%, 66%, 51%, 24% and 15% at 2.5, 5, 7.5, 12.5 and 15 months, respectively. During FU 18 (44%) patients had intracranial progression (4 local progression, 6 local progression with distant lesions, 6 only new BM and 2 had new BM with leptomeningeal disease). Thirty-one (76%) patients died due to extracranial tumor progression and only 2 (5%) patients died due to progressive intracranial disease.

Conclusion: GKRS is a good treatment option for controlling BM in patients with primary gynecological malignancies. As most patients die due to extracranial tumor progression, their survival might improve with better systemic treatment options in addition to GKRS. 

Multiple brain metastases are challenging to treat, especially if the number of tumors is large or if they accompanied with brain surface seeding.  In addition to whole brain radiation therapy (WBRT), stereotactic radiosurgery (SRS) to some larger lesions and/or volumetric modulated arc radiotherapy (VMAT) boost on part of brain was performed in such cases. In four cases of multiple brain metastases, boost radiation to the cerebellum, where many tumors existed, by VMAT in addition to WBRT.  In all seven cases, before or after WBRT, SRS was done to some larger lesions. Four cases were alive and three were dead with median follow-up period of 5 months (range, 1 to 18 months).  Four cases developed advanced cerebrospinal fluid dissemination in their course.

Additional tumor boost by SRS and/or area boost by VMAT may prolong the control of multiple brain metastases in advanced situations for a certain period. 

Objective: To examine the combined effects of prescription method and dosimetric margin on the quality of dynamic conformal arc plans for the treatment of single brain metastases.

Methods: This study consisted of two target sets: simulated targets and patient data.  The simulated targets included spheres of varying sized (0.5-3cm in diameter) and ellipses in varying orientations.  Data from 11 patients (12 lesions) were also included. Plans for the simulated targets were generated by using a standard 4 arc plan whereas for the patient data the beam arrangement used for treatment was also used in this study.  For all plans, the dosimetric margins were varied in 1mm increments from -1mm to 3mm; the prescription isodose line (IDS) was varied independently of dosimetric margin from 50-98% isodose surface (where 100% is the maximum point dose) and 90-99.8% target coverage. Plans were evaluated by target coverage, minimum and maximum dose to target, heterogeneity, gradient index (GI), and conformity index (CI).

Results: Increased dosimetric margin increased the minimum IDS required to provide 100% target coverage of the prescribed dose. For fixed margins, minimum and maximum dose to the target were scaled directly by the IDS and a higher IDS resulted in lower maximum dose. Heterogeneity increased with decreasing margin size and was independent of IDS. The GI increased with dosimetric margin, however, for a given margin, the minimum GI occurred at the same IDS for each target. For patient data and simulation, the minimum CI for a given margin occurred at a higher IDS as the dosimetric margin increased. As the IDS increased the dosimetric margin that gave the lowest CI also increased.  In the patient data when the target coverage was fixed to 99.8%, the 0mm margin had the lowest CI for 8 targets and the 1mm for 4 targets. When the maximum dose was also constrained to no more that 120%, the 1mm margin had the lowest CI for 9 targets.  Under these constraints the targets larger than 2cm in diameter required a larger margin of 2mm or 3mm.

Conclusions: The dosimetric margin and prescription method both impact the dose distribution and associated quality metrics.  Particularly, when it comes to finding an optimal conformity index, it cannot be assumed that fixing either the margin or prescription IDS to a predetermined value will result in the most desirable plan

INTRODUCTION:

Current cancer treatments confront an increasing number of patients with metastatic brain metastases. We present the result of SRS in patients with metastases in the central lobes of the brain. 

MATERIAL:

All patients submitted to SRS (Varian-LINAC) for Brain Metastases between February 2013 and October 2016 were retrospectively studied. Seventeen patients with tumors in the Central Lobes were identified.

METHOD: Clinical symptoms; MRI pre and post - SRS of each patient; tumor volume; SRS dosimetry; MRI evidence of adverse radiation effects (ARE); Local Control and Outcome were investigated. 

RESULTS:

Mean age (years) = 59.53; Median = 62.5. Most frequent primary tumor was Non SCLC = 12. No laterality predominance. Eight patients (47.5 %) had intracranial oligometastases condition (all tumors received SRS at the same day). Motor weakness, hypoesthesia and/or seizure were present in ten patients (58.82 %). Prescription doses: Mean = 18.33 Gy. PTV: Mean = 1.79 cc. Effective local control (LC) = 16/17 (88.23 %). Partial LC (>50%) in 5/17 patients (35.29%), coincident with larger PTV or shorter period of follow up. No patient was lost during the follow-up of the series. Salvage surgery of local relapse was performed in only one patient. ARE in the MRI follow up was present in six patients: four of them had brain edema in gyri contiguous to the metastases. Four patients developed epilepsy after SRS, in the absence of tumor growth, all presented gyral edema in MRI pre SRS; three of them had PTV > 4 cc  and V12 > 6.50 cc. One patient presented worsening of motor function after SRS, which improved with steroids medication. All the rest of patients with normal motor function preserve it after treatment. Three patients presented a distant relapse. A second SRS was applied in all cases. The mean overall survival was 14.41 months (range 3 – 39), median of 11 months. None patient died as result of intracranial disease.

CONCLUSIONS:

Motor deficit prior to SRS and tumor’s presence within the Central Lobe did not correlate with worsened motor function following SRS. Prescription doses ≥ 18 Gy were used in the search of better tumor control; PTV and V12 values were also determinant at the time of dose selection. Pre-SRS edema does not seem increase the risk of developing motor dysfunction. Regarding complications, epilepsy was more frequent than motor weakness. The occurrence of new lesions does not impact overall survival if salvage therapy is offered.

The aim of the study was to assess the quality of plans and comparison of the dosimetric parameters of dose distributions for the radiosurgery of multiple brain metastases ( N≥4) on CyberKnife (CK) and Gamma Knife (GK).

Materials and methods. 13 patients with 68 brain metastases previously treated on the CK radiosurgery system were selected. Then contours of targets and critical structures according to CT and MRI data were transferred from MultiPlan to Leksell GammaPlan for reproducing treatment plan. The value of dose, covering 100% of target volume, was equal for both modality. Following characteristics were calculated from dose-volume histogram (DVH): homogeneity index (HI), coverage (CO), conformity indexes (CI_RTOG, nCI), gradient indexes (GI₅₀, GI₂₅) and doses in normal tissues (V₁₀, V₁₂, V_10net, V_12net). The original software was developed for the automate DVH processing .

Results. Following results we achieved while comparing indicators calculated for two SRS system.  Depending on the targets volume: for volume [0,01;0,1) cm³ statistically reliable difference for CI_RTOG, nCI was no obtained (CI_RTOG: 1,523 ± 0,094 (GK) vs 1,623 ± 0,166 (CK), p=0,35; nCI: 1,535 ± 0,092 (GK) vs 1,623 ± 0,166 (CK) ,p=0,35); for volume [0,1;6]  cm³ GK dose distribution is more conformal (nCI: 1,262 ± 0,075 (GK) vs 1,313 ± 0,079 (CK),p<0,05). For all volumes of target CK gives more homogeneous dose distributions in target (HI: 62,43 (35,18 – 81,48) % (GK) vs 18,25 (13,82 – 24, 30) % (CK), p <0,001); the GK dose heterogeneity in target is accompanied by high dose gradient out of the target and lower V₁₀ and V_10net values. Independently on the total number of lesions (N): nCI values are equal for CK and GK (1,211 ± 0,036 (GK) vs 1,332 ± 0,066 (CK) , p=0,25), however CI_RTOG conformity is better on GK (1,151 ± 0,023 (GK) vs 1,332 ± 0,066 (CK), p<0,05). For all N dose distribution in target is more homogeneous on CK (62,22 (18,02 - 78,04) % (GK) vs 21,93 (20,16 - 23,75) % (CK),p< 0,001), while dose gradients are higher on GK allowing more effective irradiation dose decreasing in brain.

Our purpose is to address physics challenges of implementing a single isocenter radiosurgery approach for treating multifocal intracranial disease.  The technique utilizes a radiosurgery specific thermoplastic mask for immobilization, and VMAT with HD-MLCs (2.5mm leaf-width at isocenter) for treatment planning.  Physics aspects that we investigated during the implementation include: (1) quantifying and managing spatial uncertainty associated with image guidance and immobilization, (2) investigating VMAT treatment planning strategies such as ideal isocenter selection, number and geometry of VMAT arcs, and choice of optimization criteria, (3) determining whether plan quality is affected when targets are located distal from the isocenter, and whether a single isocenter is sufficient for spatially dispersed targets, and (4) revisiting tolerances for routine machine quality assurance based on the potential dosimetric effect.  Our analysis showed that after 6D correction of initial setup error, a 1mm margin was still needed to compensate for spatial uncertainty within the mask.  Conformity and gradient indices improved with decreasing distance from isocenter, with this trend being more pronounced for targets < 1 mL. Alternative isocenters aimed at decreasing distance of small targets improved their gradient index, but resulted in poorer dose indices for large targets.  Yet when targets were spatially dispersed, utilizing multiple isocenters resulted in only modest improvements in plan quality, thus indicating that a single isocenter is sufficient for most cases.  Optimal treatment plan quality could be achieved when no constraint was placed on the maximum target dose; for cases with many targets at least 4 arcs were needed for optimal plan quality.  Our investigation of routine machine QA tolerance found that in order to avoid potentially unacceptable dosimetric degradation to single isocenter multifocal VMAT plans, the gantry and collimator angular tolerance for routine (monthly) QA should be tightened to 0.5°.

Background: Radiation necrosis (RN) is a potential late complication of radiotherapy for CNS malignancy, often associated with neurological morbidity. Traditionally, high dose corticosteroids or surgical resection have been the standard of care treatment, but protracted steroid course can lead to significant side effects and resection is not always feasible. While there is evidence that the anti-VEGF antibody bevacizumab (Bev) can be effective for RN in other cancer types, evidence supporting its use for RN in melanoma patients is lacking, due to known risk of spontaneous bleeding in melanoma brain metastasis (MBM) and increased bleeding risk with Bev.

Methods:  We retrospectively identified 6 melanoma patients who received Bev for RN, and demographics, systemic and radiation treatment history were obtained.  Diagnosis of RN was established by MRI in all patients, and pathology confirmed RN in one patient.

Results:  We present 6 melanoma patients who developed RN after CNS radiation and who were treated with 2-6 doses of Bev. Median age was 58 years-old (rang 52-71). Two patients received SRS (stereotactic radiosurgery) alone, 2 patients received WBRT (whole brain radiation therapy) alone, and 2 patients received SRS plus WBRT. Four patients received current or prior immunotherapy. Radiation necrosis was diagnosed 5 months after completion of radiation (range 4-7 months).  None of these patients developed bleeding, and all patients experienced improvement in symptoms and quality of life, with concurrent improvement of imaging in 5 patients.

Conclusions: Our series suggests that in selected melanoma patients Bev may be a safe and effective treatment for RN for MBM and should be evaluated in a prospective setting.

Objective The authors present a retrospective analysis of a single-center experience with treatment of brain metastases using Gamma Knife (GK) and linear accelerator (LINAC)–based radiosurgery and compare the results.

Methods From July 2010 to July 2012, 63 patients with brain metastases were treated with radiosurgery. Among them, 28 (83 lesions) were treated with a GK unit and 35 (47 lesions) with a LINAC. The primary outcome was local progression–free survival (LPFS), evaluated on a per-lesion basis. The secondary outcome was overall survival (OS), evaluated per patient. Statistical analysis included standard tests and Cox regression with shared frailty models to account for the within-patient correlation.

Results The median number of lesions per patient was 2.5 (range 1–9) in the GK group and 1 (range 1–3) in the LINAC group (p < 0.01, 2-sample t-test). There were more radioresistant lesions (e.g., melanoma) and more lesions located in functional areas in the GK group. Additional technical reasons for choosing GK instead of LINAC were limitations of LINAC movements, especially if lesions were located in the lower posterior fossa or multiple lesions were close to highly functional areas (e.g., the brainstem), precluding optimal dosimetry with LINAC. The median marginal dose was 24 Gy with GK and 20 Gy with LINAC (p < 0.01, 2-sample t-test). For GK, the actuarial LPFS rate at 3, 6, 9, 12, and 17 months was 96.96%, 96.96%, 96.96%, 88.1%, and 81.5%, remaining stable until 32 months. For LINAC the rate at 3, 6, 12, 17, 24, and 33 months was 91.5%, 91.5%, 91.5%, 79.9%, 55.5%, and 17.1% (log-rank p = 0.03). In the Cox regression with shared frailty model the risk of local progression in the LINAC group was almost twice that of the GK group (HR 1.92, p > 0.05).

Conclusions In this retrospective study comparing GK and LINAC-based radiosurgery for brain metastases, patients with more severe disease were treated by GK, including those harboring lesions of greater number, of radioresistant type, or in highly functional areas. The risk of local progression for the LINAC group was almost twice that in the GK group, although the difference was not statistically significant. Importantly, the OS rates were similar for the 2 groups, although GK was used in patients with more complex brain metastatic disease and with no other therapeutic alternative.

Purpose: We investigated the impact of optimization in dose-limiting auto-shell function on the dosimetric quality of Cyberknife (CK) plans in treating brain metastases (BMs).

Methods: We selected 19 BMs previously treated using CK between 2014 and 2015. The original CK plans (CK_original) had been produced using one to 3 dose-limiting auto-shells, one at the prescription dose (PD) level for dose conformity and others at low-dose levels (10–30% of PD) for dose spillage. In each case, a modified CK plan (CK_modified) was generated using 5 dose-limiting auto-shells, one at PD level, another at intermediate dose level (50% of PD) for steeper dose fall-off, and the others at low-dose levels, with an optimized shell-dilation size based on our experience. A Gamma Knife (GK) plan was also produced using the original contour set. Thus, a triplet data set of dosimetric parameters was generated and compared.

Results: There were no differences among CK_original, CK_modified, and GK plans in the conformity index (mean 1.22, 1.18, and 1.24, respectively; P=0.079) and tumor coverage (mean 99.5%, 99.5%, and 99.4%, respectively; P=0.177), whereas CK_modified plans produced significantly smaller normal tissue volumes receiving 50% of PD than those produced by CK_original plans (P<0.001), with no statistical differences in those volumes compared with GK plans (P=0.345)

Conclusions: These results indicate that significantly steeper dose fall-off can be further achieved in the CK system by optimizing the auto-shell function, while maintaining high conformity of dose to tumor.

Objectives:

The aim of our study was to retrospectively report on the image guided stereotactic body

radiotherapy (SBRT) in the oligo-recurrent bladder cancer. Eleven patients  treated in our

department for urinary bladder transitional cell carcinoma with lymph node or bone

recurrence were retrospectively reviewed. The endpoint was to evaluate the safety

of SBRT, proposed as an alternative to systemic treatment, and in order to defer the start

of a second line chemotherapy.

Methods:

Inclusion criteria for our retrospective study were as follows: adult oligometastatic bladder

transitional cell cancer (TCC) patients with lymph node or bone recurrence  that underwent

SBRT but not other local/systemic therapy. Previous  radiotherapy, concomitant systemic

 therapy or  surgery on the primary tumor were allowed. Previous systemic therapy or

cystectomy was performed in all  patients.  The median treatment dose was 25 Gy (range,

20- 30 Gy) given over a median  of 5  fraction (range, 3-15 fraction).Toxicity and tumor

response were evaluated using RTOG/EORTC and RECIST criteria.

Results:

Eleven patients for a total of 19 lesions were treated with CyberKnife or Vero System-

SBRT between 2012 and 2016 in our Division. Median age at SBRT was 65.3 years

(range 45-80) and  Karnofsky performance status (KPS) was 90 (range 70-90).Mean

interval between TCC diagnosis and the first RT treatment was 3.1 years. Median follow-

up was 20.3 months. Radiological response evaluated at the first imaging assessment

was: complete response, partial response, local progression and not evaluable 11, 1, 6 

and 1 lesions, respectively.

The radiological progression of disease was registered in 7 patients at the median of 8.2

months (range 2.3-18.5 months) from the end of SBRT ; in 6 cases it was out-field and in-

field progression, while in one patient an only  out-field progression was observed.

At present (January 2017), 3 pts are alive  with no evidence of disease (median of 20.1

months from the end of SBRT), 2 alive with evidence of disease, 4 dead of cancer related

disease and 1 was lost at the follow-up. No severe acute (except for Grade 1 in 1

patient) and late toxicity were observed.

Conclusions:

SBRT on lymph node or bone recurrence from TCC offers a good in-field tumor control

with very low toxicity profile. In small proportion of patients the starting of another systemic

therapy was deferred with a reasonable control of disease. Further study are needed  to

establish a role of SBRT in the oligometastic recurrent bladder cancer.

Purpose: Although the possible impact of radiation therapy (RT) on neurocognitive functioning and quality of life (QoL) has become an important issue when choosing treatment for patients with brain metastases, related data are still scarce. The aim of this study is to evaluate neurocognitive and QoL outcome in patients treated with RT for brain metastases.

Method: Patients had neurocognitive evaluation and QoL testing before RT, 3 and 6 months after RT the first year, and then every 6 months. Mini mental state examination (MMSE), Trail Making Test (TMT A and B) and Clock Drawing Test (CDT) were used for neurocognitive assessment and EORTC QoLC30 and BN20 for QoL assessment. Differences in baseline points were analyzed with ANOVA test.

Results:77 patients were enrolled (M/F=31/46) in the last two years, with median age 58y (range 28-77y). 64% of pts were RPA class 1 and 36% class 2, 39% had GPA score between 0.5-2 and 61% between 2.5-4. Primary tumor location was lung in 43%, breast in 30%, melanoma in 12% patients. 51% of pts had no neurological symptoms at diagnosis, 16% of symptomatic patients had headache. 55% of pts had 1 lesion, 13% and 16% respectively 2 and 3 lesions. 61% of pts had cortical and 38% subcortical localizations. At presentation, a surgical approach was chosen in 40% of cases. 62% of pts received whole brain irradiation (WB). 9 of them were also administered simultaneous integrated boost and 2 had stereotactic radiosurgery (SRS) after WB. 38% of pts had SRS without WB, and 19% received concomitant chemotherapy. 69% of pts received steroids and 26% had anti-epileptic drugs before RT. During treatment, 12% and 15% of pts reported respectively G1 and G2 headache. G1 and G2 fatigue was recorded respectively in 27% and 15% of the cases. The use of steroids increased during RT in 15 pts. Median overall survival for pts treated with WB vs SRS was 209 days (range 3-1096) and 257 days (range 68-1087) respectively. Baseline MMSE was between 24-30 (no neurological alteration) in 94% of pts, 4% had a score of 20-23 (suspected neurological alteration), none scored under 20. Baseline MMSE and CDT mean points did not differ in pts treated with WB vs SRS. Mean TMT A, B, A+B in WB and SRS groups were 65, 158, 92 and 44, 91, 46 respectively.

Conclusions: The mature data and the neurocognitive functioning after RT will be presented.

The Gamma Knife Unit at the Department of Neurosurgery, Medical University Vienna looks back on 25 years of experience in radiosurgical treatment. In this timeframe several thousand patients with brain metastases (BMs) have been treated. Since the implementation of the Gamma Knife Perfexion® in 2012 the modern radiosurgical treatment era has commenced. Apart from developments in the radiosurgical treatment there has been significant progress in the oncological management of cancer patients since the introduction of immunotherapy and personalized targeted therapy. So far, even though concurrent treatment with radiosurgery and targeted drugs or immunotherapy is increasingly performed, available safety information is scarce. We have recently established a database of all patients with brain metastases treated in the modern radiosurgical and oncological era from 2012 onwards. So far the database includes 830 patients and 3079 radiosurgically treated BMs. In the modern radiosurgical era melanoma has presented itself as the second most frequent primary tumor of BMs (15%), at least in our series. Especially melanoma patients are increasingly treated with immune therapy. We provide radiological and clinical outcome data of patients with BMs treated in the modern radiosurgical and oncological era. We especially focus on complications after radiosurgery in patients with immune therapy or targeted therapy.

Introduction

Increased sophistication in machine-learning algorithms and artificial intelligence has begun to reveal patterns that were otherwise unappreciated in clinical medicine.  Here we applied one such algorithm, Factorial Analysis of Mixed Data, to better understand combinations of clinical variables that influence survival of brain metastasis (BM) patients treated with stereotactic radiosurgery (SRS).

Methods

A dataset of 6,326 BM patients were collated from four SRS centers (University of California, San Diego, Katsuta Hospital Mito GammaHouse, Tsukiji Neurological Clinic, and Melanoma Institute of Australia). We ran three models using Factorial Analysis of Mixed Data, each including one of the following in model construction: died before 3 months, survived for 1 year, or survived for 2 years. The other clinical variables included were age, Karnofsky Performance Status (KPS), cumulative intracranial tumor volume (CITV), total number of metastases, histology (breast, gastro-intestinal (GI) cancer, renal cell carcinoma (RCC), melanoma, and lung cancer), and systemic disease control.

Results

Our unsupervised machine-learning algorithm defined three groups of SRS-BM patients with different survival profiles. Dying within 3 months of SRS treatment was most significantly correlated with the group characterized by lower KPS, poor systemic disease control, higher CITV, higher number of metastases, and the diagnosis of GI cancers. Surviving beyond 24 months of SRS treatment strongly correlated with the category consisting of subsets of lung and breast cancer patients with higher KPS, controlled systemic disease, and lower CITV. A third category emerged from the analyses that consisted of young breast cancer patients with systemic disease control, but survival was not strongly correlated with this group. This group of patients was also defined independently of KPS, CITV, and the number of metastases. 

Conclusion

Clinical survival after SRS for BM is defined by combination of known prognostic factors. A prognostic factor critical for survival prognosis in one sub-population of BM patients may bear little relevance in another. 

Purpose:

This work aims to raise and evaluate the issue of beam geometric optimization in multiple targets using single isocenter technique.

Method:

OCTAVIUS 4D phantom (PTW, Freiburg, Germany) CT image set was utilized in this study. Seven targets of 2cm-diameter spheres located at central (C), right (R), left (L), anterior (A), posterior (P), superior (S) and inferior (I) and 5 cm apart from the central target. 2 mm margin was given to each target to create planning target volume (PTV) and prescribed 16 Gy to each PTV. Two single isocenter volumetric arc therapy (VMAT) plans with different collimator angles (20° and 70°) in Monaco v5.0 (Elekta AB, Stockholm, Sweden) were generated so that some of the targets were parallel to leaves direction during the trajectory. The plans were then delivered on Elekta Agility and measured by PTW OCTAVIUS 4D system and 1500 array detector. In order to obtain higher resolution, the plans were re-measured with 5 mm detector shifted. 3-dimenional (3D) dose distribution was reconstructed in Verisoft v7.0 and 3D gamma analysis (3D-γ) at 2%/2mm/local-dose of 40x40x40 mm³ region of interest for each target were performed. The measurements were further reconstructed to provide the dose-volume-histogram information of each target. Independent t-test was used to compare the γ-index and percentage change in PTV coverage (%PTV) between two groups (MLC-direction and nonMLC-direction) with statistical significant at p<0.05.

Results:

The average γ-indices of MLC-direction and nonMLC-direction were 91.2% and 96.3% respectively, while -13.3% and -5.2% were found for %PTV with both p<0.05.

Conclusion:

The targets parallel to MLC have lower γ passing rates and more loss in PTV coverage. The loss in PTV coverage for all targets indicated that the complexity of the plans. Therefore, gantry and collimator angles should be carefully selected for this advance treatment technique.

Objective

To assess the efficacy of stereotactic body radiotherapy (SBRT) and metastasectomy for treatment of pulmonary metastases

Methods

The study included 51 patients treated between January 2010 and June 2016. Among these patients, 21 patients received SBRT for 29 metastases and 30 patients received metastasectomy for 30 metastases. SBRT was performed using Cyberknife system and total radiation doses ranged from48 to 60 Gy, as delivered in three to five equal fractions. In metastasectomy group, wedge resection was performed in most patients (93.3%). Local control rate (LCR), progression free survival (PFS), and overall survival (OS) were assessed.

Results

The median follow-up duration was 13.7 months and the median patients’ age was 65 years. The median tumor size was larger in SBRT group compared with metastasectomy group (2.5cm vs 1.25cm; p=0.015) and other synchronous metastases was prevalent in SBRT group (57.1% vs 20%; p=0.006). The LCR was similar between SBRT and metastasectomy groups (61.58% vs 69.86%; p=0.163) and 1-yr LCR was 83.5% for SBRT group and 91.5% for metastasectomy group. PFS was longer in patients receiving metastasectomy compared with those receiving SRS (10.9% vs 26.59%; p=0.02). 1-yr PFS rate was 47.6% in SBRT and 60 % in metastasectomy group. OS did not show the difference between SBRT and metastasectomy group (52.75% vs 56.23%; p=0.534) and 1-yr OS rate was 79.5 % in SBRT and 95 % in metastasectomy.

Conclusion

The present study shows the SBRT and metastasectomy for pulmonary metastases had similar LCR and OS result. The patients with other synchronous metastases were more included in the SBRT group, and that may result in inferior PFS in SBRT group.

Background: Most recently, significant advances have been made in the field of in the immune check-point blockade. Nivolumab has been approved for use in patients with melanoma and non-small cell lung cancer (NSCLC). Little is known about the safety and outcomes in patients receiving immune checkpoint blockade agents and stereotactic radiosurgery (SRS) for the treatment of brain metastases (BMs).

Patients and methods: We retrospectively reviewed medical charts of 18 patients with advanced NSCLC and 23 patients with melanoma, treated with nivolumab. There were 19 men and 22 women. The mean patient age was 66.

Results: Among the 41 patients, 14 patients had BMs. SRS was administered before, during and after nivolumab in six, four, and two patinets.  One patient experienced seizure during the treatment with nivolumab, caused by perifocal edema 8 months after SRS. No other treatment-related neurologic toxicities were reported.

Conclusions: The relatively small number of patients and the heterogeneity of the population are limitations of this study. Nivolumab CNS activity warrants further evaluation.

Objectives

Preoperative radiosurgery facilitates target definition enabling a reduced planning margin. High local control rates with low toxicity are reported. Our aims were (1) to review the outcome of patients treated with postoperative hypofractionated stereotactic radiosurgery, (2) to compare matched preoperative GTV and PTV with postoperative CTV and PTV, (3) to explore dosimetry of preoperative radiosurgery, (4) to evaluate the patients in the context of the two current studies of preoperative radiosurgery.

Methods

We reviewed MRI follow-up imaging of 20 consecutive patients treated with postoperative radiosurgery (5 x 6 Gy to the 80% isodose). Brain metastases were contoured retrospectively on the preoperative MRI to generate a theoretical preoperative GTV. A planning margin of 1 mm was added to the preoperative GTV and 2mm to the postoperative CTV. 12 Gy and 15 Gy plans were created (iPlan, Brainlab) for brain metastases > 3cm diameter. Patients were evaluated against the eligibility criteria for current Phase I and Phase II trials of preoperative radiosurgery (RAD001, NCT02514915).

Results

5 patients declined MRI follow-up. MRIs showed 80% local control of the tumour bed [12 of 15 evaluable patients, median FU 15 mths, (range 3-55 mths)]. 3 patients (20%) had a local recurrence at the time of distant intracranial brain failure (median 8 mths), of whom one patient had leptomeningeal failure (6.6%). These outcomes are comparable to reports of preoperative radiosurgery (local control 85.6%, leptomeningeal relapse 4.5%). The postoperative CTV was larger than the preoperative GTV in 13/20 of our cases. Correspondingly, the postoperative PTV was larger than the preoperative PTV in 14/20 patients. The mean increase in PTV was 160% (18-530%), with increases >200% in the case of metastases <2 cm³. Where a decrease in volume was seen postoperatively, the mean preoperative GTV (often cystic) was 12.9 cm³ (11.3-76.2 cm³) and the average volume decrease was 37.6% (18-60.5%). 5 patients had synchronous brain metastases that were treated with radiosurgery. 13/20 patients would be eligible for current studies.

Conclusions

Preoperative radiosurgery is compelling due to more precise target definition, the potential for less normal brain toxicity due to a smaller irradiated volume and a dose reduction, a low incidence of leptomeningeal disease, enhanced patient convenience, reduced disruption of systemic therapy and radiosurgery of synchronous brain metastases. We propose a multidisciplinary registry study of preoperative radiosurgery for brain metastases indicated for resection with < 5mm midline shift.

Introduction: Early change in tumour perfusion and diffusion following stereotactic radiosurgery (SRS) is a potential biomarker of response.  However, efforts for quantitative model-based measures of Dynamic Contrast Enhanced (DCE) and Diffusion Weighted Imaging (DWI) parameters have shown variable findings to-date that may reflect variability in the MR acquisition and/or analysis method. This work describes the use of a voxel-based, multi-modality GPU architecture to include various complimentary solute transport processes such as perfusion and diffusion into a common framework. This is anticipated to improve accuracy and robustness of the early imaging biomarker predictions.

Methods: Patients treated with SRS as part of REB-approved clinical trials underwent volumetric DCE-CT, DCE-MRI and DWI-MRI scans at baseline, then 7 and 20 days post-SRS.  As DCE-CT is considered a good standard for tracer-kinetic validation given its signal linearity, we compared 3D pharmaco-kinetic parameter maps using a modified Tofts model (k_trans, v_e, V_b, AUC) from both modalities as well as the correlation between apparent diffusion coefficient (ADC) values from DWI-MRI and the extravascular, extracellular volume (V_e) from DCE imaging. A total of 14 tumours in 9 patients were evaluated. All imaging was co-registered to T1-Gad tumour contours and voxelwise correlations evaluated inside the GTV by Pearson correlation and Bland-Altman comparison.

Results: Voxel-wise analysis showed statistically significant correlations in K_trans (P<0.001) between DCE-CT and DCE-MRI over all imaging time points as well as excellent agreement with very little bias. Statistically significant correlations were also present between ADC/V_e and peaked at Day 7 (R=0.85, p<0.001) but a large variation was present across tumors (R²: 0.15-0.8) and disappeared altogether when reviewing the mean ADC only hence disregarding tumor heterogeneity.

Conclusion: Utility of a common analysis platform has shown statistically higher correlations between pharmaco-kinetic parameters than has previously been reported but is highlighting the need for a better understanding of the tumor microenvironment to improve biomarker sensitivity.

Background:  We have previously reported that patients with non-small cell lung cancer (NSCLC) harboring epidermal growth factor (EGFR) mutations have improved overall survival compared to patients with EGFR wild-type NSCLC.  With the advent of second- and third-generation tyrosine kinase inhibitors patients are living longer, have better systemic control, and have a greater opportunity to develop subsequent brain metastases and toxicity.  We present the clinical outcomes and toxicity of patients whose data were collected in a prospective registry. 

Methods:  Over a two year interval, 112  NSCLC patients had stereotactic gamma knife radiosurgery (GKRS) as their initial brain metastasis management.  Fifty-six of these patients had EGFR mutations. Overall survival, intracranial progression-free survival, and time to intracranial failure were determined.  Univariate and multivariate analysis were performed to determine factors affecting OS.  Toxicity of care was evaluated. 

Results:  Median follow-up and overall survival were 15 months (range: 1-98 months) from initial GKRS.  Twenty-one percent of patients lived longer than 2 years with 11% living longer than 3 years.  Twenty-three patients had mutations in exon 19, 8 in exon 20, 21 in exon 21, and 4 patients were unknown.  Twelve patients were symptomatic from intracranial disease at presentation.  Twenty-one patients were on targeted therapy at the time of GKRS and nine were on concurrent chemotherapy.  The median number of tumors treated at initial GKRS was 3.5 (range: 1-20) and median tumor volume was 0.88 cc (range: 0.03-7.05 cc).  The intracranial failure free survival was 7 months (range: 1-50).  Thirty-one patients developed distant brain failure, 2 failed locally, and 1 additional patient had local tumor progression and synchronous distant brain failure.  Twenty eight patients underwent salvage GKRS with a median number of GKRS treatments of 1 (range: 1-6).  Two patients required salvage surgery and 14 had salvage whole brain radiotherapy.  Twenty patients acquired biopsy confirmed T790M resistance mutations.  Total tumor volume > 0.75 cc didn’t correlate with overall or failure free survival. Median survival was non-significantly higher in patients who underwent subsequent immunotherapy (24 vs. 16 months). No patients had toxicity from initial GKRS regardless of concurrent chemotherapy or targeted therapy.  One patient developed seizures, required corticosteroids and anti-epileptic medication after a second GKRS.

Conclusions: With advances in targeted tyrosine kinase inhibitors and molecular testing patients are experiencing longer overall survival.  The management of such patients with brain metastases should be based on the potential for longer term survival with an emphasis on maintenance of function. 

Purpose:            Level 1 evidence now supports stereotactic radiosurgery (SRS) as the sole treatment in patients with up to four metastases. Prospective data on the use of SRS as an alternative to whole brain radiation therapy (WBRT) in patients with more than four metastases remains forthcoming. However, in order to avoid the deleterious effects of WBRT, it has been used in select patients with many more metastases. Gamma Knife (GK) SRS has been used to treat up to 43 targets in a single patient[i], and linear accelerator (linac) SRS has been used to treat up to 16.[ii] We present here the treatment of 27 individual targets in a single fraction with a single isocenter, to the authors’ knowledge, the highest reported number for single-fraction linac based treatment to date.

Methods:           A 63 year old Caucasian female with Stage IV (ER/PR+, HER2-) breast cancer, status post treatment with six rounds of traditional chemotherapy, CDK4/6 inhibition, and tamoxifen, presented to the emergency department with two-week history of new onset dizziness and falls, with associated headaches, weakness, and fatigue. MRI revealed numerous tiny new foci of enhancement since her previous scan 18 months before.  After detailed discussion of the risks and benefits of SRS and WBRT, the patient chose focal therapy. Each lesion was prescribed 18Gy, and a 4-non-coplanar RapidArc plan was generated in Eclipse for our Edge linear accelerator (Varian Medical Systems) according to our previously described template. The plan was normalized such that at least 99% of total target volume (TV) received the prescription dose. Patient specific quality assurance was performed on the plan using radiochromic film and our standard phantom, confirming excellent agreement between calculated and delivered dose.

Results:                             Total TV was 2.2 cm³. V_12Gy to brain was 13.9 cm³_. Mean dose to brain was 4.81Gy. Paddick conformity for the entire plan was 0.73. Maximum doses to brainstem, chiasm, and optics were 7.82, 7.52, and 4.20 Gy, respectively. From patient positioning on table to thermoplastic mask removal, beam on time was approximately 6 minutes, and total treatment time was approximately 20 minutes. The procedure was well-tolerated and without incident.

Conclusion:        Improvements in linear accelerator technology and treatment planning capability have led to high-quality and highly efficient treatment of many more metastases than previously deliverable. 

Objectives

The aim of the study was to assess the efficiency and safety of the CyberKnife stereotactic radiosurgery and radiotherapy, as well as to evaluate prognostic factors in patients with lung cancer brain metastases.

Methods

The analysis included 53 patients of median age of 61 (range 43 – 84), treated due to lung cancer brain metastases between 2011 and 2015. The time between the diagnosis of the primary tumor and brain metastases ranged from 0 to 89 months (mean 20 months, median 14 months). In 9 patients brain metastases were found prior to diagnosis of the primary tumor. All patients had high KPS score (mean and median – 80; 22.6% - KPS 100, 64.2% - KPS 80, 13.2% - KPS 70). The number of irradiated lesions in one patient ranged from 1 to 6. Mean total dose was 18.3 Gy, median – 18.0 Gy (range 6 – 26 Gy). 51% of the lesions were treated with single fraction, 49% using fractionated regimens. The planned dose was delivered in 1 – 4 fractions. Medical examination along with radiological evaluation of the treatment effects was performed in 23 patients (43%). The survival probability was calculated with the use of the Kaplan – Meier estimator.

Results

The median overall survival (OS) in the whole group was 15.1 months. Patients with a single lesion lived longer (median 24.0 months) than those with multiple brain metastases – median 7.2 months (p = 0.0079). Longer OS was observed in patients with controlled systemic disease (median 19.7 months) as compared to progressive systemic disease – median 5.5 months (p = 0.058). The analysis showed no association between KPS score or the histological type of the primary tumor and OS. Single fraction and fractionated treatment resulted in similar OS. Adding whole brain radiotherapy did not influence the OS. During radiological follow-up, progression of the irradiated lesions was found in 10 patients, 7 to 27 months after the treatment. Median local progression-free survival was 19.8 months.

Conclusion

Stereotactic radiosurgery and hypofractionated stereotactic radiotherapy have similar effectiveness in the treatment of lung cancer brain metastases. Patients with single metastasis and controlled systemic disease have the most favourable prognosis.

Purpose: Stereotactic body radiation therapy (SBRT) is increasingly applied to oligometastatic disease, like bone and lymph node metastases., We developed a comprehensive set of treatment guidelines for bone and lymph node locations eligible for SBRT, based on our clinical experience with common SBRT sites (such as lung, spine and liver). To our knowledge, this is the first study on IGRT positioning data of bone- and lymph nodes SBRT treatments in the pelvic region.

Materials and Methods: Positioning data for 32 patients (n=16 bone, n=16 lymphnode) treated with SBRT were reviewed. Radiotherapy schedules ranged from 24 -45 Gy in 3 fractions to 25-50 Gy in 5 fractions. Patients were immobilized with a personal vacuum bag, knee-fix, head rest and arm support. The Gross Tumor Volume (GTV) was expanded with a 5mm and 7mm Planning Target Volume (PTV) margin for bone and lymph node treatments, respectively. Treatments were performed on an Elekta linear accelerator, using a 10MV, coplanar, dual arc, volumetric Modulated Arc Therapy (VMAT) technique. A Cone Beam CT (CBCT) based online imaging protocol was used for set-up, couch correction verification and intra-fraction motion (IFM) assessment. Rigid registrations were performed on the bony anatomy adjacent to the GTV. If the residual translation setup error (i.e. after couch correction) was larger than 2 mm, the correction-verification procedure was repeated and if residual rotation setup errors were larger than 3° the patient was repositioned. The coverage of GTV within PTV was checked visually. To calculate the IFM, the difference between translation and rotation errors of the inline (i.e. during treatment) or post treatment CBCT and the residual setup errors was calculated.

Results: Residual setup and IFM errors (translations and rotations) are summarized in the table 1 for bone and lymph node cases. Mean (M) errors are close to zero. Systematic (S) and random (s) components of these errors remain well below 0.5mm, 1.4° and 1.6mm, 0.7°, respectively. The correction-verification procedure was repeated in 3.8% and 10% of the fractions for bone and lymph node cases.

Conclusion: The residual setup and IFM errors of patients treated with SBRT for oligometastatic disease in the pelvic region (for bone or lymph nodes locations) are very small, demonstrating the reproducibility and robustness of the positioning protocol. Consequently, the contribution of these errors to the GTV-PTV margin is limited and margins may be reduced. Ongoing investigations include clinical data, robust planning techniques and image registration methods.

Objectives : To compare dose distribution qualities for TMR 10 and TMR classic Gamma plans.

Methods: Leksell Gamma knife radiosurgery equipment was installed at PSMMC in Sep. 2013. The planning system  available was TMR classic with which patients were planned and treated till Jan, 2015 for a period of 15 months. Since TMR 10 became available and the system got upgraded for use till to date. A dosimetric comparison was carried out for dose distributions performing single and multiple shots plans, controlling all other variables/factors using both planning algorithms. Diameters of multiple Isodose lines/levels and point doses at fixed coordinates were calculated to find any difference in measurements. Output, decay, time, shot sizes, shot numbers and shot volumes/weightages were kept consistent.

Results : Comparison was made through calculating doses on two algorithms TMR classic and TMR 10 and revealed a difference of mean 0.5 mm constricted isodose lines for TMR classic  compared to TMR 10. This difference was consistent with all forms of factors evaluated including single versus multiple shots, rounded versus oblong shaped targets and for all three sizes of collimators. Effects of different grid sizes did not differ in two algorithmic calculations. Point doses calculated at different dose levels (within target, prescription point and at dose gradient), a mean difference of 0.7 Gy was found in favour of TMR 10 . All calculations were made on MRI images without inhomogeneity corrections.

Conclusion : In an era of precision radiotherapy fractional dose distribution differences should be marked and understood before embarking on to treat sensitive organs at risk.

Objective

To investigate the residual errors of the ExacTrac X-Ray 6 Degree-of-freedom (6D) IGRT system for spine SBRT

Methods and materials

Forty four spine SBRT treatments from January, 2016 to March, 2016 were analyzed retrospectively. The treated sites included 6 cervical, 24 thoracic, 12 lumbar and 2 sacrum spines. Lesions encompass 1 to 2 vertebral sections. The patients were aligned with ExacTrac X-Ray 6D system and then followed by a CBCT scan. No shifts applied in between CBCT and ExacTrac images. CBCT images of 44 treatments were imported into Pinnacle treatment planning system and they were co-registered to planning CT. The translational and rotational shifts obtained by image registration, from CBCT to the planning CT, were considered residual errors of ExacTrac X-Ray 6D system.

Results

The residual translational errors in RMS for the 44 treatments were 0.63, 0.63 and 0.82 mm at left-right, anterior-posterior and superior-inferior directions, respectively. The overall residual translational error was 0.70 mm. The residual rotational errors were 0.20, 0.10 and 0.19 degree in pitch, yaw and roll, respectively. The overall residual rotational error was 0.17 degree.

The standard deviations (SD) of residual translational errors were 0.45, 0.50 and 0.51 mm at left-right, anterior-posterior and superior-inferior directions, respectively. The SD of the overall residual translational error was 0.50 mm. The SD of residual rotational errors were 0.20, 0.10 and 0.19 degree in pitch, yaw and roll, respectively. The SD of the overall residual rotational errors was 0.17 degree.

Conclusion

This study demonstrates the residual translational and rotational errors of Exactrac X-ray 6D IGRT system is very small.

PURPOSE

In this study, we make a change Dynamic Conformal Arc Therapy (DCAT) for Volumetric Modulated Arc Therapy (VMAT) system. We estimated Conformity Index(CI) and Homogeneity Index(HI) with accuracy, usefulness using APEX system, Multi shape phantom and QA phantom for the radiosurgery.

MATERIAL AND METHOD

We used the multi shape phantom and APEX system. we make a changed DCAT for VMAT System. We simulated multi shape(Star,Diamond,Half moon) phantom and Mim software(ver. 5.0) Contouring VOI(Virtual Organ at Interest). MONACO (ver. 5.0) was utilized as a radiation treatment planning system (RTPs). APEX system has the micro multi leaf collimator (mMLC) with leaf thickness of 2.5 mm. INFINITY Tx Machine was used. We performed two different techniques for the comparison.

RESULTS

We confirmed difference between the DCAT and VMAT plan. the 3D assessment showed a significant difference between the DCAT and VMAT (Multi shape 98% with 3% / 3 mm criteria, Multi shape pass rate of above 95% with 2% / 2 mm criteria, both CI 0.3 and HI 1.0 results is accept ).

CONCLUSIONS

DCAT plan can not modulate the intensity of radiation while the gantry is rotating. And mMLC shape and gantry rotating speed is fixed at the each control point. However,we  make a VOI and used another MONACO plan skill is changed the VMAT effect results.we can modulate the intensity of radiation through the optimized mMLC shape and gantry rotating speed. Therefore, our results show that the VMAT plan is superior to the DCAT plan. we conclude that APEX system should adapt the VMAT system for the accurate radiosurgery

For functional radiosurgery using Gamma Knife, the maximum dose of 80Gy~130Gy is delivered to the target. Therefore, besides the area of 50% isodose, the penumbra area of 25% isodose is important because the 20Gy~32.5Gy is delivered to the area. The dose can cause the direct effect of necrosis and the indirect effect that the surrounding tissue will die from the vessel is occluded from radiation. The Gamma Knife model C has the hemispherical 5 ring distribution of 201 cobalt source, the first ring has 35 sources, the second ring has 39 sources, the third ring has 39 sources, the fourth ring has 44 sources and the fifth ring has 44 sources. All sources are focused at SAD (source axis distance) of 40cm. The distribution of the sources makes a nonlinear ellipse of radiation in X, Y and Z direction. The FWHM(full width at half maximum) and 25% of isodose was 4.8mm and 5.8mm in Z direction but in X and Y direction the value was 5.9mm and 8.8mm which meant that the penumbra of X and Y direction was bigger than in Z. To reduce this effect, we made a special distribution of sources by plugging the 5^th ring and the 5^th and 4^th ring. As the treatment time in using all 5 rings is unit of 1, the time of plugging of 5^th ring and of plugging of 5^th ring and 4^th ring was 1.28 and 1.79 respectively. The size of 50% in X and Y slightly decreased to 5.8mm and 5.7mm with plugging but in Z the size slightly increased to 4.9mm and 5.0mm. For the penumbra area of 25%, the size in X and Y much decreased to 8.3mm and 7.8mm but in Z, the size increased to 6mm and 6.2mm. This results showed the reduction of nonlinear ellipse in dose distribution. The comparison of 25% volume and 50% volume was calculated with GI (gradient index). The GI was first a 2.73 in using all 5 rings and changed to 2.69 with plugging 5^th ring and changed to 2.7 with plugging 5^th ring and 4^th ring. This study showed the much reduction of penumbra and the increase of 50% size in Z direction with ring plugging. This technique will be applied to the functional Gamma Knife radiosurgery such as trigeminal neuralgia radiosurgery, Gamma Knife thalamotomy and pallidotomy.

AIMS:
LOT (lung optimized treatment) is an evolution in Cyberknife® technology which allows treatment of lung cancer without invasive fiducial implantation procedures. The aim of this analysis was to evaluate the technical feasibility, toxicity profile and clinical outcome.
METHODS:
Between 1/2014 and 10/2016 115 patients (pts) (M/F 79/36) were treated with Cyberknife® using LOT at European Institute of Oncology (IEO). The median patient age was 72.6 years (range 31.8-90.3). Treated lesions were 120; 50 with histopathological confirmation (44 primitive pulmonary cancer, 8 pulmonary mets), 66 were untyped tumors. 101 pts treated a single lesion, while 5 pts treated multiple target lesions. For 13 pts treatment using LOT was a re-irradiation for a recurrence in field. Concomitant systemic therapy was administered in 3 pts. Three tracking methods were used: 0-View tracking method (treats an ITV using Xsight Spine tracking for patient alignment) in 59 pts, 1-View tracking method (tracks targets that are visible in only one X-ray image) in 33 pts, 2-View tracking method (tracks targets that are visible in two X-ray images) in 28 pts. Median dose/fraction was 15 Gy (range 4-18). In most cases the isodose prescription was at 80%. The median PTV was 24.3 cm3 (range 2.7-161.1). Toxicity was evaluated by RTOG/EORTC and CTCAE V4.1. Tumor response was evaluated with RECIST V1.1 criteria.
RESULTS:
Toxicities and follow up data are available for  52 patients.The median follow-up was 5.1 months (range 1- 15.4). Acute toxicity (within 6 m.) was observed in 21 of 44 pts with follow-up (47.7%): according to RTOG/EORTC criteria only G1 and G2 toxicity was registered (no G3 orG4); in CTCAE V4.1 two events of G3 toxicity were observed (cough, dyspnea) . Late toxicity (after 6 m.) was observed in 10 of 19 pts with follow-up (52.6%): all events were G1 and G2 RTOG/EORTC events. (no G3 or G4); in CTCAE V4.1 one event of G3 toxicity was registered. According to RECIST V1.1 guideline complete response, partial response, stable disease and progressive disease was observed in 23.9%, 26%, 43.5% and 4.3% respectively.
CONCLUSIONS:
This first analysis demonstrated high feasibility and minimal toxicity of LOT in lung cancers. Promising response rates have been registered. Further studies are necessary in order to confirm our results.

Introduction: The aim of this study is to develop a new evaluation method that quantifies a degree of risk at each organ at stereotactic body radiation therapy (SBRT) plan. A risk index method was designed to evaluate the possibility of under-dose to the treatment target or over-dose to the organs at risk (OAR) at the plan.

Materials and Methods: A risk index value was designed to quantify the possibility of the under-dose or over-dose at the target and the organs from the small set-up errors. The risk index values at all voxels in the treatment target and organs at risk were calculated and the voxels were categorized to three status of “Pass”, “Fail” or “Pass with Risk”. Passing rates were calculated with the risk index values at 3D volumes of the treatment targets and the organs at risk to show the degree of dosimetric safety from the under-dose or over-dose. Various 3D dose distributions were acquired from a spine SBRT plan and a lung SBRT plan by simulating setup errors of 1 mm, 2mm and 3 mm, and the degree of the risk from setup errors were evaluated to show the feasibility of the suggested method.

Results:  The risk index values at 3D volumes could show increased risk at the target or organs at risk quantitatively. For the lung SBRT plan, the passing rate of the treatment target was decreased from 93.47% to 85.14% as the setup errors were increased. The passing rate of some organs at risk showed slightly increased or decreased tendencies according to the position of the organs. The plan of spine SBRT showed similar results. The passing rate of the treatment target, spinal cord and esophagus were decreased, but left lung, right lung, and heart showed almost same passing rate as the setup errors were increased.

Conclusion: The developed method could give 3D information about where the possibility of under-doses in the treatment target and the possibility of over-doses to the organs at risk exist. The risk index method is expected to give great potential to planning strategy which can give safer and more deliverable treatment plan under setup errors.

Objectives

The aim of this work was to characterize the imaging system of a CyberKnife VSI® which consists of two X-ray tubes mounted to the ceiling, coupled with two in-floor Flat Panel Detectors (FPDs). The central axes of the X-ray beams intersect the centre of the FPDs with a 45-degree angle.

The quality controls of X-ray tubes were the ones recommended by Report AAPM TG 135, while the performance of the detectors was evaluated applying the Report No.6  of the Italian Association of Medical Physics (AIFM). 

Methods

The X-ray beam of each tube was characterized using a multimeter with clinical  parameter setting.

All the images were acquired with the standard radiation quality RQA5 at different exposure levels. It was estimated the response function, the local and global non-uniformity of the signal, the local and global non-uniformity of the signal-to-noise-ratio, the presence of uncorrected bad pixels.

Images were converted into dose in order to analyze the noise components (electronic, quantum and structural) and to obtain the Noise Power Spectrum (NPS). The spatial resolution was evaluated by the Modulation Transfer Function (MTF) using both an edge test and a grid pattern phantom. Detective Quantum Efficiency (DQE) was calculated from MTF and normalized NPS. Furthermore, the geometric distortion was investigated and the Contrast to Noise Ratio (CNR) and contrast-detail curves were obtained through specific phantoms.

Results 

The X-ray tubes measurements showed values well within the limits indicated in the Report AAPM TG 135.

The analysis of the images exhibited a linear response function (R²> 0.99) with a negative slope and all the uniformity indexes were far below the reference limits set in the Report AIFM No.6 .

Each component in noise analysis had the typical trend expected for FPDs: the quantum noise was the predominant one and it decreased increasing the dose. This result was also confirmed by the NPS curves. MTF values obtained from both methods confirmed the specifications provided by the manufacturer, but the grid pattern was chosen as the best candidate because of practical reasons. The CNR values increased with the incident exposure and the contrast detail curves trend was the expected one. No geometric distortion was detected.

Conclusion

The performance of a Cyberknife imaging system was analyzed and characterized in detail; in this way some reference values were here suggested. This study is a good starting point for the implementation of a quality control protocol.

Introduction: SRS is complexed procedure requires accurate target definition, correct radiation beam simulation, exact target localization and precise dose delivery. In LINAC based SRS, on board imaging system such as CBCT and KV-pair was used for target localization. And the dose delivery accuracy was determined by accuracy of the localization and the isocenter discrepancy between radiation isocenter and geometric isocenter. In this study, a Lucy 3D QA Phantom was applied to estimate the geometrical variation of a practice with a LINAC-based SRS system.

Methods and Materials: Quality Assurance (QA) procedures of five SRS cone cases were selected for this study. The cone range in diameter were from 10mm, 12mm, 16mm, 18mm and 24mm. In each QA procedure, two GAFCHROMIC films were attached to both sides of the dosimetry inset for the chamber in the Lucy Phantom, and the thickness of the inset was 1cm.  After alignment, a single 360 degree ARC plan was delivered to the phantom, so both of the film were irradiated with a chamber reading.  Then the films were scanned with a Vidar DosimetryPro scanner, which was controlled by the RIT Software by Radiological Imaging Technology.  Due to the isocenter variation for setup, the two film images showed difference in optical density, which reflected the dose difference due to different distances from the radiation isocenter. The images were analyzed by RIT Classic version 6.4.  The ratio of distances from the two films to the radiation isocenter were computed with linear relationship assumption based on the maximum raw pixel values from each film image.  And the distances from the film to the radiation isocenter were also calculated.

Result: In the selected cases, the average ratio of maximum raw pixel values from these two film was 1.11 with standard deviation at 0.14, and the computed average distance differences from the two films to the radiation isocenter was 0.48mm with standard deviation at 0.60mm.  The chamber reading were satisfied the dosimetry requirement after this geometrical correction were applied.

Conclusion: A 3-dimension geometrical accuracy evaluation was reached by adding two 2-dimension-planes within a Lucy Phantom during a dosimetry accuracy estimation of quality assurance procedure. This simple philosophy could be applied to other quality control and quality assurance for image registration between different modalities and spatial distance between image slices during the SRS procedure. 

Objectives

The aim of this work was the definition of a quality control (QC) protocol for the imaging system of a Cyberknife VSI^® (G4 configuration) to detect changes from its original level of performance. A deviation from the proper baseline may result in a clinically significant degradation in image quality, that contributes to a loss of target accuracy and/or a significant increase in radiation exposure.

AAPM Report TG 135 “Quality assurance for robotic radiosurgery” suggests only basic indications about quality controls for imaging system. Moreover, there are only few papers about this topic available in the existing literature.

Methods

The quality checks suggested in AAPM TG 135 were performed and the whole imaging system (X-ray tubes and Flat Panel Detectors FPDs) was characterized in order to set some reference values. Further measurements on FPDs were executed applying Report No.6 of Italian Association of Medical Physics as guideline.  

For X-ray tube measurements a multimeter was used. Different radiological specific phantoms,adopted for image quality assessment, were employed and both original X-ray images and virtual projections were analyzed. Finding a suitable and reproducible setup for measurements was not easy: a home-made device was realized in order to position phantoms and multimeter at a fixed distance, perpendicular to  the X-ray beam.

Thanks to that, it was possible to select the QCs to be performed, their frequency, tolerances and limit values. A rigorous protocol was drafted, implemented and tested in practice.

Result

Here the image QCs for FPDs included in the protocol are reported. The frequency of all the QCs is annual except for the last which is performed monthly

Aspect ratio  <± 3%

Conversion function R² > 0.99

Noise analysis Reference value

Local & Global Non Uniformity of Signal NULS <4%  NUGS <8%

Local & Global Non Uniformity of Signal-to-Noise-Ratio NULSNR <8%  NUGSNR <20%

Artifact analysis No artifacts

Spatial resolution          Deviation from reference value ± 10%

Low contrast sensitivity Deviation from reference value ± 20%

Contrast detail curves           Reference values

Lag effect<0.5%

Low & high contrast sensitivity and Spatial resolution     Reference values: number of visible contrast details and lp/mm

Conclusion

In this work a QC protocol for the  imaging system of Cyberknife VSI^® was implemented. It will be interesting to verify if the selected checks and frequencies have a sufficient degree of sensitivity to pre-emptively detect small deviations from the baseline . Furthermore it will be mandatory to identify levels of action to prevent serious patient damages.

Purpose: In this study, the dose spillage level was explored by adjusting surrounding tissue limitation function to reach the optimal conformity of the target in liver Stereotactic Body Radiotherapy (SBRT) with RapidARC treatment technique.

Materials and methods: A liver SBRT case was selected for this study. The PTV was 72.48cc, and liver volume was 1113cc.  A subsidiary structure (shell) was utilized to avoid dose spillage during the optimization procedure. Three dose shells were added around the PTV from central to peripheral. The optimization setting radius of using one shell was 3mm, 5mm, 8mm, 10mm, and 15mm,  two shells was 3mm and 3mm; 5mm and 5mm; 8mm and 8mm; 10mm and 10mm; and 15mm and 20mm.  When 3 shells were used, the radius of the shells were combination of the 3mm, 3mm and 3mm; 5mm, 5mm and 5mm; 8mm, 8mm, and 8mm; 10mm, 12mm, and 15mm; and 15mm, 20mm, and 20mm. The plan optimization strategy was to add different priorities to the shells, and then to evaluate the dose spillage level, liver function index, and related dosimetry performance. Optimization was that 98% of the prescription dose would fall within the first (inner) shell, 50% of the prescription dose would fall within the second shell, and 40% of the dose would fall within the third shell when 3 shells were used. Dosimetry performance was described by the conformity index, which was defined as PIV/TV, where TV is target volume, and PIV is prescription isodose volume. The spillage index was defined by concerned dose level, which is defined to be 12Gy volume divided by target volume in this study. And the liver function index was defined to be liver volume subtracted volume of 15Gy and then divided by liver volume.

Result: When one shell was applied, the optimized plan  resulted in conformity index = 0.97, minimum spillage index = 14.9, and corresponding functional index = 2.65; when two shells were applied, [setting at 8mm and 8mm] yielded minimum spillage index = 13.53 with conformity index= 1.01 and function index = 2.41; When three shells structure were applied, [setting at 8mm, 8mm, and 8mm]  resulted in minimum spillage index = 13.62, conformity index = 1.02 and function index = 2.41.

Conclusion: Shell dose volume limit setting at 8mm was the best planning strategy for this SBRT RapidARC plan. A similar strategy could apply to different treatment sites. 

Purpose: The developments of highly conformal and precise radiation therapy techniques promote the necessity of more accurate treatment target localization and tracking. It is unclear which system provides the best accuracy for tracking. The aim of this study is to evaluate the role of lung optimization treatment (LOT) simulation for Cyberknife stereotactic lung radiotherapy.

Material and Methods: From September 2014 to December 2016, 315 consecutive patients with ling metastases and lung primary cancer,referred to our department for Cyberknife stereotactic radiotherapy, were set out for lung optimization treatment simulation. For all patients  a CT scan was performed in expiratory and inspiratory phase and registered to the planning CT. During the simulation Two x-ray tubes arranged orthogonally to each other locate the lung lesion and align it to the digitally reconstructed radiographs derived from CT scan obtained earlier in the same day. The Simulation process allow the clinician to determine the visibility of the target in three different tracking methods . If the target is visible in two projections,  the Xsight® Lung Tracking System is used for treatment, if it is visible in only one projection, the 1-View Tracking algorithm is used.  No visible target in either projection. X sight spine tracking  or better fiducial marker are requested.

Results: According to the accuracy of the LOT system we observed these results: for 165 patients (52%) the Xsight® Lung Tracking System was used for Cyberknife lung SBRT treatment, for 99 patients (31%) 1 view modality. in all others 51 patients (17%) fiducial markers have been necessary for tracking modality. Target margins were individualized according to tumor respiratory motion as follow: the GTVs were expanded by 3 mm in all directions to create the CTVs .We used different margins for PTVs. In the 2-view modality the CTV on expiratory CT was expanded by 2mm in all directions, while for 1 view modality two different CTVs were generated on both CT scan to include the entire inhale-to-exhale tumor motion, and added together to create an ITV expanded by 2 mm in the direction followed by the XRay camera and 3mm providing the most flexible treatment options

Conclusion: LOT simulation system and Xsight lung are considered the best choice in the management of lung lesions in our clinical practice with the advantage to dramatically reduce PTV margins and consequently the risk of potential toxicities related to the high level of doses delivered during SBRT practice.

Introduction

In vivo dosimetry (IVD) represents the last step of quality assurance process in radiotherapy clinic. This strategy has shown unique features to trace deviations between planned and actually delivered dose distributions. Extracranial stereotactic radiotherapy (SBRT) involves the delivery of high doses in a few fractions (1-5) for ablative purposes. SBRT treatments may strongly benefit from IVD as any uncertainties in dose delivery are more detrimental for treatment goals. We assessed the feasibility of EPID-based IVD for complex clinical VMAT treatments for SBRT.

Methods and materials

15 patients with lung, liver, bone and lymphnodal metastases treated with Elekta VMAT were enrolled. All plans were generated with a single 360° arc. All targets were irradiated in 5 consecutive fractions, with total doses ranging from 40 to 50 Gy depending on anatomical sites. All patients passed pre-treatment 3%/3mm γ-analysis verification. IVD was performed using SOFTDISO (Best Medical Italy), a dedicated software implemented in our clinic for conformal, IMRT and VMAT techniques. IVD tests were evaluated by means of (i) R ratio between isocenter daily in-vivo dose and planned dose and (ii) γ-analysis between EPID integral portal images in terms of percentage of points with γ-value smaller than one (γ%) and mean γ-values (γmean), using a global 3%-3 mm criteria. Alert criteria of ±5% for R ratio, γ% <90% and γmean > 0.67 were chosen. 

Results

A total of 75 transit EPID images were acquired. The overall mean R ratio was equal to 0.999 ± 0.021  (1 SD) for all patients, with more than 98% of tests within 5% alert criteria. The portal images γ-analysis shows an overall γmean of 0.29±0.13 with 100% of tests within alert criteria, and a mean γ% equal to 96.9±5.2% with 96.0% of tests within alert criteria. In contrast to our past experience of patients with head-neck and pelvic treatments, where the systematic use of IVD revealed some discrepancies due to major anatomical variations or random anatomical changes, no relevant discrepancies were detected in SBRT patients. The results are supplied in quasi real-time, with IVD tests performed and displayed after only 1 minute from the end of arc delivery. 

Conclusions

The present EPID-based IVD algorithm provided a fast and accurate procedure for SBRT-VMAT delivery verification in clinical routine. This strategy allows physics and medical staff to promptly act in case of major deviations of dose delivery.

The intraoperative radiotherapy (IORT) is a treatment technique to deliver radiation during surgery. The comfort of patient can be increased compared to fractionated external radiation therapy. The radiation generator of INTRABEAM system (Carl Zeiss Meditec AG, Germany) generates x-ray in an isotropic direction. We designed and developed a dedicated applicator to apply the INTRABEAM system to the thoracic and abdominal region where there are many critical organs. The applicators were customized for the irradiation direction as 0 and 45 degrees and depending on flattening or non-flattening by 3D printer. Dosimetric characteristics were evaluated, such as profiles, flatness, and leakage as well. The applicator was also designed to shield any other sides except the desired direction. Without applicator, the flatness tended to be 1.28 on the surface, while 1.47 on the 5 mm depth. However, the resultant flatness of applicator developed in this study was uniformly maintained from 1.22 to 1.30 up to the depth of 15 mm. It was also found that the leakage dose on the side except the beam direction was negligible. We applied the applicator to four patients with rectal, breast, and pancreas cancer treated with IORT from January to December 2015. The validity of the applicator had been assured to be effectively applied to the thoracic and abdominal region.

Background and purpose: Small field sizes are increasingly used in new radiotherapy(such as SRS/SBRT) to deliver higher dose gradient to patients.Estimating dosimetric parameters for such fields in non-reference conditions based on the conventional protocols used at large fields, as used in the reference condition, lead to significant errors.This study was to determine and compare small fields correction factors (KNR and KNCSF)measured with different types of active detectors.

Materials and Methods: Small field sizes were defined by circular cones down to 30 and 5mm diameters.Then, the KNR and KNCSF correction factors proposed recently for small field dosimetry formalism (TG155) were determined for different active detectors in a homogeneous as well as a non-homogeneous phantom. The non-homogeneous phantom was designed and made by using Perspex as the soft tissue and appropriate lung and bone tissue equivalent materials. Dosimetric measurements were made by using high resolution diodes, and ionizing chambers. The 6 and 18MV beams were produced by a 2100C/D Varian  linear accelerator system with the circular collimators fixed at its head.Variation of the central axis dose in the 5 and 30 mm small fields,in the inhomogeneous phantom constructed of different inhomogenous layers (composed of Cork and PTFE)for the 6 and 18MV energies was also investigated.

Results: The KNR correction factors for the circle field of 30mm estimated for the Pinpoint ionizing chambers, EDP-20 and EDP-10 diodes were 0.993, 1.020 and 1.054 at 6 MV; and 0.992, 1.054 and 1.005 at 18 MV respectively.The KNCSF correction factor for the circle field of 5mm estimated for the Pinpoint ionizing chambers, EDP-20 and EDP-10 diodes were 0.994, 1.023 and 1.040 at 6MV; and 1.000, 1.014 and 1.022 at 18MV respectively.The maximum variation in the percentage depth dose in the non-homogeneous phantom relative to the homogeneous phantom in the 5 and 30 mm field sizes due to the presence of 30mm Cork heterogeneity were 23.5 % and the 62.1%,respectively,while the PTFE heterogeneity caused a maximum variation of 8.17%, 7.15% for the 5 and 30 mm field size respectively.

Conclusion: Implementing the correction factors based on the new dosimetry protocol proposed for the small fields increases the dosimetric precision and accuracy of small field’s radiotherapy procedures of such small fields. In addition,the dosimetric measurements with the diodes and ionizing chamber indicated that the perturbations of doses at the central axis in the small fields increases due to the presence of heterogeneities within the non-homogeneous region and thereafter.

The aim of this presentation is to report a case of radionecrosis of aggressive evolution.
We present a female, 13-years-old patient, with incidental diagnosis of right parieto-occipital AVM (Spetzler 2), treated by endovascular embolization and later radiosurgery (LINAC, dose 16 Gy, volume 12 cm3) in the first year since the diagnosis. Two years after radiosurgery she presented epilepsy with partial visual crises; the EEG demonstrated right parieto-temporo-occipital seizures. Neuroimages studies demonstrated sequelar expansive lesion with a necrotic-cystic component and another nodular one, with mass effect, post-gadolinium enhancement and surrounding edema. It was interpreted as symptomatic radionecrosis; pharmacological antiepileptic treatment and high doses of steroids were indicated. One year later she persisted with seizures and growthing necrótic-cystic that required stereotactic evacuation and CSF derivation placement. Despite of short transitory improvement , she added left hemianopsy, brachial palsy and contralateral neglect. The neuroimages demonstrated progression of the nodular lesion with increase of the mass effect, for this reason the post-actinic injury was resected.
In spite of marked improvement in the neuroimages and resolution of the brachial weakness, she persisted with daily partial seizures (average 3 per day) evolving in lost of consciousness sometimes. Actually epilepsy surgery is considered.
We conclude that the radiosurgery is a valid tool for the treatment of cerebral AVM, controlling the primary disease. Nevertheless, the radionecrosis can appear as one of the most serious associated complications, generating persistent neurological alterations in spite of an intensive clinical-surgical managing, demonstrated in this case as well as in previous reports. In view of eventual neurosurgical complications, the indication, planning and execution of cerebral radiosurgery must be realized with participation of a neurosurgeon.

Purpose: With the development of a cone-beam CT (CBCT) image guided Icon system, ablative doses of SRS can be delivered using a frameless technique. However, as is typical with CBCT systems using filtered back-projection (FBP), the standard Icon images suffer from blurriness, CT# inaccuracy, and low soft-tissue contrast detectability.  In this study, an image reconstruction framework is proposed utilizing a novel high-resolution iterative reconstruction algorithm together with pre- and post-processing steps to overcome some of the image quality issues inherent in the Icon CBCT system.

Materials and Methods: The pre-processing step includes a shading and beam-hardening artifact correction applied on the X-ray projections. The projections are then reconstructed using a novel iterative algorithm that improves spatial resolution by explicitly accounting for image un-sharpness caused by different factors in the reconstruction formulation. The images are post-processed to correct the dome/capping artifact caused by the spatial variations in X-ray energy generated by the bowtie-filter. The proposed shading correction algorithm relies solely on the acquired projection images (i.e. no prior information required). It utilizes FBP reconstructed images to generate a segmented bone and soft-tissue map. Ideal projections are estimated from the segmented images and a smoothed version of the difference between the ideal and measured projections is used to reduce the image in-homogeneity and to improve the soft-tissue visibility. The proposed framework was tested on CatPhan phantom, as well as patient images acquired on the Icon system.

Results: The CatPhan results show that the linearity of the reconstructed attenuation-map was improved from 0.8 to 0.95, resulting in more accurate CT#. The CT# mean absolute measurement error was reduced from 76.1 to 26.9HU. The contrast to noise ratio of the acrylic insert in the sensitometry module was improved from 1.75 to 7.76. The line-pair resolution was improved from 12 line-pair/cm in FBP to 14 line-pair/cm. Moreover, the resulting clinical brain images show substantial improvements in soft tissue contrast visibility, revealing structures such as ventricles and lesions which were otherwise un-detectable in FBP-reconstructed images.

Conclusion: The proposed reconstruction framework improved the CT# measurement accuracy and the soft-tissue contrast/visibility. Improved visibility of the brain structures can improve the evaluation of MR to CBCT co-registration. Moreover, the improved CT# values can make the application of the CBCT for daily dose delivery measurements more reliable.

Objectives:

We investigated the Monte Carlo (MC) method of dose calculation for the planned target volume (PTV) in intensity-modulated radiotherapy (IMRT) for brain tumors in comparison with Pencil Beam (PB) method.

Methods:

Twelve cases of skull base tumors (volume, 8.3 to 51.0 ml), including acoustic tumors, meningiomas, and metastatic tumors (4 cases each), were selected for dosimetric comparison. They were treated by Novalis IMRT during a period from January, 2013 through April, 2016. A total dose of 37 to 50 Gy was delivered in 10 to 20 fractions. Radiotherapy planning was made using iPlan version 4.1.4 (BrainLAB, Tokyo) with coplanar multi-beam IMRT.  We investigated the dose for the PTV (mean dose, D98%, D95% D5% and D2% of PTV) calculated by MC, in comparison with that by original PB method.

Results:

The differences in the mean values of the mean dose, D98%, D95%, D5% and D2% of PTV between PB and MC were minimal (mean dose, 2.99±0.627 Gy and 2.96±0.506 Gy; D98%, 2.93±0.619 Gy and 2.93±0.620 Gy; D95%, 2.84±0.602 Gy and 2.83±0.601 Gy; D5%, 0.148±0.035 Gy and 0.146±0.033 Gy; D2%, 0.058±0.010 Gy and 0.059±0.010 Gy). In the two cases, D5% of PTV was smaller by about 7.5% and 10% in MC compared with that in PB, while the difference was minimal in the remaining 10 cases. On the one hand, D2% of PTV was higher by 16% in 1 case, but when compared in absolute dose between two methods, the difference was only 0.01Gy. 

Conclusions:

In this study, we did not find remarkable differences in the PTV doses calculated by the two different methods, PB and MC, in IMRT planning for skull base tumors.

Objectives:

Stereotactic Radiosurgery requires precise target localization and positioning accuracy. We evaluate the real time isocenter positioning accuracy of the target using optic-guided patient positioning system during spinal radiosurgery. The system, ExacTrac(BrainLab, Germany), was based on real time detects multiple passive body markers attached on the selected patient skin landmarks

Materials & Methods

We investigate the selected 8 metastatic tumor cases commonly closed to thoracic spinal code. All patients were stabilized with head & neck mask, vacuum cushion and treated with supine position. To evaluate the target point positioning accuracy in spinal radiosurgery, real time patient’s position monitoring was performed by comparing the current 3-dimensional positions of body markers in the treatment room with those of an initial reference positions in the CT scan images. For selected patient cases, we have checked isocenter translations per every 20 millisecond for 50 seconds during radiosurgery.

Results

The overall results of real time isocenter positioning translations were -0.24±0.17mm, -0.3±0.12mm, -0.24±0.15mm in x-, y- and z-directions and average body rotation about the x-, y- and z-axis were 0.17±0.09, -0.26±0.07, 0.04±0.06 degrees, respectively.

Conclusion

The optic-guided isocenter positioning accuracy was acceptable in spinal radiosurgery procedure, even if we consider the systematic errors. Real time position monitoring system provides more reliable precision in clinical application.

Objective:

Simulation plan of VMAT (volume modulated arc radiotherapy) was made for multiple brain metastases on Eclipse (Varian, Tokyo).  It was compared with original multi-isocenter dynamic conformal arc (MI-DCA) plan on iPlan (BrainLAB, Tokyo) (iPlan-MI-DCA) and simulation single-isocenter DCA plan by Automatic Brain Metastasis Planning (ABMP, BrainLAB, Tokyo) (ABMP-SI-DCA).  

Material & Method:

Dosimetric comparison was made among 3 plans on 2 cases of 9 small brain metastases (totally 18 tumors).  The mean volume of GTV (gross tumor volume) was 0.06 ml and the mean volume of PTV (planning target volume with 2 mm margin) was 0.40 ml.  D95 [dose covers 95% volume of PTV] was set with 95% dose of 22 Gy (100%).   Non-coplanar 3 arcs, couch angle of 90 degree (from vertex), 330 (from right), and 30 (from left) with 179 degree range, were employed in VMAT.   Besides Normal Tissue Objective function of Eclipse, virtual OARs (organs at risk) in rings surrounding each PTV were made during optimization for better dose concentration to each target. 

Result:

Conformity for each PTV was good in all 3 plans.  The means of gradient index, V[1/2 of prescription dose] / V[prescription dose], were 7.3, 5.4, and 4.6 in VMAT, iPlan-MI-DCA, and ABMP-SI-DCA respectively, showing tendency of wider low-dose spread around PTVs in VMAT   The means of maximum dose in PTV were 31.4 Gy, 22.8 Gy, and 24.5Gy respectively, showing tendency of higher dose delivery inside PTVs.

Conclusion:

VMAT is expected to treat multiple brain metastases with a short treatment time.  More detailed evaluation in various situations, such as different numbers and sizes of tumors, and different arc geometries in planning, is necessary to clarify the efficacy of VMAT. 

Study’s objectives

The scope of this study is to investigate the effectiveness of a 3D relative dosimetry end-to-end test for auditing purposes to Stereotactic Radiosurgery (SRS) modalities.

Methods

Anonymized CT DICOM data of a real patient were used to construct two identical hollow head phantoms of a radiologically bone equivalent material using a 3D printer. Both phantoms were filled with water equivalent polymer gel 3D dosimeters. Irradiations were performed as end-to-end audit tests for the Gamma Knife Perfexion (GK PFX) and CyberKnife G4 (CK) SRS treatment modalities and were implemented in collaboration with the Greek Atomic Energy Commission. Corresponding irradiation plans consisted of 4 and 7 small brain targets, simulating hypothetical multiple metastases cases. Gel dose read-outs gels were performed by departments’ MRI units using specially developed multi-echo pulse sequences, lasting a reasonable scan time of the order of 20 minutes. Measured relative 3D dose maps for both phantoms (derived by corresponding R2 relaxation rates) were compared against calculated ones (exported from the Treatment Planning Systems) in terms of Dose Volume Histograms (DVHs), 1D dose profiles, 2D isodoses maps and 3D gamma indices using passing criteria of 5% dose difference and 2 mm distance to agreement.

Results

Excellent spatial agreement (well within 1mm) was observed for all the irradiated targets in both modalities with maximum disagreement of approximately 0.8 mm observed for a small peripheral target located away from the MR scanner’s isocenter and irradiated with the GK PFX unit,  due to the MR-related geometric distortions and specifically Bo inhomogeneity which affect polymer gel/MRI measurements. In addition, dosimetric comparison between measured and calculated datasets showed a slightly increased gel over-response in low dose areas (<2 Gy) of the CK modalitydue to the limited low-dose resolution of the polymer gel. However, the mean 3D gamma index passing rate for all targets and modalities was 93%. DVH comparison resulted in an acceptable agreement within all targets (except for the aforementioned small target with the 0.8 mm spatial disagreement, as expected) and organs-at-risk for both phantoms/irradiations.

Conclusion

An end-to-end patient-specific audit test was implemented by a national external auditor to validate 3D spatial and dosimetric accuracy of SRS treatment techniques and specifically for demanding multiple brain metastases cases. The introduced methodology was found efficient for the purpose of external auditing in terms of workflow and accuracy despite the fact that gel showed an over-response in low dose areas.

Introduction

Uncertainty in geometric accuracy of mask-immobilized treatments on the Gamma Knife® Icon^TM is primarily caused by intra-fraction movement of the patient. Patient position is monitored using the High Definition Motion Management (HDMM) system via an infra-red reflective marker on the nose. This research uses clinical data from multiple Gamma Knife centers to address the unknown relationship between nose movement and intracranial target movement, building upon the established method from a preliminary phantom study[1]. This will allow Gamma Knife centers to make an informed choice when choosing a threshold level, above which, treatment is paused.

 Methods

Log files of patient treatments are interrogated using an in-house  MATLAB® script. Information regarding stereotactic coordinates of the nose marker, center of matrix and shot positions, plus the transformation matrix for each CBCT made prior and during treatment is extracted.

 Prior to treatment delivery, or if patient movement exceeds the pre-defined threshold, a CBCT is required to correct for positional changes. The vector shift of stereotactic coordinates of the nose marker, center of the matrix and shot positions are calculated for each CBCT using the transformation matrix.

 For all patients the ratio between the vector shift of the nose marker and the center of matrix is displayed in stereotactic space in order to determine any spatial variation in the ratio. This is repeated for patients with target volumes larger than 20 cc in order to determine any spatial variation in the ratio within the individual target, taking shot positions instead of center of matrix.

 Results

Whilst a preliminary phantom study[1] suggests that intracranial targets generally move less than the nose, in certain scenarios target movement could be larger. The stereotactic reference space showing the ratios for all patients is presented, along with variations of the ratio within large targets.

 [1]Wright G, Harrold N, Hatfield P and Bownes P (in-press) “Validity of the use of nose tip motion as a surrogate for intracranial motion in mask-fixated frameless Gamma Knife® Icon™ therapy”, JRSBRT

Objectives: To assess patient movement during frameless volumetric arc therapy (VMAT) of intracranial stereotactic radiotherapy using an Optical Surface Monitoring System (OSMS).

Methods: OSMS uses three cameras to track a region of interest on patients to determine translational (lateral (LAT), longitudinal (LNG), vertical (VRT)) and rotational (Rotation, Pitch, Roll) deviations, called Real Time Deltas (RTDs), from a reference position. After kV image-guided patient setup, an OSMS reference surface was captured and used to track patient movement throughout treatment. RTDs were recorded at discrete time points during treatment including before and after each beam, mid-beam when an OSMS camera is blocked by the gantry, and at end of treatment after the patient was returned to the initial reference position.

Results: We observed 106 fractions of 63 treatment plans delivered by flattening filter free (FFF) beams on an Edge linear accelerator (Varian, Palo Alto, CA).  The average change in position from reference capture to end of treatment was LAT = 0.01 ± 0.19 mm, LNG = -0.05 ± 0.34 mm, and VRT = 0.04 ± 0.14 mm with a mean magnitude of 0.34 mm and 95% less than 0.77 mm.  Rotational offsets from reference to end of treatment were Rotation = 0.02 ± 0.21°, Pitch = -0.01 ± 0.25° and Roll = 0.02 ± 0.16°. The average time between reference capture and end of treatment was 4.76 minutes.  For non-zero couch angles, average RTDs before beam on were LAT = -0.02 ± 0.4 mm, LNG = -0.66 ± 0.52 mm, and VRT = -0.11 ± 0.15 mm. Couch walkout was investigated as a potential source of the increased longitudinal offset but was not found to be correlated during phantom studies.

Conclusion: Overall, intrafraction patient motion assessed by OSMS is consistent with previous kV image guided studies. The largest deviations reported by OSMS were in the longitudinal direction, suggesting systematic error in tracking patient position along this direction when the couch is rotated. This was not found to be due to couch walk-out; therefore, other sources for this discrepancy are currently under investigation.

Purpose: In oktober 2016, the first ‘dry’ installation of the Gamma Knife was performed at our hospital: the ICON was installed without the 192 Co-60 sources. We will describe how this opportunity enabled us to create a unique workflow and integrate the GK into our radiotherapy department well before system start.

Method: Except for the sources, a complete and functional ICON was installed by Elekta. We customized the ICON workflow to include off-site mask making, Philips Ingenia 3T MR-RT scanner, dedicated contouring software, Mosaiq based treatments and automated procedures. Further, we were able to pre-test CBCT and HDMM functionalities, design procedures, and train the ICON-team.

Results:

We designed an efficient mask making routine in a specially equipped mouldroom that corresponds perfectly with the ICON procedure.

A priori, the Philips Ingenia 3T MR-RT scanner is not compatible with the G-frame. Instead of a head coil, we use flex coils in combination with the anterior and posterior coil. A customized adaptor for patient positioning accommodates both frame- and mask-based setups. Sequences are fine-tuned such that geometric integrity is guaranteed.

We use dedicated software that combines advanced registration and contouring tools and allows for creating margins and scripting. Depending on indication and preference, this software is now integrated in the GK workflow.

Workflows for frame- and mask-based treatments with Mosaiq were developed. We found Mosaiq to provide a reliable environment under normal conditions and established robustness in ad hoc scenarios.

We integrated the GK into our ICT-infrastructure to ensure reliable data transfer in accordance with current information security requirements. E.g., we developed a digital, PDF-based approval routine and prepared for automated instead of manual data transfer.

Pre-testing of CBCT and HDMM demonstrated the robustness and reproducibility of these systems and allowed for comparison of linac and ICON CBCT-QA procedures.

Planning-, workflow procedures were tailored to patient population, logistics and safety standards. For example, planning and image guidance protocols were developed for mask-based treatment of brain metastases based on clinical experience and dry-ICON sessions.

Hands on training sessions were organized to train the GK team on all procedures. This included extensive practice of emergency procedures without any exposure to radiation.

Conclusion: The ‘dry’ installation of the GK-ICON offered a unique opportunity to thoroughly prepare the infrastructure, department and team so that the clinical start may follow the loading of the machine swiftly and smoothly. We expect to have our ‘system start’ late march.

The objectives of this work were to investigate the differences between the two calculation algorithms available on Leksell GammaPlan (LGP), TMR10 and Convolution, and dosimetrically verify them in realistic patient-like conditions. 

An anthropomorphic and tissue equivalent head phantom (STEEV, CIRS, Norfolk, VI, USA) was employed to conduct end-to-end tests at four Gamma Knife centres in the UK. Four treatment plans were generated for an 8cc target on LGP using TMR10. All plans were delivered to the phantom and eight alanine pellets were used to measure point doses for each plan, four in the target and four in the brainstem. The plans were subsequently recalculated on LGP with the convolution algorithm and the measured doses were scaled using the ratio of beam-on times between the two plans. The accuracy of calculation algorithms was assessed by comparison of measured doses to LGP for TMR10, and comparison of scaled-doses to LGP for convolution. 

The measured doses show, on average, clinically acceptable agreement with TMR10 calculated doses both inside the target (1.1%) and the brainstem (2.4%). The rescaled doses compared to convolution show, on average, higher percentage differences both in the target (5.4%) and the brainstem (7.2%).

                                                     Measurement Vs TPS

                      Mean of 4 pellets in Target                Mean of 4 pellets in Brainstem     

                     TMR10           Convolution                      TMR10        Convolution

Centre 1          0.9%                5.3%                            1.7%                7.9%

Centre 2          1.7%                4.9%                            3.3%                6.7%

Centre 3          1.0%                5.5%                            2.9%                7.8%

Centre 4          0.9%                5.7%                            1.6%                6.2%

Average           1.1%                5.4%                            2.4%                7.2%

The results call for further investigation of the convolution algorithm. Further work should be conducted to evaluate the agreement of two-dimensional and three-dimensional measured dose distributions to convolution algorithm predictions.

Purpose: Nowadays  planning of radiosurgical treatment is impossible without neuronavigation systems. Basic diagnostic tools are the data  of  MRI visualization. For precise surgery planning most neuronavigation programs make specific demands to the MRI that differ from  standard scanning conditions. Our network of diagnostic centers suggests that a patient can have an opportunity to be scanned in any of them without a geographical reference. So a strict compliance with all technical aspects of MR imaging on any system is required. We set up a unified Protocol of examination on1.5 T MR tomographs of different classes  to achieve a unified quality standard.

Materials and methods: Created examination Protocol represents a set of three sequences, one of which was performed after the injection of a contrast media. At that , depending on the pathological changes, the examination can be supplemented with some specific sequences, on the recommendation of a radiosurgery physician. Depending on the type of the study – before or after a treatment, different coils, and, accordingly, different sets of adapted sequences are used . The difference in the choice of coils is due to the necessity of scanning a patient before a treatment with a special localiser  for a  further coregistration of tomography and planning  station coordinates .

Results: the main specifications are as follows:

- square matrix and the scan field;

- slice thickness 1 mm and isotropic resolution;

- a sufficient number of slices when scanning in the sagittal plane to cover bone structure and labels

- axial plane is preferred

- the use of slab-selective mode

- using turbo spin echo and 3d vibe sequence

- slice oversampling is not less than 20% to compensate  overlay crosstalk

- spatial resolution of at least 1.0x1.0x2.0 mm for T2 and 1.0x1.0x1.0 mm for T1

The preferred choice of MR visualization is made in favor of the 1.5 T compared to 3T due to the lesser susceptibility to distortion. Three-dimensional sequence is executed one time, the other projection can be recovered by using MPR without loss of quality due to isotropic resolution. The whole survey does not exceed 30 minutes including  contrast .

Conclusion: Compliance with the specific requirements during MR visualization and the use of a unified standard imaging allows you to collect diagnostic data compatible with any neuronavigational software which increases the accuracy of the planning of a radiosurgical intervention.

Objective: Gamma knife radiosurgery is safe and effective in treating patients with metastatic brain tumors. Metastatic brain tumor Gamma knife therapy is difficult to distinguish from meta lesion of less than 1mm because of difficulty in accuracy of boundary dose in treatment. In this study, we aimed to present the appropriate range of treatment for gamma knife radiosurgery by comparing 3D T1, delayed FLAIR and 3D T1 20 minute delayed images with MRI contrast injection. Materials and Methods: From January 2016 to December 2016, MRI images of 15 patients with less than 1 mm of brain metastasis were analyzed. Symphony 1.5T was used and images were acquired with a Brain coil. After administration of MRI contrast agent, 3D T1 tra, delayed FLAIR and delayed 20 minute 3D T1 tra were examined. The images were analyzed by Syngo Program. Results: The enhancement range of the 20-minute delayed image was increased from a minimum of 3% to a maximum of 8%, and the average volume of the 20-minute delayed images increased by 5% In delayed enhancement FLAIR images, 12 of 15 patients were found at the same lesion (80%) and three new lesions were found on the 20-minute delayed image. Conclusion: If metastatic lesions are bordered by 3D T1, FLAIR, and 3D T1 20-minute delay images after contrast injection, the 20-minute delayed image will be helpful for planning the gamma knife radiosurgery treatment. It is useful to detect lesions.

Background. Stereotactic body radiotherapy (SBRT) delivers a highly conformal, hypofractionated radiation dose to a small target with minimal radiation applied to the surrounding areas. Therefore, using the proper treatment planning techniques for SBRT is important. Intensity modulation techniques, such as static intensity-modulated radiation therapy (IMRT), modulated arc therapy (mARC), and helical tomotherapy (HT), are useful for spinal SBRT because of a rapid dose fall-off and spinal cord avoidance. This study aimed to compare the planning characteristics for hypofractionated spinal SBRT administered using three treatment techniques.

Materials and Methods. The factors evaluated for spinal SBRT planning were dose coverage, cord avoidance, target conformity, homogeneity, and dose spillage.

Results. Target coverage was 82.74 ± 3.35%, 80.92 ± 80.95%, and 85.01 ± 7.274% for IMRT, mARC, and HT, respectively. HT was therefore a powerful technique with respect to target coverage. The spinal cord dose for HT (mean, 1763.96; standard deviation, 164.48) was significantly different from those for mARC (mean, 1991.75; standard deviation, 248.00) and IMRT (mean, 2053.24; standard deviation, 164.48). In addition, the partial spinal cord volume at 20 Gy for HT (mean, 0.12, standard deviation, 0.01) was significantly different from those for IMRT and mARC (0.5 ± 0.10 and 0.56 ± 0.25, respectively). The conformity index was 1.30 ± 0.118, 1.08 ± 0.052, and 1.36 ± 0.231 for IMRT, mARC, and HT planning, respectively. mARC showed the highest conformity (p = 0.000).

Conclusions. HT used a narrow field pan beam and exhibited remarkable improvement of target coverage and cord dose, offering an important benefit to spinal SBRT. mARC had the highest target conformity and better high- and intermediate-dose spillage than HT and IMRT did, respectively. The HT and mARC planning techniques each have different advantages and disadvantages. Therefore, users should choose the appropriate planning techniques accordingly.

Purpose:

To study the advantage of using simultaneous kilo-voltage (kV) during mega-voltage (MV) treatment in spine stereotactic radiotherapy (SBRT)

Methods:

9 spine SBRT patients were treated for three to five fractions in Elekta Agility MLC with altogether 25 images using simultaneous X-ray volumetric imaging (XVI) during MV irradiation. The  intra-fractional positional errors were analyzed and input to treatment planning system (Monaco v5.0, Elekta AB, Stockholm, Sweden) to evaluate the dosimetric effect of not applying the shifts. Paired student t-test was used to evaluate the statistical significance of maximum dose to cord+2mm and PTV coverage after applying the shift. Statistical significant was considered with p < 0.05.

Results:

The average intra-fractional positional errors were 0.02 mm, -0.11 mm and -0.17 mm in X, Y and Z directions respectively. p < 0.05 was found in cord+2mm while > 0.05 in PTV coverage.

Conclusion:

Simultaneous XVI could be served as an alternative for intra-fractional monitoring and it has the potential to minimize the positional errors in spine SBRT in order to protect the organ-at-risks and provide more accurate dose delivery. The moderate conformity of the plans may be the reason of statistical insignificance of PTV coverage.

Purpose:

To study the dosimetric effect of different prescription isodose levels (IDLs) in spine stereotactic body radiotherapy (SBRT).

Methods:

6 spine SBRT (2 cervical spine, 2 thoracic spine and 2 lumbar spine) cases at 65%, 75% and 85% IDL with prescription of 24 Gy in three fractions were optimized in Monaco v5.0 (Elekta AB, Stockholm, Sweden). The three IDLs were achieved by using target penalty on the PTVs while keeping the beam geometry and the organ-at-risk parameters the same. The plans were then delivered on Elekta Agility MLC linear accelerator and measured with Octavius 4D system and SRS1000 (PTW, Freiburg, Germany). Maximum dose to cord (Cord), PTV coverage, Modulation (M) reported in Monaco, Paddick's Conformity Index (CI), Gradient Index (GI), Monitor Unit (MU) and Gamma Analysis (GA) with 1.5%/1.5mm (local dose) were compared among the three groups.

Results:

The average values at corresponding IDLs are: Cord = 21.3Gy, 21.2Gy and 21.2Gy; PTV coverage = 94.4%, 95.6% and 94.1%; M = 3.6, 3.0 and 2.9; CI = 0.75, 0.76 and 0.79; GI = 4.79, 4.74 and 5.1; MU = 4010, 3267 and 2752; GA = 93.3%, 93.1 and 93.2%.

Conclusion:

Different IDLs provided similar OARs doses, PTV coverage and gamma analysis results. On the other hand, the highest IDL gave the best conformity but the worst fall-off, although the magnitude is not significant. Besides, the lower the IDLs, the more the modulation and MU required.

Purpose:

To study the dosimetric impact of modulation on 3D gamma index and dose-volume histogram (DVH) due to the imperfection of delivery system in spine stereotactic body radiotherapy (SBRT).

Methods:

Eighteen plans of spine SBRT with prescription of 24 Gy in 3 fractions at different reported modulation levels were created in Monaco v5.0 (Elekta AB, Stockholm, Sweden) by using target penalty to boost the dose within PTV to 65%-85% isodose level. They were then delivered on Elekta Agility MLC linear accelerator and measured with liquid-filled detector array (PTW, Freiburg, Germany) having 2.5 x 2.5 mm² resolution. The measurements were subsequently reconstructed into a 3D dose distribution on both phantom and patient's CT images. Correlation (r) between isodose level (IDL), Gamma Index (GI) (1.5mm/1.5% local dose), absolute percentage change in cord (%cord) and in PTV coverage (%PTV) against modulation (M) with statistical significant at p<0.05.

Results:

Both IDL and %cord showed moderate correlated with M (r = -0.54 and -0.48 respectively with p<0.05) while other two were statistically insignificant.

Conclusion:

Gamma passing rate is not sensitive to the plan modulation. The weak correlation of PTV coverage with modulation is perhaps due to the imperfect conformity. The optimal modulation would be around 3.5 to compromise the change in dose to cord and the risk of vertebral fracture from extremely low isodose level. 

With the advent of SBRT, metastatic spine lesions became well suited targets for this new treatment modality. Thanks to the advances in image guided radiotherapy, durable local control is today often the aim. This in turn has pushed the surgical management.

However, the management of spinal metastases with spinal cord compression or impression in the thecalsac has been posing challenges, in particular in a post surgical scenario where a segmental resection is followed by the vertebral stabilization with titanium implants.

The challenges of delivering an ablative target dose and restrict dose to medulla/cauda equine (OAR (Organs At Risk)), requires small geometrical uncertainties. Components of the latter are; delineation of the tumor  and  OAR, especially difficult amid the presence of titanium implants. MRI T2 based sequences has become the imaging modality of choice for the delineation task. Furthermore, the registration of MR and CT images (for dose calculation) and uncertainties in the CBCT-online imaging at set-up implies the use of a PRV (Planing Risk Volume) for OAR. The PRV-margin has a bearing with the fixation method and has also to account for internal motion of the OAR. This margin and the relative position of the tumor to the OAR steers the dose gradient needed between target and medulla.

Target and OAR delineation is performed as high resolution segments on MR T2 with 3mm thick slice and 0.16 mm transversal resolution. These structures are transferred to a co-registrated  CT with 2mm slice thickness and 0.17 mm transversal resolution. PRV 2mm around medulla is defined, based on the observed behaviour of patients on the stereotatic bodyframe. At the onset of the  post-operative SBRT in vertebras at Karolinska an additional step was introduced, which is a co-registration of the CT with a pre-planning CBCT. This step prior to planning reveals important information of the inter fractional positional errors that can be expected for a specific patient. If no rotation (below 0.2 degrees) is present this CBCT image is used as reference image for the ensuing online matchings. The reason for this is the high spatial resolution of the CBCT when a reduced field of view mode is used. Automatic matching between CBCTs rules out user induced matching errors.

Matching errors and intra fractional movement was observed to be less than 0.5 mm. Pre planning CBCTs and co-registration to CT provides information of paramount importance when discussing margins and the acceptable and safe dose gradient. 

Preliminary results of stereotactic radiotherapy (SRT) for spinal arteriovenous malformation (AVM) were presented in 5 cases.  Two were male and three were female.  The median of age was 32 years (range 23 to 54 years).  Spinal intramedullary AVM were located in cervical spine in three and in thoracic in two.  SRT with 20 Gy in 4 fractions was delivered to the nidus in every case.  At the end of the median follow-up period of three years (range 2 to 7.5 years) the neurological symptoms and signs were improved or same as before treatment, though temporary adverse effect was developed between a half year and one year after SRT in one case. The nidus was disappeared three years after SRT in one case. The nidus was decreased in size in one. In the other three cases the niduses were unchanged. In conclusion, SRT with 20Gy in 4 fractions was thought to be a safe treatment, though this study dealt with only a small number of patients with short term follow-up period. 

Purpose: To suggest optimal planning target volume (PTV) margins in single-fraction stereotactic radiosurgery (SRS) of the spine.

Patients and Methods: From December 2014 to July 2016, 40 patients received 42 fractions of single-fraction SRS for spinal tumors in thoracic and lumbosacral spines using a volumetric modulated arc therapy technique and patient immobilization. Before each treatment, kilovoltage cone-beam CT (CBCT) images were obtained for a 4 degrees of freedom (DoF) correction of patients alignment (translations and yaw) using planning CT as reference. After each correction, additional CBCT was acquired to verify adequate patients alignments just before treatment delivery (pretreatment CBCT). Immediately following SRS, CBCT was acquired again (posttreatment CBCT). Residual setup errors (SE) between planning CT and CBCT were determined by a 6 DoF manual matching. Intrafraction motions (IM) were calculated as differences in SE between pretreatment and posttreatment CBCT. For comparison, three clinical target volumes (CTV) were created by translating and rotating original CTV by pretreatment SE alone (CTV_SE), IM alone (CTV_IM) and both SE and IM (CTV_SEIM), respectively. The impact of various PTV margins on CTV coverage was evaluated. A provisional criterion of adequate CTV coverage was that PTV encompasses at least 97% of CTV.

Results: Time interval between pretreatment and posttreatment CBCT was 6.8±2.5 min (mean±2SD). The 2SDs of translation were 0.7 mm, 0.8 mm, 1.1 mm (SE) and 0.7 mm, 0.8 mm, 1.1 mm (IM) in lateral, vertical and longitudinal directions, respectively. The SDs of rotation were 1.7°, 1.1°, 1.6°(SE) and 1.1°, 0.8°, 1.1°(IM) for pitch, roll and yaw, respectively. Without margins, PTV showed adequate coverage for CTV_SE, CTV_IM, and CTV_SEIM in 48% (20/42), 71% (30/42) and 48% (20/42) of fractions, respectively. With 1-mm uniform margins, PTV was adequate in 95% (40/42), 98% (41/42) and 100% (42/42) of fractions for CTV_SE, CTV_IM and CTV_SEIM, respectively. A 2-mm uniform margin was adequate for three CTVs in all fractions.

Conclusions: With appropriate immobilizations and 4DoF corrections, a uniform 1-mm PTV margin may ensure adequate CTV coverage in most treatment sessions of spine SRS. Combination of short treatment time and small IM may obviate the need of treatment interruption for additional image guidance. A 1-mm PTV margin is still needed to address IM even in case of ideally perfect 6 DoF patient alignments.

With a purpose to evaluate the current status of Gamma Knife surgery (GKS) for ocular and/or extraocular intraorbital diseases in Europe, European Gamma Knife Society performed a dedicated survey.

In autumn 2014 the electronic questionnaire comprising 12 questions was sent 4 times to all European Gamma Knife centers (49 as on October 2014). Responses were obtained from 20 centers (41%). Among the latter 8 centers mentioned that they do not perform GKS for ocular and/or extraocular intraorbital diseases, whereas 12 centers did it.

The overall number of treated patients was >100, 30-100, and <30 in 4, 2, and 6 centers respectively. Uveal melanoma was the most common pathology treated with GKS (11 centers), followed by other ocular or extraocular intraorbital tumors. The main factors influencing decision making on application of GKS were lesion size, location, stage of disease, absence of advanced metastatic disease, proximity of the critical intraocular structures to the target, visual acuity, patient general condition and co-morbidity. Ophthalmologists were constantly involved in treatment-decision making in 10 centers. Eye immobilization technique during irradiation included invasive suturing of the rectus muscles (7 centers), retrobulbar anesthetic blocking (2 centers), and fixation on the light source (1 center); two centers did not apply eye fixation during GKS. The marginal dose depended on the tumor size and type of pathology, and frequently was at least 35 Gy (7 centers). Postradiosurgery treatment strategy included observation (8 centers), pre-planned local tumor resection (1 center), and enucleation (1 center). Second-time GKS in case of disease progression was applied in 4 centers. Marginal dose for second-time GKS was either similar to primary treatment (3 centers), or corresponded to the cumulative dose of 56-57 Gy (1 center).

In conclusion, there is definite variability in indications, technique, and general treatment strategy between European Gamma Knife centers performing GKS for ocular and extraocular intraorbital pathology. Development of the optimal radiosurgical treatment strategy in such cases may be done by comparative evaluation of outcomes in collaborative multicenter studies.

Objective : In order to review cases treated with Gamma Knife Radiosurgery in an year on a facility being used for the first time in the Kingdom.

Methods :Leksell Gamma Knife Perfexion system, has been installed at PSMMC in July, 2013.

First patient treated at this machine was on Nov.3^rd, 2013. And 19 patients have been successfully treated at the end of one year, which includes; 7 patients of Arteriovenous malformation (AVMs), CP angle Schwannomas ; 6, CP angle Meningioma; 2, Trigeminal neuralgia; 2, Pituitary Adenoma; 1, and one case of Recurrent single focus of metastasis from Colon cancer.

A Radiosurgery data-base has been established. All the patients who received their treatments are kept on regular follow up, being updated in the database based on their follow up visits in combined clinic at Radiation Oncology Department, PSMMC.

Results : Mean Doses delivered were as follows : AVMs;  18 Gy, CP Angle tumours : 12 Gy,  TNs : 80 Gy, Pituitary Macroadenoma : 13Gy, Metastasis : 16 Gy. These doses are marginal doses prescribed at 50 % isodose levels. RTOG/ Timmerman’s Tables of dose constraint levels using single fraction treatment for Organs at risk in brain were used.

All the patients, had safe course of treatment with no acute complications.

As Expected, Trigeminal Neuralgias(N : 2) treated at our facility have been found to experience quickest relief of symptoms within 2-3 months achieving BNI scores of II and III.

All the CP angle Tumors  were found to have stable radiological findings on 6 month follow up MRI with 37.5% (N: 3/8)patients commented stable hearing compromise but marginal improvement in tinnitus and vertigo.

28.5% (N : 2/7) patients of AVMs felt better in headaches and requirement of anticonvulsants (both had elapsed 6 months or more of follow up).

One patient of AVM(referred from another hospital)reported to have  worsening headaches, although reduced seizure activity, for which he underwent work up and found no complications. He felt improvement in symptoms on neuropathic medications prescribed by his neurologist at his primary hospital.

Conclusion : Gamma Knife Radiosurgery is a safe and effective modality of treatment for benign and malignant brain diseases with certain specific indications. More patients including metastatic brain lesions should be considered for treatment with this modality expecting good cognitive outcomes. 

Introduction: Icon is the newest model of Gamma Knife equipped with a cone beam CT (CBCT) and high definition motion management system. It is possible to perform a frameless mask based Gamma Knife surgery (GKS). The authors report the early experience of clinical cases of GKS executed with Gamma Knife Icon.

Materials and Method: Between March 2016 and December 2016, 670 procedures were performed with GK Icon and 199 (29.7%) cases were mask-based procedures. There were 78 brain metastases, 66 meningioma, 31 vestibular schwannoma cases. Among the mask based GKS, 75 (37.7%) cases were fractionated into 3 to 5 sessions. Various aspects on mask-based GKS such as tumor volume, prescription dose, irradiation time, number of lesions, and patient age were compared with frame-based procedures. The accuracy of mask based GKS was assessed by co-registering conventional CT images of an anthropomorphic phantom to Icon CBCT images and measuring locations of the irradiated points. The error related with the image co-registration was in order of a millimeter.

Results: Distribution of indications of mask-based GKS was similar with that of frame-based system. Though the mean tumor volume of mask-based procedures was similar to that of the frame-based procedures (3.6 ± 6.0 cm³ versus 3.0 ± 5.9 cm³), shorter irradiation time per procedure was used for mask-based patients (29.9 ± 18.7 min versus 49.1 ± 33.7 min). It is because more lesions were treated in fractionated procedures in mask-based GKS (37.7% versus 25.7%). While only four (5.5%) patients of vestibular schwannoma patients were treated with fractionation in frame-based GKS, 16 (51.6%) patients were fractionated in mask-based procedures. The main reason for fractionation was that the target was closed to an organ at risk or it was too big for a single session irradiation. The mean target volume of fractionated cases was 6.1 ± 8.4 cm³ while it was 2.2 ± 3.3 cm³ in single session treatments among mask-based procedures. The average number of interrupts during an irradiation session was 1.4 +/- 1.9 times. Three patients were converted from the mask-based to frame-based because of involuntary movements during CBCT imaging before irradiation.

Summary: Frameless mask-based GKS could be performed with Gamma Knife Icon with accuracies comparable to frame-based GKS. With careful selection of patients to keep the irradiation time not to be too long, most of the mask-based patients could finish the frameless GKS with high satisfaction rate.

Objective

Cardiac implantable electronic devices (CIEDs) are more and more common among patients referred to radiotherapy. Due to continuous development of new models of the CIEDs and introduction of new types of medical accelerators there is an urgent need for up-to-date information on the susceptibility of modern CIEDs to ionizing radiation, especially in patients subjected to new techniques of radiotherapy or irradiated with new kinds of devices.

The aim of our study was to test a set of modern CIEDs in the setting of robotic radiosurgery for target volumes localized in the cranium and cervical spine.

Material and methods.

A set of four CIEDs from two different vendors was placed on an anthropomorphic phantom to simulate the real position of the devices in a patient. Five treatment plans were made to simulate irradiation of a convexity meningioma, brain metastasis, trigeminal neuralgia, and tumors located in the upper and lower cervical spine. The treatment plans were optimized to avoid direct irradiation of the CIEDs. The phantom was irradiated with 6 MV photons and after each treatment the devices were interrogated to test for possible malfunctions. A set of thermoluminescent dosimeters was also placed on the chassis of each device in order to verify the calculations of the treatment planning system.

Results.

All the devices functioned correctly after the irradiation in spite of proximity of the irradiated target volumes and movement of the working accelerator above the CIEDs which potentially could be a source of electromagnetic interference.

The mean values of dosimetric measurements made on the surface of the devices were as follows: 4.3, 4.9, 5.6, 9.5, and 8.5 cGy for the convexity meningioma, brain tumor, trigeminal neuralgia, and tumors located in the upper and lower cervical spine, respectively. The doses calculated by the treatment planning system were 1.8, 2.3, 8.2, 11.2, and 10 cGy, respectively.

Conclusions

The preliminary results indicate that an operating CyberKnife device poses no direct threat to modern CIEDs even in case of irradiation of target volumes close to a CIED. The results require confirmation on a larger set of CIEDs from various vendors to make them more generalizable.

The treatment planning system underestimates the dose to CIEDs in case of locations distant from the target volume which should be taken into account in clinical practice.

Objectives: Gamma Knife Radiosurgery (GKRS) is generally performed under minimal sedation with local analgesia provided by non-anesthesiologists. Patients with restricted airway access who are isolated from medical personnel are a considerable risk. Especially in the pediatric population, careful observation of both respiration and ventilation is required because moderate sedation is frequently used. Thus, early detection and intervention are essential for safe respiratory management. There is no standard for basic monitoring during GKRS, with peripheral oxygen saturation (SpO₂) as the sole respiratory monitor. According to the practice guideline for sedation and analgesia by non-anesthesiologists released by American Society of Anesthesiologists, the adequacy of ventilation should be evaluated by continuous observation of qualitative clinical signs and monitoring for the presence of exhaled carbon dioxide during moderate or deep sedation. Since 2016, we have adopted Capnostream^® (Medtronic) to evaluate both oxygenation and ventilation during GKRS. This dual parameter monitor of SpO₂ and end-tidal carbon dioxide (EtCO₂) is used in all cases as a standard respiratory monitor. We herein demonstrate a pediatric case to clarify the effectiveness of EtCO₂ monitoring during GKRS.

Methods: A 9-year-old boy underwent GKRS due to arteriovenous malformation on left medial occipito-temporal lobe (Spetzler-Martin grade 2, Pollock-Flickinger score 0.75) with 22Gy at 50% isodose (PIV 0.98 cc). As the patient was considered too young to receive GKRS under minimal sedation with local analgesia, monitored anesthesia care with a combination of dexmedetomidine and remifentanil was planned. Supplemental oxygen was provided through nasal cannula and SpO₂ and EtCO₂ were continuously monitored using Capnostream^® along with electrocardiogram, heart rate, and noninvasive blood pressure.

Results: Conscious sedation with stable cardiorespiratory status was successfully provided at the beginning of the procedure. However, additional sedation (midazolam, 1 mg) and analgesia (fentanyl, 120 mcg in total) were required because the patient was agitated when we fixed a Leksell frame with topical analgesia. Approximately 1 minute after midazolam administration, respiratory depression was detected as his respiration rate decreased below 10 and EtCO₂ level dropped from 38 to 16 mmHg, while SpO₂ remained 98%. Hypopnea was detected by capnogram almost 2 minutes in advance of desaturation, therefore prompt respiratory support with jaw-lift could recover his respiratory status without pharmacological reversal or invasive airway management.

Conclusion: Sole monitoring of SpO₂ is insufficient to detect respiratory decline. From an anesthesiological point of view, simultaneous EtCO₂ monitoring with SpO₂ enhances patient safety and improves the clinical efficiency of GKRS.

Introduction:

3D printing technology helps to make of prostheses, assistive devices, production of hearing aids, prostheses, nose, ears, eyes, teeth, breast reconstruction, of bone substitutes. Method of 3D printing cells (bioprinting) also helps in treating skin injuries and regenerative medicine in which the patient's stem cells form filling printers.

Methods and Results:

3D printing we used in neuro-oncology at radiosurgical planning performance. Tumour’s shape and boundaries with virtual model of 3D printing increases the accuracy of the therapeutic doses and subsequently significantly reduces irradiation to risk structure.

3D printing helps in deciding which modality of irradiation is the safest therapy for patient. It helps the surgeon to decide the way of radiotherapy (e.g. Linac treatment, Gamma knife, Cyberknife or proton beam irradiation).

Mentioned 3D methods contribute to a better understanding of localization and progression of tumours compared to only 2D view, which also helps for undergraduate and postgraduate education.

In collaboration with physicians from the Ophthalmological clinic of the Faculty of Medicine in Bratislava, we have created a set of 3D physical models of eye and tumours for radiosurgical operations, that we have also been used for the postgraduate teaching of physicians and medical students.

We are preparing a new model in rehabilitation and physiotherapy for better differentiation of the shoulder joint pain origin which will be determined on therapeutic pain procedure with the shoulder joint for targeted rehabilitation, using the model comparison of 3D printing shoulder joint and cervical spine.

Conclusion:

We used 3D printing in neuro-oncology with radiosurgical planning performance. With the 3D model printing, we increased the accuracy of the therapeutic dose and at the same time we calculated the lowest dose to risk structures. 3D printing method helps determine which modality radiation is optimal. For operations of the eye’s melanoma we created a set of 3D physical models of the eye and tumour. In the near future, 3D printing will also help us for targeted rehabilitation.

Purpose: In Japan, since Leksell Gamma Knife Type B was introduced in 1990, version up has been repeated from type C, 4C to Perfexion. Leksell Gamma Knife Icon was developed and it became possible to treat without frame. Our hospital has been using from December 2016, and we will report on its initial use experience.

Material and results: We started treatment with Gamma Knife Icon from December 5, 2016. At our hospital, treatment of the large lesion requiring fractionated irradiation or treatment for cases with less than 1 hour irradiation time has been performed with mask fixation system. During the 77 gamma knife treatments performed by February 24, 2017, the mask system was applied to 28 patients (36.4%) according to the above-mentioned criteria. The breakdown was 8 cases (28.5%) with large tumor requiring fractionation, 19 cases (67.6%) with a few lesions, and 1 the other case (3.6%). In principle, irradiation was performed at 35 Gy / 5 fx for large metastatic brain tumors, whereas for tumors with a small number of small lesions, margin of about 1 mm was taken and a single irradiation at 20 Gy was performed. Treatment was rarely interrupted by patient movement, and it was able to be treated smoothly. In the only one case, it was not able to withstand compression by the mask, requiring sedation.

Conclusion: Even with fixation with a mask, the movement of the patient was unexpectedly small. The range of treatment could be expanded from fractionated irradiation to day treatment. Gamma Knife Icon was useful for widening the range of treatment.

ABSTRACT
Introduction: Patients with traumatic brain injury (TBI) need frequent computed tomography (CT) of the head for assessment
and management. In view of the associated polytrauma, hemodynamic instability, and various in‑dwelling catheters and
tubes, shifting of patients for CT scans may be difficult.
Aims and Objectives: To assess the role of mobile CT (Ceretom®; NeuroLogica Corporation, Boston, MA, USA) in a trauma
center with respect to patient management.
Materials and Methods: In this retrospective study over 67 months (June 2009 to January 2015), the number of CT scans done,
the time taken for CT and downtime were evaluated. Also, for the first 1000 mobile CT scans, the clinical and radiological records
of all patients with TBI who underwent imaging using the mobile CT scanner in the intensive care units (ICUs) were analyzed.
Observations and Results: A total of 10,000 mobile CT scans were done on the mobile CT scanner till January 5, 2015. Of
the first 1000 patients evaluated, 75.3% had severe TBI, 15.1% had moderate TBI, and 9.6% had mild TBI. 78.1% patients
were on ventilator, with 80.2% requiring sedation and 8.4%, an inotropic support. An in situ intracranial pressure monitoring
was present in 21.1% of patients. In all, 12.4% of patients had long‑bone fractures requiring skeletal traction; and, the tube
thoracostomy was in-situ in 7.4%. No adverse events related to line malfunction/pullout occurred. The mean time for the
performance of imaging using the mobile CT scan was 11.6 minutes compared with 47.8 minutes when patients were shifted
to a conventional CT scan suite. The machine was nonfunctional 94 times, with an average downtime of 4.2 hours (range
2–72 hours). The life‑cycle cost per mobile CT scan was Rs. 1340.
Conclusions: A mobile CT has considerably changed the management response time in the neurosurgical intensive care
unit (ICU) setup and decreased patient transfer times and the associated complications. Inclusion of a mobile CT scanner
in the armamentarium of a neurosurgeon as a “bedside tool” can dramatically change decision making and the response
time. It should be considered as the standard of care in any large‑volume emergency department or neurosurgical facility.

Ferrante P, Lagmari M, Carron R, Hayashi M, McGonigal A, Daquin G, Villeneuve N, Laguitton V, Bartolomei F, Chauvel P, Regis J.

Background: HHs associated to epileptic comorbidity are highly variable from patient to patient in term of clinical presentation severity and quality of responds to Radiosurgery.

Materials and Methods:
Between October 1999 and October 2007, 57 patients were investigated, included and treated by Gamma Knife in Timone University Hospital.

Inclusion criteria were: patients presenting with a conspicuous HH on their MRI scan, an ongoing or past history of gelastic seizure, the demonstration of drug resistance of the underlying epilepsy and patients being aged between 3 and 50. The parents (or patients when possible) were required to sign the informed consent. Exclusion criteria were: patients already included in another official trial, patients with a psychiatric disorder too severe to be amenable to radiosurgery and to allow satisfactory follow up, patients unable to carry out EEG recording with a minimum of one video-recorded seizure and pregnant women.

Follow-up of more than 3 years was available for 48 patients.
We evaluated for each patient the electroclinical presentation and the severity of the epilepsy, the pattern of psychiatric comorbidity and the neuropsychological performances before the Radiosurgery. At the last follow up the quality of epileptic, psychiatric and cognitive improvement was assessed. The topography of the HHs, its extension to mammillary tuberal and supraoptic hypothalamic, its lateral extension contact to pituitary stalk were evaluated.  

Results: Preliminary results are showing that all the patients have an involvement of the mammillary hypothalamus and mammillary bodies. All the patients with Precocious Puberty have HH involving the tuberal hypothalamus. Statistical study of the relationship between our anatomic parcellation and clinically relevant features will be presented. 

Conclusions: Our preliminary data are demonstrating the importance of taking into account the precise topography and anatomical extension of HH in order to better predict clinical presentation, evolution, optimum surgical strategy and prognosis.

Yanez L., Fariña A. (Radiation oncology, Fundación Arturo López Pérez), Cantillana P., Diaz B., Zurita V., Ruiz A., Torzok K., Marsiglia H., Sallabanda K.

In Santiago the first Cyberknife equipment (model M6) was installed at Fundación Arturo López Pérez in 2016. The first patient was treated on March 21, 2016.
So far sixty adults have received Cyberknife radiosurgery, intra and extracranial.

Interestingly three cases have been young adults (mean 35 years old, 2 female one man) with the diagnosis of cervical schwannomas. None of them has stigmata or diagnosis of Neurofibromatosis. Two patients had a confirmatory biopsy.

Since this treatment has not very extensibly been published in the literature the aim of this work is to review the pathogenesis, incidence, diagnosis, radiology characteristics and treatment of spinal schwannomas and the possible indications, therapeutic doses and results of radiosurgery. A brief bibliographic review will be presented, specially a discrete number of case reports informing about radiosurgery treatments in similar cases.

We also will present our three treatments in detail, regarding contouring and doses.

In summary we will offer a review of an infrequent pathology treated with Radiosurgery and the characteristics surrounding it ́s therapeutics.
For this complicated surgical location, radiosurgery presents as a reasonable and safe option for patients with symptomatic or enlarging schwannomas, clinically effective.

Purpose:

The purpose of this study is to estimate the dose distribution to the normal brain parenchyma and radiological response following cyber knife Radiosurgery of large AVMs.

Material and Methods:

A total of 15 patients were included in this study that underwent staged volume cyber knife Radiosurgery. All patients were pre-diagnosed for arterio-venous malformation (AVM), with a surgical and neuro interventional opinion. An immobilization device was made for every patient and then was performed a contrast enhanced Computed Tomography and Magnetic resonance Imaging. Tumor delineation was performed on Multiplan 4.0 with division of AVM nidus into two volumes.

Results:

In our study included patients were 13(86.7%) male and 2(13.3%) female with mean age of 41 years with a mean AVM volume for first stage Radiosurgery of  12.44ml and a median volume of 11.20ml. however the second stage mean and median  AVM volume stood, 12.99ml and 11.66ml respectively. The isodose line ranged from 73 to 81 for first stage of AVM and 62 to 82 for the second stage with a prescription dose of 16 Gy. The maximum normal brain volume 1643ml and a minimum volume of 1219ml. post Radiosurgery MRI T2 weighted hyper intensity was observed in 4(26.66%) patients which remained asymptomatic. No neurologic deficit or post contrast hyperintensity observed in the patients on follow-up MRI.

Conclusion:

In a mean follow-up period of 6 months patients treated with staged volume technique at our institute were not observed with radiation induced necrosis or any neurologic deficit. The only complication was post radiation T2 weighted changes which were asymptomatic. In our patients with large AVMs staged volume Radiosurgery reduced the risk of radiation necrosis.

Background To assess the efficacy of stereotactic radiosurgery (SRS) for intracranial dural arteriovenous fistulas (DAVFs), the authors retrospectively reviewed our 7-year experience. We evaluated obliteration rate, improvement in clinical symptoms, and complications.

Methods Thirty patients with DAVF underwent SRS using Gamma Knife between 2009 and 2015. Twenty-three patients were treated with SRS alone and seven patients underwent SRS for residual or recurrent DAVFs after embolization or surgery. Chemosis, diplopia, and pulsating tinnitus were the most common symptoms. Median target volume was 2.9 cm³ (range, 0.8–13.6 cm³) and median radiation dose to the target was 17 Gy (range, 12–20 Gy). Median follow-up period was 33 months (range, 6–82 months).

Results At the last neuroimaging follow-up, DAVFs were totally obliterated in 23 patients (77 %) and subtotally in seven (23%). At the last clinical follow-up, 21 patients (70%) showed complete recovery and nine (30%) showed incomplete recovery in symptoms or signs. None experienced worsening symptoms or signs. Asymptomatic perilesional edema after SRS occurred in one patient (3%). Total obliteration rates after SRS were 43% at 1 year, 79% at 2 years, and 95% at 5 years. Improvement rates of neurological function after SRS were 12% at 1 month, 52% at 2 months, 72% at 3 months, and 96% at 6 months. A multivariate analysis revealed that Borden type 1 (p = 0.019, hazard ratio, 3.254, 95% confidence interval, 1.216-8.707) was significantly associated with symptom improvement.

Conclusions SRS for intracranial DAVFs provided a high obliteration rate and a relatively low risk of radiation-induced complications. In selected benign cases without cortical venous drainage, SRS is a safe and effective treatment for symptom relief and fistula obliteration, even though the time course of improvement is longer than those of embolization and surgery.

Objective: We reviewed our 7-year experience to assess the efficacy of stereotactic radiosurgery (SRS) for dural carotid cavernous fistulas (DCCFs). We analyzed the clinical outcome, complications, and angiographic results.

 Methods: We performed a retrospective analysis of 18 consecutive patients with DCCF treated by SRS alone using Gamma Knife between 2009 and 2015. Median target volume was 2.6 cm³ (range, 0.6–11.6 cm³) and median radiation dose to the target was 17 Gy (range, 14–19 Gy). Median follow-up period was 30 months (range, 6-65 months).

Results: Fifteen patients (83%) achieved a total obliteration of the DCCF and a subtotal obliteration of the DCCF was achieved in three patients (17%). Total obliteration rates after SRS were 53% at 1 year, 90% at 2 years, and 100% at 5 years. Twelve patients (67%) showed complete recovery from symptoms or signs and six patients (33%) showed incomplete recovery. Improvement rates of neurological function after SRS were 56% at 1 month, 72% at 3 months, and 94% at 6 months. None of the patients experienced radiation-related complications. A multivariate analysis revealed that absence of cortical venous drainage (p = 0.023, hazard ratio, 3.902, 95% confidence interval, 1.157-13.166) was significantly associated with symptom improvement.

Conclusion: SRS for DCCFs offered a high obliteration rate with low risk of radiation-induced complications. In patients with benign DCCFs that are not amenable to embolization or microsurgery, SRS is a safe and effective treatment for complete obliteration of the arteriovenous shunt and for improving the quality of life.

Objectives : To assess the treatment outcomes in the patients diagnosed with arteriovenous malformations( AVMs) treated by LINAC based stereotactic radiosurgery(SRS).

Methods and Materials : We retrospectively analyzed 30 patients diagnosed with AVM treated with LINAC based radiosurgery between 2008 to 2016. Male to female ratio was 23 : 7. Out of 30 patients 18 patients had only SRS , 12 patients had onyx embolization followed by SRS .There were 9 (30%) patients with hemorrhages ,5  patients (17%) presented with neurological deficits. The median age at presentation was 30 years ( range 14-60 years). The median PTV volume was 6.8 cc (range 0.9 -54 cc).The median dose prescribed  was 18Gy (range 16 – 24Gy) Modified radiosurgery based AVM grading score was calculated for all the patients. The median modified AVM score was 1.47(range 0.5 -5.68). The follow up was done by clinical examination as well as  serial MRI angiography or DSA for all the patients.

Results : The median follow of the entire cohort was 60 months ( range 24-96months) .The obliteration rates for patients followed up for  3 years and 5 years was 75%  and 86.1% respectively.  The obliteration rates  based on Modified radiosurgery based  AVM grading scale for  a score of ≤ 1, >1 to ≤ 2, and >2  was 84.7%, 60% and 54% respectively. On Logistic regression , there was no  statistical correlation between obliteration rates and AVM score, but there was a trend favouring better obliteration rates for AVM score of ≤ 1. On univariate analysis , age (<35 years) (p=0.007) and PTV volume (<7cc) (p=0.04), had better obliteration rates.  3 patients had hemorrhage, from the AVM after irradiation. None of them were fatal. 3 patients had neurological deficits post irradiation which was reversible. 

Conclusion : SRS is a preferred  non invasive treatment modality for younger group of patients (<35 years)  and for smaller volume AVM (<7cc) , with acceptable morbidity. Patients with low modified radiosurgery based AVM score have better obliteration rates, though more patient numbers and  long term follow up is warranted for better outcomes with SRS

Objective: To evaulate late complications after Gamma knife radiosurgery(GKRS) for cerebral arteriovenous malformation(AVM) other than delayed cyst formation or signal change in MRI

Methods and Materials: Between May 1992 and June 2012, 844 GKRS was done for cerebral AVMs. We retrospectively analyzed our data and evaluated any complications 3 years after GKRS.

Results: Among 844 procedures, 2 cases (0.2%) of dural arteriovenous fistula and 11 cases (1.3%) of growing organized hematoma were noted. DAVF was emerged at 16 and 36 months each. Growing organized hematoma was emerged at mean 14.4 years (range 7.6~24.4) after 1st GKRS. Eight among 11 patients under went open craniotomy for the removal operation, and 9 patients had repeated GKRS for the remnant AVMs (2 times: 6, 3 times:1, 4 times: 2).

Conclusions: Late complications after GKRS for cerebral AVMs are well known though the incidence is low. We recommend long-term follow-up of patients who undergo GKRS for cerebral AVMs, since cyst formation or expanding hematoma could be emerged irrespective of obliteration status. Once these delayed lesions emerged, the follow-up should be done with caution, and it seems that early operative management is needed than conservative especially for the growing organized hematoma.

Stereotactic radiosurgery (SRS)  is  useful in treating small AVM, especially when the  nidus of the AVM is compact.  However, it does not offer protection from AVM rupture during the latent period before the obliteration of the malformation. This study addressed the risk of AVM rupture after SRS, cyberknife,  treatment.

Method: We reviewed all  patients with AVM treated with SRS from July 2010 to September 2013 in a single institution . We evaluated the efficacy of the treatment, the complication rate, symptom reduction, and bleeding/re-bleeding rate.

Results: The study population included 13 patients; Eleven of these patients had prior bleed. The median age was 41.5 yeras (11-67). The median Spetzler-Martin grade was 3 (2-4). The median prescription dose was 16.5 Gy(16-18.5).The median target volume was 9.0 cm3 (3.50-18.26). All but two are treated with a single session; two patients were treated with volume staging. The median follow up was 3.7 years (mean 3.3+/- 1.5).  Six of the treated AVM obliterated after single treatment (50%) including the one that had volume staging. One AVM was partially obliterated with a small residual nidus. The median time to response was 16 months (range 9-29) The median time to obliteration was 45 (14-52).  Two of the treated AVM ruptured at 6 and 13 months after treatment, both of these events were rebleed.  We estimated the odds ratio of symptom reduction is 0.055 (95%CI 0.0029-1.050, p=0.026) and a Risk ratio of 0.28 (95% CI 0.077-1.010, p=0.036) in favor of bleed reduction,

Conclusion: This a small study lacking the statistical power. But, it underscores the usefulness of SRS as an effective therapeutic option, and the peril of catastrophic  bleed.  Even then, the annual  bleeding rate in the latent period after SRS  is estimated to be 5.2% /yr, and the rebleed rate of 8.6%/yr. These are very compatible to those reported in the literature.

Objective: Arteriovenous malformations (AVM) initially treated by stereotactic radiosurgery (SRS) can be further treated with repeat SRS, embolization, and open surgery. Whether a minimally invasive SRS-retreatment strategy yields the safest long-term results has yet to be determined. We investigate how the hemorrhage rate following repeat CyberKnife SRS compares to that of other treatment strategies.

Methods: A single-institution retrospective review was conducted over the period spanning February 2002 to August 2013, during which we identified 114 patients who underwent primary treatment for AVM, among whom 13 received repeat SRS and four received an alternative adjuvant therapy (embolization or open surgery). Patients undergoing repeat SRS were treated with a median single fraction at a prescription dose of 19 Gy and a median max dose of 24.7 Gy.                       

Results: Patients were followed for a median 20.6 months after treatment of lesions measuring a median 2.9 cm (0.8 – 8.0 cm) and a median Spetzler-Martin grade 3 (1-5). There were 42.4% who had previously experienced hemorrhage. Prior to the repeat SRS session, half (50%) of the patients had undergone embolization treatment and 21% had undergone open surgery. At first follow-up, 7.8% were with radiographic evidence of obliteration and 33% with partial obliteration. Among those with repeat SRS and no additional intervention, 30.7% (4/13) and 2% (2/101) experienced subsequent hemorrhage (p = 0.002), respectively, with repeat SRS being the only significant predictor in our cohort. Repeat SRS did not predict obliteration.

Conclusions: Our experience with repeat SRS for incompletely obliterated AVM after primary SRS suggests there are subpopulations that may exhibit an increased rate of hemorrhage. Meanwhile, this elevated hemorrhage rate is on par with the already high overall hemorrhage rate from prior to retreatment, perhaps indicating a continuation of AVM-natural history. Thus, ongoing studies with larger cohorts are needed to identify the individual risk factors or optimal sequence of treatment modalities that will lead to safe obliteration of these complex lesions.

Objectives. The aim of this study is to analyze the dosimetric characteristics which are relevant to clinical outcome in treatment of large and giant cerebral arteriovenous malformations (AVMs) using single (SRS) and hypofraction (HFRS) stereotactic radiosurgery.

Methods. Since December 2005 to April 2015, a total of 17 patients with large (10-15cm³, n=9) and giant (>15 cm³, n=8) AVMs were included in this study. For large cerebral AVMs with median target volume of 13.3 cm³, SRS was perfomed and HFRS was performed for giant cerebral AVMs with median volume of 37.0 cm³. Clinical features including demographic features, type of nidus, angiographic features, obliteration rate, complication rate and radiosurgery-based modified AVM score were comparatively evaluated with dosimetric features, GTV and BED. The 12 patients of 17 patients were followed more than 12 months. Median follow-up period was 36 months (range, 15-104 months).

Results. Complete obliteration was achieved in 3 cases with SRS and 3 cases with HFRS. Mean modified AVM score was 1.86 (range 1.4-2.31) in SRS group and 4.73 (range 2.49-8.32) in HFRS group. Obliteration rate showed significantly proportional distribution according to the prescription dose (BED>126), and showed inverse distribution according to the modified AVM score (>3.0). Complications are one radiation necrosis with cyst formation,

Conclusions. Although the optimal therapeutic strategies are still remaining great challenge to achieve successful obliteration with low complication, multi-modal approaches with volume staged SRS was expecting to overcome the obstacle of the intractable large cerebral AVM.