Background:

Advanced age has been shown to be a factor predicting poor survival in patients with brain metastases (BM). There have been only a few studies focusing on stereotactic radiosurgery (SRS) for elderly BM patients. The present study aimed to investigate the efficacy and limitations of SRS for very elderly BM patients.

Methods:

This was a retrospective observational study analyzing 106 consecutive patients (69 males / 37 females) age 80 years and older who received upfront Gamma Knife SRS for BM between January 2009 and October2015. The median age was 84 years, and the median Karnofsky performance status (KPS) was 70. Fifty-two patients had a solitary BM, and others multiple BM. The median cumulative tumor volume was 3.9 mL and the median dose prescribed was 20 Gy. Overall survival (OS), neurological death rates and distant and local intracranial tumor control rates were analyzed.

Results:

No patients were lost to follow-up. Six-month and 12-month OS rates were 54% and 32%, respectively. The median OS time was 7.1 months. Competing risks analysis showed that 6-month and 12-month neurological death rates were 8% and 11%, respectively. In total, 245 / 311 tumors (79%) in 82 patients (77%) with sufficient radiological follow-up data were evaluated. Six-month and 12-month distant BM recurrence rates (per patient) after SRS were 17% and 25%, respectively. Six-month and 12-month rates of local tumor control (per lesion) were 94% and 89%, respectively. Repeat SRS, salvage WBRT and surgical resection were subsequently required in 25, 4 and 1 patient, respectively. Proportional hazard regression analysis showed that KPS ≥ 70 (HR: 0.444, P < .001), controlled primary disease/no extracranial metastases (HR: 0.361, P < .001) and female sex (HR: 0.569, P = 0.028) were independent factors predicting better OS. Similarly, tumor volume (> 2 mL) was the only factor predicting a higher rate of local control failure (HR: 12.8, P = 0.003).

Conclusions:

The present study suggested an upfront SRS strategy to offer a feasible and effective treatment option for very elderly patients with limited BM. In the majority of patients, neurological death could be delayed or even prevented. 

Purpose: Recently, with the aging of the population, an increasing number of elderly brain metastasis (BM) patients have been treated with stereotactic radiosurgery (SRS).

Methods: For this IRB-approved, retrospective cohort study, we used our prospectively accumulated database including 3102 consecutive patients undergoing SRS for BMs during the 1998-2015 period. Among these 3102 patients, 1684 were over age 65 years. The patient numbers for the 65-69, 70-74, 75-79, 80-84 and over 85 (max. 96) years of age groups were 557, 524, 382, 166 and 52, respectively.

Results: There was a significant increase in the proportion of elderly patient undergoing SRS in 2006 or earlier (51.1%) versus in 2007 or later (58.0%, p=0.0001). Median survival times (MSTs, months) of the aforementioned age groups were 7.4, 8.1, 7.2, 5.9 and 4.1, respectively (stratified p<0.0001). Although MST differences between each pair of neighboring age groups failed to reach statistical significance, there was a relatively large MST difference between patients under 79 and those over 80 years of age (HR; 1.439 [95% CI: 1.242-1.657], p<0.0001). Furthermore, MSTs of the age groups 65-69, 70-74, and 75-79 years did not differ significantly from the MST of the under-65-year group (8.4 months). Time-to-event outcome analyses showed that these age groups did not differ significantly in cumulative incidences of neurological death (p=0.47), neurological deterioration (0.42), SRS-related complications (0.42) or local recurrence (0.47).

Conclusions: Our results suggest that patients under 79 years of age are not poor candidates for SRS as compared to those over 80 years old.

Background: Leptomeningeal disease (LMD) presents symptomatically in approximately 5% of patients with metastatic brain cancer. The presence of LMD is conventionally viewed as an indication for whole brain radiation therapy (WBRT) and not suitable for stereotactic radiosurgery (SRS). The purpose of the study was to evaluate the local control rate and overall survival of patients who underwent SRS to focal LMD.

Methods: Thirty-two patients with brain metastases and LMD were identified in our prospective Gamma Knife radiosurgery database, from a total of 465 patients that underwent SRS between 2013 and 2015. For 16 patients, focal LMD was targeted with SRS. The median imaging follow-up time was 7 months.  The median volume of LMD was 372 mm³ and the median margin dose was 16 Gy. Five patients had undergone prior WBRT. Histology included non-small cell lung cancer (8), breast cancer (5), melanoma (1), gastrointestinal cancer (1) and ovarian cancer (1).

Results: Follow-up MR imaging was available for 14 patients.  For 13 of the 14 patients, LMD was stable (35.7%) or partially regressed (57.1%) at follow-up. Only one patient had progression of LMD associated with hemorrhage 5 months after SRS. Seven patients developed distant LMD at a median time of 7 months. The median actuarial overall survival from SRS for LMD was 10.0 months. The 6-month and 1-year actuarial overall survival was 60% and 26% respectively. Six patients underwent WBRT after SRS for LMD, at a median time of 6 months, with an overall survival of 3.5 months after WBRT.

Conclusion: Focal leptomeningeal metastatic disease may be treated successfully with radiosurgery permitting delay or avoidance of WBRT in some patients.

Objective:  Spinal metastases recurrent after conventional palliative radiotherapy have historically been difficult to manage due to concerns of spinal cord toxicity in the retreatment setting.  Spine SBRT, also known as stereotactic radiosurgery, is emerging as an effective and safe means of delivering ablative doses to these recurrent tumors. To determine the clinical efficacy and safety of spine stereotactic body radiation therapy (SBRT), specific to previously irradiated spinal metastases, a systematic review of literature was performed.

Methods:  A systematic literature review was conducted specific to SBRT to the spine using Medline, Embase, Cochrane Evidence Based Medicine Database, National Guideline Clearinghouse and CMAinfobase with further bibliographic review of appropriate articles.

            Research questions:

1.      Is retreatment spine SBRT efficacious with respect to local control and symptom control?

2.      Is retreatment spine SBRT safe?

Results: Initial literature search retrieved 2263 articles; 160 were potentially relevant, 105 selected for in-depth review and 9 studies met all inclusion criteria for analysis. All studies were single institution series: 4 retrospective, 3 retrospective series of prospective databases, 1 prospective, and one phase I/II prospective study (low or very-low quality data). The results indicate that spine SBRT is effective with a median 1-year local control rate of 76% (range, 66-90%).  Improvement in patient pain scores post-SBRT were observed to range from 65-81%. Treatment delivery was safe, with a crude rate of vertebral body fracture of 12% (range, 0-22%) and radiation myelopathy of 1.2%.   

Conclusion:  This systematic literature review suggests that SBRT to previously irradiated spinal metastases is safe and effective with respect to both local control and pain relief. Although the evidence is limited to low quality data, SBRT can be a recommended treatment option for re-irradiation.

Background: Stereotactic radiosurgery (SRS) for a few brain metastases (BM) is acknowledged as the optimal method for sparing critical structures, such as the hippocampus.  However, in the setting of 4 or more BM, the ability of SRS to spare the hippocampus and, thus, reduce neurocognitive deficits has been questioned.  This study reports hippocampal dose from single-fraction, multi-target SRS for 4-10 BM and assesses the feasibility of hippocampal-sparing SRS via plan reoptimization.

Materials/Methods:  Patients with four to ten brain metastases receiving single-isocenter, multi-target single-fraction SRS were identified in this IRB-approved study. Hippocampi were contoured using the RTOG 0933 atlas. RTOG 0933 dose constraints were converted to a biologically effective dose using an alpha/beta of 2 (D₁₀₀ 421 cGy, D_max 665 cGy). Number of metastases, total target volume, prescribed dose, and distance of nearest metastasis [dmin] were analyzed as risk factors for exceeding hippocampal dose constraints. If hippocampal dose exceeded constraints, the SRS plan was reoptimized. Key dosimetric parameters were compared between original and reoptimized plans. To determine if a single target would exceed dose constraints, all targets but the metastasis closest to the hippocampi were removed from the plan and dosimetry was compared.

Results: 40 plans were identified. 15 hippocampi (19%) exceeded constraints in 12 SRS plans. Hippocampal sparing was achieved in 10 of 12 replanned cases (83%). Risk factors associated with exceeding hippocampal constraints were decreasing dmin (24.0 v 8.0mm, p=0.002; OR 1.14, 95% CI 1.04-1.26) and higher total target volume (5.46 cm³ vs 1.98 cm³, p=0.03, OR 1.14, 95% CI 1.00-1.32). There was no difference in exceeding constraints for 4-5 vs 6-10 metastases (27% v. 21%, p=0.409) or prescribed dose (18 Gy, p=0.58). For reoptimized plans, there were no significant differences in PTV coverage (99.6% vs 99.0%, p=0.17) or conformity index (2.03 vs 2.09, p=0.78). Six (50%) plans exceeded dose constraints with a single target.

Conclusion:  A substantial proportion of hippocampi may receive a relatively high radiation dose from SRS when treating 4-10 BM. Decreased distance of the closest metastasis to the hippocampus and higher total target volume are associated with exceeding hippocampal constraints. Reoptimizing these plans spares dose to the hippocampi and still yields acceptable dosimetric characteristics. Prospective evaluation of the impact of hippocampal dose on neurocognition in the setting of SRS to 4 or more BM would be valuable.

Introduction: Recent reports suggest that immune checkpoint therapy (ICT) combined with stereotactic radiosurgery (SRS) have been associated with greater lesion regression of melanoma brain metastases and decreased local failure. When given concurrently (within 4 weeks), combined SRS and ICT may result in improved freedom from additional new brain metastases.

Objective: To investigate the feasibility and efficacy of ICT administered with SRS in patients with brain metastases of non small cell lung cancer (NSCLC) and evaluate if synergistic effect observed for melanoma is also identified in these patients. 

Methods: 101 individual patients with brain metastases from lung cancer, being treated 185 times by SRS between March 2014 and November 2016, and followed in the prospective cohort of the brain metastases clinic at the CHUV where included in this analysis. Systemic treatment use within 6 weeks of SRS was noted. The prescription was 20 Gy in single fraction or 33 Gy in 3 fractions, according to the volume of PTV.

Results: Among the 185 SRS performed, 66 corresponded to single metastases, 48 to 2 - 4 metastases, 27 to 5 - 10, and 10 to > 10 in every single event. The median volume for the PTV was 1,68 ml, the average volume 10,17 ml. Fifty-six patients received only one treatment, 29 of them two and 16 three or more (max. 5).

1-year OS was 56,11% and 72,06% for patients in the chemotherapy and immunotherapy groups, respectively (p=0,26); 1-year brain-DFS was 32,16% (chemotherapy) and 33,84% (immunotherapy; p=0,95). Toxicity was minimal, with 3% grade 2 and no adverse event > grade 2. 

Conclusions: SRS in combination with ICT is feasible, without an increase in toxicity. Even if there was a trend for improvement of OS between patients treated with immunotherapy, the brain-DFS was identical in both groups, suggesting that there was no synergistic effect between ICT and SRS in patients with brain metastases of NSCLC.

Objectives: Main purposes of this study was to analyze treatment results for a group of glioblastoma patients treated in our center on Leksell Gamma Knife (LGK) over past more than twenty years.

Methods and materials: Altogether 126 patients were treated for glioblastoma during 1992-2014 in our institution and analyzed in this study. There were 69 (55%) male and 57 (45%) female with median age of 56 years (range 17-80 years). From this group 123 (98%) underwent surgery, 126 (100%) radiotherapy and 116 (92%) chemotherapy before the LGK radiosurgery.

Results: Median Karnofsky score before the LGK radiosurgery for patients in this group was 90 (range 50-100), median time from diagnosis of glioblastoma to the LGK radiosurgery was 12 months (range 1- 96 months). Median tumor volume was 3.75 cm³ (range 0.04 – 37.10 cm³) and location: 45 (36%) temporal, 31 (25%) frontal, 21 (17%) parietal, 12 (9%) occipital and 17 (13%) other location. All patients in this study were treated by the LGK radiosurgery performed in one single fraction with median minimal tumor dose of 12 Gy (range 10 – 25 Gy) on median 50 % (range 40 – 86 %) isodose line. Two and more LGK radiosurgeries were performed in 19 (15 %) cases. Median patients’ survival since glioblastoma diagnosis was 20 months (range 6 – 237 months) and median patients’ survival after the LGK radiosurgery was 7 months (range 1 – 223 months).

Conclusion: Treatment on the LGK was well tolerated by all patients. It appears a good final therapeutic option for glioblastoma after previous surgery, radiotherapy and chemotherapy for a small residual volume.

OBJECTIVES: We determined the maximum tolerated dose (MTD) of 5-fraction stereotactic radiosurgery (SRS) with concurrent and adjuvant temozolomide (TMZ) in glioblastoma (GBM).

METHODS AND MATERIALS: Patients received 5 consecutive days of SRS in a 3+3 design with 25, 30, 35 or 40 Gy targeting the GTV with a 5 mm CTV and 0 mm PTV (maximum size 150 cm³).  A dose limiting toxicity (DLT) was Grade 3-5 CNS toxicity within 30 days, with life-long assessment for late SRS-related adverse radiation effect (ARE). Secondary endpoints included progression free survival (PFS), overall survival (OS), and health-related quality of life (HRQOL). Changes from baseline for 9 HRQOL measures were calculated at baseline and in follow-up. 

RESULTS: From 2010 to 2015, 30 patients were enrolled.  Median age was 66 years (range 51-86) with median KPS of 80 (range 50-100).  Median GTV was 26.8 cm³ (range 3.8–81.0) with a PTV of 60.2 cm³ (range 14.7–137.3). DLTs occurred in 2 patients and ARE in 12 patients: 5 developed pseudoprogression at a median of 3.0 months and 7 developed grade 1/2 radionecrosis (RN) at a median of 8.0 months after SRS. RN was not associated with dose (p=0.8) or PTV size (p=0.09). With a median follow-up of 14.0 months, median OS and PFS were 14.9 (95%CI 10.9-21.4; range 1.7 - 62.4) and 8.2 (95%CI 4.6-10.5; range 1.7-55.8) months, respectively. RN was associated with improved median survival (33.0 vs. 11.3 months; p=0.05). HRQOL assessment compliance was 76% at 12 months. Communication deficit worsened over time (-1.7 points/month, p=0.008), with no significant changes in the other 8 HRQOL scales.  While RN was not associated with a significant decline in any HRQOL scale, disease progression was associated with communication deficit (p=0.01). 

CONCLUSION: Five-fraction SRS with 5mm margins with TMZ for newly diagnosed GBM has similar OS and HRQOL compared to conventional fractionation. Patients with radiation necrosis, the primary ‘toxicity’, had longer survival with no decline in HRQOL.

Treatment of recurrent lhight grade gliomas includes a constellation of options including surgery, radiation therapy, cytotoxic chemotherapy, and white therapies. In the case of reirradiation, it is necessary to have clear the volume to be irradiated, the dose received previously so it is necessary to use a technique that allows to achieve a therapeutic dose and at the same time protect the surrounding brain tissue, the above only Is possible through techniques such as radiosurgery and hypofractionation. Which by means of stereotactic localization allows an accuracy of the target to be treated, presents a high dose gradient and protects normal brain tissue.

Patients were analyzed from May 1992 to November 2014 diagnosed with high-grade, recurrent gliomas (WHO varieties III and IV). A total of 43 patients obtained in this study were evaluated with at least 1 year of follow-up from the radiosurgery treatment, in whom radiosurgery was used as treatment for relapse.

There was symptomatic improvement in 44% and stabilization in 35.2%, the progression was observed in 14 of 43 patients, radiologically the reduction was obtained in 22.66% and stabilization in 40.71%, radionecrosis in 20%.

In conclusion, the rescue of patients with recurrent high-grade gliomas should be taken into account for the use of radiosurgery, however, further studies are needed to define possible subgroups of greater benefit

Introduction: Glioblastoma (GBM) is the most malignant form of astrocytoma with a dismal outcome.  Despite advances in treatment, the average survival in patients with recurrent GBM remains 6-10 months. In this study we evaluated the role of stereotactic radiosurgery (SRS) in the management of recurrent GBM.

Materials and Methods:  A retrospective review of the Cleveland Clinic brain tumor database (1997-2016) was performed following IRB approval. Overall survival (OS) and progression free survival (PFS) from salvage SRS were the primary and secondary end points, respectively. Molecular analysis was performed using standard techniques. Response to SRS was assessed on T1with contrast and T2-FLAIR MRI images. The log rank test and Cox proportional hazard models were used for analysis.

Results:  Fifty-three patients with 75 lesions underwent salvage SRS. Eleven patients (21%) had multiple lesions and one patient (1.8%) underwent treatment in three stages for a single lesion (12 Gy each). The median age at diagnosis and SRS was 58.9 years and 60.5 years, respectively. Overall, 68% (n=36) of patients were male and the majority (70%, n=34) had good performance status (KPS>80). Most patients (74%, 56/75) had either gross total tumor resection (57%, 43/75) or subtotal resection (17%, 13/75) and all surgery was followed by chemoradiotherapy prior to salvage SRS. Most lesions were treated with at least two additional procedures, following index procedure prior to salvage SRS (56%, 42/75). Majority (85%, n=11/13) of patients had wild type IDH, 53% (n=10/19) had extensive Ki-67 staining (>30%) and 71% (n=10/14) were MGMT unmethylated. Median OS post SRS was estimated to be 11.0 months (95% C.I. 7.1-12.2) and median per-lesion PFS was 4.1 months (95% C.I. 2.6-4.4). Median tumor diameter and volume were 2.55 cm and 3.80 cm³, respectively. Median prescription dose was 18 Gy (12-24 Gy) and homogeneity index was 1.90 (1.11-2.02). KPS>80 was independently associated with longer OS (HR: 4.42, CI: 1.89-10.30, p=0.0006). Small tumor size (<3 cm) was positively correlated with PFS (p=0.03) whereas small tumor volume (<15 cc) was independently associated with both OS and PFS (p=0.02 and 0.04 respectively). Higher homogeneity index (>1.75) was independently associated with longer PFS (HR: 2.38, CI: 1.02-5.56, p=0.04).

Conclusions:  Good performance patients with smaller tumor volumes and treated at higher homogeneity index were associated with longer OS/PFS despite multiple prior treatmentsfor recurrent GBM. SRS for recurrent GBM is reasonable salvage treatment option for these patients.

Objectives: The study aimed to analyze radiological responses of recurrent high-grade glial tumors treated with hypofractionated stereotactic radiotherapy (HSR) and to estimate patient overall survival (OS) and progression-free survival (PFS) after the treatment.

Methods: We retrospectively analyzed 104 patients managed with HSR for recurrent high-grade gliomas between 2011 and 2016. All patients had experienced progression after a complex treatment with surgery and chemo-radiotherapy. 49% of patients were diagnosed with glioblastoma and 51% with grade III glioma. The indications for HSR were locally progressing tumors in 77% or new distant tumors in 23% of patients. Tumor recurrence was determined and localized with MRI matched against C¹¹-methionine PET/CT. The mean target volume was 36.4 cm³, ranging from 2.8 to 127.2 cm³. The prescription dose was volume-dependent from 18 to 30 Gy delivered in three fractions with interfraction intervals from 2 to 10 days. HSR was performed with Cyber Knife (Accuray, Sunnyvale, CA, USA) and linear accelerator TrueBeam STX (Varian Medical Systems, Palo Alto, CA). After treatment, the patients underwent follow-up imaging examination (contrast-enhanced MRI and/or PET/CT with C¹¹-methionine) every 2 months. Post-treatment MR and PET images were fused with initial pre-treatment images and volumetrically analyzed with Gamma Plan software (Elekta AB, Stockholm, Sweden). Radiological response was evaluated using RANO criteria. PFS and OS were calculated using the Kaplan-Meier method. The median follow-up time was 11 months.

Results: We found that patients with recurrent glioblastoma and grade III glial tumors demonstrated a similar pattern of radiological response to HSR: none of the patients achieved a complete response, 9% of patients with glioblastoma and 11% with grade III glioma showed a partial response, 32% and 42% stable disease, 23% and 19% progressive disease, 36% and 28% pseudo-progression. Median progression-free survival from the date of HSR was 8.6 months for patients with glioblastoma and 13.6 months for patients with grade III glioma (p=0.001). Median overall survival after HSR was 13.3 months for glioblastoma and 32.8 months for grade III glioma patients (p=0.011). The invasion of the tumor into deep brain structures appeared to be a crucial factor affecting OS (HR=6.18, p<0.001).

Conclusion: Recurrence often means a fatal outcome for the patient, as standard treatments are no longer effective. This is why HSR may be considered an effective salvage treatment for patients with recurrent high-grade gliomas. A precise determination of recurrent active parts of the tumor is essential for the success of this approach.

Study objective: retrospective analysis of results of CyberKnife radiosurgery and hypofractionated irradiation for small to medium-size progression of supratentorial malignant glioma in 61 selected patients treated in Burdenko Neurosurgical Institute in 2009-2016. 

Patients and methods: 32 men and 29 women was included, mean age was 47,2 years. 39 patients had primary glioblastoma, 5 - secondary glioblastoma, 6 - anaplastic astrocytoma, 7 - anaplastic oligoastrocytoma and 4 had anaplastic oligodendroglioma. 60 patient underwent tumor removal and 1 stereotactic biopsy. 57 patients received postoperative radiotherapy with 58-60 Gy in 29-33 fractions and other 4 had shorter courses (33-45 Gy in 12-18 fractions). Patients with glioblastoma and anaplastic astrocytoma received temozolomide 75 mg/m² during radiotherapy. After completion of radiotherapy 54 patients received adjuvant chemotherapy (35 had temozolomide-based regimen and 19 - other regimens). Mean time from completion of radiotherapy to first progression was 8,2 months in glioblastoma group (44 patients) and 16,3 months in anaplastic glioma group (17 patients).

First progression as single growing lesion in primary tumor region (local type of "monofocal" progression) was observed 31 of 44 (70%) glioblastoma and 14 of 17 (82%) anaplastic glioma patients. 6 glioblastoma and non of anaplastic glioma patients had distant type of monofocal progression (single new distant lesion with absence of progression in primary tumor region). Other 12 patients had "multifocal" progression (2 had several local growing foci, 2 had several distant foci and 8 had at least 1 growing local lesion and 1 new distant lesion).

Lesions with volume less than 11 cm³ was treated with single median dose of 20 Gy, bigger lesions (up to 58 cm³, median volume - 12,7 cm³) were irradiated with 3 to 7 fractions up to total dose of 21-39,5 Gy (every day or every other day). Mean follow-up was 13,3 months after CyberKnife salvage irradiation.

Results: mean time from salvage irradiation to second progression was 8,2 months in glioblastoma group and 17,2 months in anaplastic glioma group; overall survival after salvage irradiation was 16,5 and 31 month respectively. In 6 of 92 irradiated lesions (6,5%) clinically significant adverse radiation effect developed, all were treated successfully with bevacizumab. In primary glioblastoma group (39 patients) addition of more than 3 infusions of 400 mg bevacizumab to CyberKnife treatment was statistically significant associated with better overall survival (p=0.01).

Conclusion: CyberKnife irradiation with bevacizumab is an effective option for monofocal and multifocal forms of supratentorial glioblastoma progression.

Objectives:

The aim of this systematic review is to provide an objective summary of the published literature pertaining to the use of stereotactic body radiation therapy (SBRT) specific to previously untreated spinal metastases.

Methods:

A systematic review of the literature using PRISMA guidelines was performed up to March of 2015 from MEDLINE, PubMed, Embase, and the Cochrane Library.  The search strategy was limited to publications in the English language.

Results:

            A total of fourteen full-text articles were included in the analysis.  All studies were retrospective, except for two prospective studies.  A total of 1,024 treated spinal lesions were analyzed. The median follow-up ranged from 9-49 months. A range of dose-fractionation schemes were utilized, with the most common ones being 16-24Gy/1fraction (fx), 24Gy/2fx, 24-27Gy/3fx, and 30-35Gy/5fx. For studies reporting crude results regarding in-field local tumor control, 346 out of 407 lesions (85%) remained controlled.  For studies reporting actuarial values, the weighted average result demonstrated a 90% one-year local control rate.  Only three studies reported data on complete pain response, and the weighted average of these results yielded a complete pain response rate of 54%. The most common toxicity was new or progressing vertebral compression fracture, which was observed in 9.4% of cases, and two cases (0.2%) of myelopathy were reported.

Conclusion:

            There is a paucity of prospective data specific to SBRT in patients with spinal metastases not otherwise irradiated. This systematic review demonstrates that SBRT is associated with favorable rates of local control of approximately 90% at one year, and rates of complete pain response of approximately 50%, with low rates of serious adverse events. Practice guidelines are summarized based on these data and ISRS consensus.

Purpose/Objective(s):

Surgery followed by conventional external beam radiotherapy (cEBRT) remains the standard of care in treating patients with metastatic epidural spinal cord compression. However, only one series has evaluated the impact of standard fractionated cEBRT vs. stereotactic radiosurgery (SRS) in the post-surgical setting. SRS may lead to suboptimal local control (LC) due to incomplete tumor bed coverage while there is evidence that cEBRT may impair healing and hardware integrity. The purpose of this study was to assess oncologic control, wound complications, and mechanical failure after SRS vs. cEBRT.

Materials/Methods:

An IRB approved retrospective review of patients treated with surgery followed by cEBRT or SRS for spinal metastases was performed. Local failure was defined as failure within the originally involved vertebral level(s). Surgical bed failure was any oncologic failure within the operative bed. Wound complications included post radiation infection or hematoma. Mechanical failure was defined as clinically significant hardware lucency or breakage, dynamic motion on imaging, loosening of bone-graft interface, or new pain not explained by oncologic progression.

Results:

63 patients were treated from 2006-2014: 37 received post-operative SRS; 26 received post-operative cEBRT. Instrumentation was performed in all 37 SRS patients and 18/26 (69%) cEBRT patients. Median follow-up post-SRS and cEBRT was 7.4 (range 1-39) and 4.4 (range 0-73) months, respectively (p=0.29). Karnofsky Performance Status ≥70 post-SRS and cEBRT was 86% vs 44%, respectively (p<0.01). The most common dose schedules were 16 Gy single fraction (SRS) and 30 Gy in 10 fractions (cEBRT). 1-year local failure was 9% vs. 30% (p=0.026) for SRS and cEBRT respectively, with 6/9 cEBRT patients demonstrating multilevel failure within and beyond the originally involved vertebral bodies; surgical bed failure at 1 year was 26.8% versus 30.0% (p=0.61). No mechanical failures were observed. Wound complications occurred in 2 SRS patients and 1 cEBRT patient; projected 1-year cumulative incidence rate of 2.8% and 6.3% respectively (p=0.91).

Conclusion:  

Post-operative SRS provides excellent LC in patients who have undergone surgery for spinal metastases. The 9% LF is less than that for SRS alone for patients with epidural disease, suggesting a benefit to post-operative SRS in carefully selected patients. The absence of significant complications in either group suggests that both SRS and cEBRT are safe approaches in this population.

Objectives: Currently, there is no consensus fractionation scheme for spine SBRT. We report mature outcomes for a cohort of patients with no prior radiation (de novo) treated with 24 Gy in 2 daily fractions, which represents an emerging Canadian standard.

Methods: The cohort consisted of 279 de novo spinal metastases in 145 consecutive patients treated with 24 Gy in 2 SBRT fractions, between 2009 and 2015, identified from a prospective database. All vertebral segments were treated with an institutionally standardized linac-based approach using cone-beam CT image guidance and six degrees-of-freedom online setup correction. The endpoints were overall survival (OS), local control (LC), and the rate of vertebral compression fractures (VCF). OS rates were obtained using Kaplan-Meier methods and cumulative incidences of LC and VCF were obtained from competing risk analysis using death as a competing risk event. Evaluation of tumor control was based on serial spine magnetic resonance imaging (MRI) as per the SPIne response assessment in Neuro-Oncology (SPINO) criteria recommendations.

Results: The median follow-up was 17.0 months (range, 0.1–71.6 months). The 1-year and 2-year OS rates were 73.1% and 60.7%, respectively. Presence of epidural disease (p < 0.0001), lung (p = 0.0415) and renal cell (p < 0.0001) primary histologies and diffuse spinal metastatic disease as opposed to oligometastatic disease (p = 0.0034) were significant prognostic factors. The 1-year and 2-year LC rates were 90.3% and 82.4%, respectively, and the median time to local failure (LF) was 9.2 month (range, 0.4–31.3 months). Only the presence of epidural disease predicted for LF (p < 0.0001). The cumulative risk of VCF at 1 and 2 years were 8.5% and 13.8%, respectively. Lytic (p = 0.0143) or mixed lytic/blastic (p = 0.0214) lesions, spinal misalignment (p = 0.0121), and the dose to 90% of the planning target volume (PTVD90) (p = 0.0085) were significant predictors of VCF.

Conclusion: 24Gy in 2 daily fractions is safe and effective in achieving high tumor control rates for de novo spinal metastases. This fractionation scheme is currently the standard SBRT arm on an ongoing Phase 3 randomized Canadian national trial (CCTG-SC 24) comparing it to a conventional radiation dose of 20 Gy delivered in 5 daily fractions.

Introduction:

Lung cancer is a leading cause of oncologic death in our country and commonly metastasize to spine which often resulted in spinal cord compression. Although recent advance in systemic therapeutic modalities including chemo- and targeted therapy, the prognosis of these metastatic disease is still poor and overall survival length is often less than 1 year. Recent radiosurgical treatment arm often provide better pain palliation and effectively durable local tumor control. Depending upon the patients’ clinical status, therapeutic strategies should be tailored to improve quality of life for these patients with considering prognostic prediction and overall survival.

Method:

Fifty-six patients with metastatic spinal cord compression from NSCLCa were treated with stereotactic radiosurgery between 2005 and 2014. Male to female ratio was 32 to 25. Median age was 66 year-old (range of 43 to 82 year). Pretreatment performance scale over 80 versus less than 70 was 25 versus 29. Cox regression model was utilized to analyze prognostic factors affecting overall survival including sex, age, KPS, No. of bone metastasis, No. of visceral metastasis, grade of metastatic epidural compression, history of prior radiation treatment and interval between initial diagnosis and spinal metastasis. In addition, based on the individual preradiosurgical scoring status and survival length, concordance rate were calculated by applying the three different prognostic systems such as Revised Tokuhashi, Tomita, and Van der Linden. By comparing the factors and outcome prediction of each system, useful relevant prognostic factors were evaluated.  

Results:

Average survival length was 8.5 months (range of 1~32 months). Concordance rates of predicting survival length with applying Tokuhashi, Tomita and Van der Linden scoring systems were 73%, 50% and 65% independently. Systemic oncologic statue and performance scale are the important parameters in evaluating tools to anticipate overall survival and outcome. Time to diagnosis of metastatic disease and primary lung cancer also showed significant predicting factor affecting survival length (< 2mos vs > 6mos, HR: 2.115, 95% CI 1.17~3.81, p=0.127).

Conclusions:

The significant prognostic factors associated with survival after spinal SRS for metastatic NSCLCa are including pretreatment performance scale, primary disease control, and time between first metastasis and performance status. Relevant outcome factors should be carefully considered and evaluated in order to determine optimal therapeutic strategies including stereotactic radiosurgery.

Radiosurgical epidural decompression has been demonstrated with in 80% with complete or significant reduction of epidural tumor volume, indicating that these patients can be treated with non-invasive radiosurgery. In order to help a better communication and unifrom decision-making of treatment, we also developed a dual grading system of spinal cord compression with graces 0 - V radiographic (anatomical), and grades a - e neurological (functional) grades. While open surgery is able to decompress the spinal cord immediately, radiosurgical spinal cord decompression occurs gradually and is suitable for patients with no neurological deficit. However, the current practice is to make such decision based on the MR imaging study rather than the patient’s neurological status, and controversity exists whether high grade spinal cord compression can be safely treated with radiosurgery. Practioners are indeed concerned about the presence of spnal cord compression. Therefore, the current study was performed to demonstrate the role of radiosurgery in patients with high-grade spinal cord compression on MRI imaging with grade IV (significand compression and displacement of spinal cord, T2-weighted CSF signal is still visible, aka partial block) and V (no visible t-2 weighted CSF signal, aka complete block on conventional myelogram). Total 33 patients with 35 lesions with radiographic grades IV-V were inrolled, with minimal or no neurological deficit (grade a-b) except one patient with neurological grade c (nonambulatory). The patients were treated with single dose spine radiosurgery 18-20 Gy, prescribed to the tumor margin. Spinal cord contraint was 10 Gy to the 10% cord volume defined 6 mm above and below the epidural tumor target. Median followup time of 6.4 months with MR imaging study and neurological examination every 2 months. Radiosurgery resulted in 75% spinal cord decompression at 2 months MRI scan post-radiosurgery, and 70% neurological improvement. One patient with neurological grade c became fully ambulatory. Only 2 patients required open surgery due to neurological decline, and 2 patients due to intractable pain. There was no other complication. Progression-free survival was improved in patients who responded to the radiosurgery treatment. Neurological grade was the only prognostic factor. Radiographic high grade did not affect the outcome. The results strongly support the use of radiosurgery for high grade spinal cord compression with no or minimal neurological deficit.

Objectives:

Malignant melanoma is the most frequent intraocular tumor in adults. One day session LINAC based stereotactic radiosurgery (SRS) of intraocular malignant melanoma is a method of "conservative" attitude to treat uveal melanoma.

Experience with 3D printed model of the eye with intraocular tumor used by planning on linear accelerator LINAC based stereotactic radiosurgery.

Methods:

Retrospective clinic-based study - clinical findings of patients with posterior uveal melanoma (choroid) in stage T1/T3 who underwent stereotactic radiosurgery (SRS) at LINAC  in Slovakia. Patients were not randomized either to radical or to “conservative” procedure, but the treatment was determined exclusively on a case-by-case basis. Tumor stage, volume, maximum elevation, localization presence of secondary retinal detachment, general status, age, gender, the functional tests were taken into consideration. The stereotactic frame was fixed to the head and the sutures were tied to the stereotactic frame. We used model Clinac 600 C/D Varian (system Aria, planning system Corvus version 6.2 verification IMRT OmniPro) with 6 MeV X by rigid immobilization of the eye to the Leibinger frame. The patient underwent CT and MRI examination with the fixed eye to the frame. The stereotactic treatment planning after fusion of CT and MRI was optimized according to the critical structures (lens, optic nerve, also lens and optic nerve at the contralateral side, chiasm).

The best plan was applied for therapy at C LINAC accelerator. The planned therapeutic dose was 35.0 Gy by 99 % of DVH (dose volume histogram).

In the software for segmentation (3DSlicer) created virtual 3D model of eye globe with tumorous mass based on tissue density from CT and MRI data. Virtual model was then processed in the slicing software (Simplify3D) and printed on 3D printer using FDM (fused deposition modeling) technology. Material used for printing was polylactic acid.

Results:

In period 2001 - 2015 the group of 150 patients with uveal melanoma (139 choroidal melanoma, 11 ciliary body melanoma) was treated. The median tumor volume was 0.5 cm³ (0.2 - 1.6 cm³). The radiation dose was 35.0 Gy by 99 % of DVH (dose volume histogram). Since 2015 stereotactic planning scheme was optimized with help of 3D printed model of the patient’s eye with intraocular tumor.

Conclusion:

Our 3D printed model of eye with tumor was helpful in planning process to achieve the optimal scheme for irradiation which requires high accuracy of defining the targeted tumor mass and critical structures.

Objectives: To describe the general follow-up of patients and to analyze the evolution pattern of melanomas choroidal treated with radiosurgery.

Material and methods: A retrospective and descriptive analysis of 8 patients (4 men, 4 women, mean age of 58 years) diagnosed with choroidal melanoma who were treated with radiosurgery, from January 2015 to January 2017, was performed at the Radiosurgery Service of the Hospital Español, Mexico City. All patients were previously assessed with a complete ophthalmologic evaluation and with imaging studies to rule out metastases. A dedicated Novalis 600 N, 6 MV linear accelerator (BrainLab, Heimstetten, Germany) was used to impart doses ranging from 35 to 40 Gy (mean 35.75 Gy). Melanomas were located in the right eye in 87.5% of patients and the mean treated volume was 0.882 cc (0.202-1.819 cc). The main symptoms included a decreased visual acuity and quadrantanopsia. The mean follow-up was 12.8 months (4-22 months). Tumor size (area) was measured using ultrasound imaging during follow-up.

Results: After treatment, the percentage in tumor control, survival and metastasis was 87.5%, 100% and 0%, respectively. The TNM stage for choroidal melanomas of the patients was T2 and T2a. The evolution after radiosurgery was variable: 4 patients showed a progressive decline in tumor size, 3 patients had an initial tendency to increase tumor size followed by a significant reduction three months after treatment. Finally, one patient only had a 50% increase in tumor size and was considered as a disease progression. Visual acuity was reported to decrease 3 months after treatment. Complications: 50% developed retinopathy between 9 and 12 months, one patient had cataract/papillopathy and one patient showed no response to treatment. Follow-up of 7 functional patients continues.

Conclusions: Radiosurgery is a very minimal invasion alternative for the treatment of choroidal melanomas providing good tumor control and health-related quality of life. In our study some patients showed an initial increase in tumor size followed by a significant reduction. Thus, it is important to recognize tumor size changes after the radiosurgery to properly assess its efficacy. A larger number and longer follow-up periods of patients with choroidal melanomas treated with radiosurgery are required to better quantify the treatment success.

Aims: We report our clinical experience of a hypofractionated Cyberknife Radiosurgery schedule for uveal melanoma treatment.

Methods: Between April 2014 and March 2016 14 patients (pts), mean age 65 years (range 36 – 83 years) suffering from uveal melanoma (11 choroidal melanoma and 1 ciliary body melanoma) were treated at Cyberknife Center, Centro Diagnostico Italiano, Milan. All of the pts had received a diagnosis and referral from an ophthalmologist. Cyberknife radiosurgery was performed delivering a total dose of 54 - 60 Gy (mean 60 Gy) given in 3 or 4 fractions (mean 3) of 15 - 20 Gy (mean 20 Gy) prescribed to the 79 - 82% (mean 80%) isodose surface. All pts underwent orbit MRI with gadolinium for coregistration with the planning CT scans. The planning target volume (PTV) included the contrast-enhancing lesion on MRI plus a 2.5 mm margins in all directions. All pts were irradiated eyelids closed, using a contention with a thermoplastic mask. The mean PTV volume was 2037 mm³ (range 701.82 – 5792 mm³), mean tumor base measured ultrasonographically 11.36 mm (range 7-15 mm), mean thickness 4.79 mm (range 2.5 – 10 mm), with a mean distance of 5.25 mm (range 0 – 15 mm) from fovea and 5.55 mm (range 0 – 13 mm) from optic nerve.

Results: After a mean follow-up of 17 months (range 7 – 30) local control was achieved in 100% of pts. No patient underwent enucleation and none developed distant metastases (all pts underwent abdomen ultrasound and liver blood examination once every six months and chest CT once a year). We observed a reduction of 13% in median base and of 44% in median thickness that were respectively 10 mm (range 4.8 – 13 mm) and 2.45 mm (range 0.5 – 5 mm) at last follow-up. Visual acuity was reduced in 64 % of pts, while in the others no change was found. Four pts suffered of radiation maculopathy, associated in one case with atrophy and in three cases with cystoids macular edema. Moreover radiation-induced optic neuropathy and radiation vasculopathy occurred respectively in 3 and 4 cases. 7 pts developed choroidal ischemia and 3 retinal detachment. At the last follow-up none had corneal anomalies.


Conclusions: These initial results of our Cyberknife schedule are consistent with data in literature and show a safe, minimally invasive and well tolerated method for treating uveal melanoma. Further follow-up is necessary.

The objective of the present study was retrospective evaluation of results of Gamma Knife surgery (GKS) for uveal melanoma performed in 8 European Gamma Knife centers (Bucharest, Florence, Istanbul, Moscow, Prague, Saint Petersburg, Sheffield, and Zurich), which agreed to participate and provided required information. Study design presumed collection of the various data (62 investigated variables) on individual patients with creation of the integrated database for further statistical analysis. Primary end-points were overall survival, local tumor control, eye retention rate, and morbidity during follow-up. Secondary end-points were tumor response, preservation of visual function on the affected eye, and incidence of metastatic disease after GKS. 

In total 349 cases were collected. All patients were treated between July 2001 and October 2015. The stage of treated tumors corresponded to I, IIA, IIB, and IIIA in 26%, 43%, 23% and 7% of cases, respectively. Eye fixation was attained with retrobulbar anesthetic blocking in 65% of cases or suturing of rectus muscles in 24% of cases; in 11% of cases no eye fixation was done. Median marginal dose was 30 Gy (range, 25-30 Gy), median maximal dose was 50 Gy (range, 30-80 Gy), median maximal dose to the ipsilateral optic nerve was 8.7 Gy (range, 0.7 – 61.2 Gy). In 23 patients GKS was done before planned eye-preserving endoresection of the tumor. Follow-up information was available in 314 cases and median length of follow-up was 40 months (range, 1-132 months). 

Actuarial survival rates at 3 and 5 years after GKS were 91% and 89%, respectively. Crude tumor control rate was 98%. Complete response was noted in 19% of cases, partial response in 46% of cases. Crude eye-retention rate was 95.5%. In overall 13 enucleations were done owed to tumor progression (5 cases), complications (2 cases), or unknown reasons (6 cases). Complications were noted in 67% of patients, and exudative retinal detachment (15%), cataract (13%), neovascular glaucoma (11%) and retinopathy (11%) were the most common. The risk of loss of the useful vision on the affected eye during follow-up after GKS was 40%. Suturing of the extraocular muscles for eye fixation during irradiation and two-staged treatment (GKS followed by planned endoresection of the tumor) were associated with significantly lower risk of complications and visual loss on the affected eye. Distant metastases after treatment were disclosed in 10% of patients.

In conclusion, GKS seems effective treatment option for management of uveal melanoma at early stage of disease.

Objectives: Visual function preservation is a secondary endpoint of choroidal melanomas (CM) treatment. We aim to identify factors predictable of better visual acuity after radiotherapy treatment.

Methods: A total of 21 patients with unilateral CM were treated in LINAC based with stereotactic radiosurgery since 2014. Sixteen patients, with tumor height ≤8 mm and base ≤16 mm,  and median FU of 20 months were enrolled in this retrospective analysis. SRS was delivered to a dose of 50 Gy in five fractions. An eye monitoring system was applied for the acquisition of planning CT and treatment delivery. Standardized A- and B-scan echography and MRI of the eye were performed at baseline and during follow up. Toxicity was graded using the CTCAE v4.0. Structures at risk in the eye, including the delineation of the macula region, were analyzed in the setting of maximum dose received.

Results: Thirteen patients had severe visual loss at baseline and three had preserved (>20/40) visual acuity. Of these three patients, one remained with 20/25 visual acuity at last FU after 10 months, despite development of radiation induced grade 2 cataract (Macula Dmax 45,3Gy; Optic Nerve Dmax 24,7 Gy). One maintained good visual function (20/40) at last FU after 31 months with no treatment toxicity (Macula Dmax 59,7Gy; Optic Nerve Dmax 31,4 Gy). Another had pre treatment cataract and maintained good visual outcome (≥20/25) until 9 months, but developed blindness at 10 months (Macula Dmax 52,7 Gy; Optic Nerve Dmax 21,6 Gy). At 12 months, this patient underwent enucleation due a residual image on MRI. Histological specimen showed residual CM. Of the 13 patients who had severe visual loss pre treatment, four had improvement in visual acuity from 20/80 to 20/25 at 6 months (Macula Dmax 53,5 Gy; Optic Nerve Dmax 21,5 Gy), 20/60 to 20/40 at 4 months (Macula Dmax 58,2 Gy; Optic Nerve: Dmax 31 Gy), 20/80 to 20/30 at 13 months (Macula Dmax 12,2 Gy; Optic Nerve Dmax 36,1 Gy), and 20/70 to 20/50 at 15 months (Macula Dmax 55,2Gy; Optic Nerve Dmax 46,8 Gy). Two of them had pre treatment cataract. One patient developed radiation induced neuropathy (Macula Dmax 51,5 Gy; Optic Nerve Dmax 51,5 Gy). Six patients developed grade 3/4 radiation induced retinopathy, retinal detachment or cataract.

Conclusion: Optic nerve lower Dmax correlates with good visual outcome. Despite high macula Dmax, patients showed improvement of visual acuity. Cataract does not predict worse results.

Objectives: To estimate the reliability of different techniques of eye globe fixation and to determine the most appropriate MRI sequences for tumor visualization and localization.

Methods: From January 2013 to December 2016, 87 patients with uveal melanoma underwent Gamma Knife radiosurgery. The affected eye was immobilized with different techniques (retrobulbar anesthetic block, fixation of two or three rectus muscles), which were tested in order to determine their reliability. For the verification of eye globe position MRI was performed immediately after the Gamma Knife procedure. The stereotactic images before and after radiosurgery were compared using the Leksell Gamma Plan software in terms of the position of the eye globe and the tumor. For visualization of choroidal melanomas, surrounding eye structures and optic pathways different MRI sequences (T1, T2, CISS and T1 with contrast enhancement) were examined to identify the most appropriate ones.

Results: Gamma Knife radiosurgery of uveal melanoma requires reliable eye immobilization, which can be achieved by complete fixation of at least three rectus muscles. Post-radiosugery MRI revealed that retrobulbar anesthesia could not provide reliable immobilization as significant shift of the eye was observed. MRI visualization of uveal melanoma is a complicated task. The tumor may appear with different signal intensity on T1- and T2-weighted images. In all cases, only the combination of T1, T2 and CISS images allowed us to identify the tumor clearly and to distinguish between the tumor and retinal detachment.  We did not find any advantages in T1 with contrast enhancement for tumor visualization.

Conclusion: Reliable eye globe fixation is extremely crucial for radiosugical treatment; only rigid fixation with suturing of the rectus muscles can give one confidence in precise irradiation. Clear visualization of the tumor can be achieved with T1, T2 and CISS sequences.  A combination of precise stereotactic visualization and complete immobilization of the eye makes it possible to perform conformal high-dose irradiation. 

Objectives

            Ten years ago, we reported a retrospective cohort study.^[1] National health service numbers cross-referenced patients treated with radiosurgery against the national cancer database. This ascertained how many of our patients developed malignant tumours, and what the predicted background rate was. Then, with 30,000 patient-years of follow-up, no increased risk was identified. A criticism was the lack of long-term follow-up. We therefore revisited this study, with 10 years more data accrual and follow-up; but are aware that our practice and the resultant questions are changing.

Methods

            The methodology was unchanged. UK patients were identified, and those with abnormal tumour suppressor genes (mainly type 2 neurofibromatosis) excluded. A vital status check corrected for loss of follow-up due to death. Patients were then stratified for sex, age and the year of follow-up data available. This generates, with national incidence rates, a background risk of developing different malignancies. This is compared with new reported malignancies in our treated patient cohort.

Results

            Statistically, with 110,297 patient-years of follow-up, on an age, sex and time matched basis, we would have predicted 11.76 intracranial malignancies, and have identified 11. Eight of these were astro-glial, two were malignant transformations of vestibular schwannomas(VS) 9 and 10 years after the radiosurgery, and one malignant transformation of a meningioma.  In interpreting this, the cohort includes 3012 VS patients, 4489 patients having more than 10 years follow-up, and 1132 more than 20 years.

Conclusions

            This study supports the long-term safety of Gamma Knife radiosurgery. Overall there is no statistical increased risk of malignancy with a data set exceeding 100,000 patient years of follow-up. The two malignant VS transformations are of concern. The background risk of malignant nerve sheath tumours of the eighth cranial nerve without irradiation has been estimated at 1 in 1041 VS,^[2] so with 3012 VS patients treated, this may reflect the background rate rather than an increased risk related to radiosurgery. The more radiosurgery becomes the main primary treatment for VS, the more likely it is that any VS undergoing malignant transformation will have been irradiated.

1. Rowe et al. (2007) Neurosurgery 60:60-66.

2. Carlson et al (2016) J Neurosurg 125:1120-9.

Background

A major concern of patients undergoing Gamma Knife radiosurgery (GKS) for benign tumors is the risk of a separate secondary malignancy or malignant transformation. Long-term follow up studies are lacking and the exact incidence of radiosurgery-associated malignancy is not known. This study quantifies this risk and compares it to estimates of population risk based on the Central Brain Tumor Registry of the United States.

Methods

Data on all patients who have undergone radiosurgery for arteriovenous malformations (AVM), trigeminal neuralgia or benign intracranial tumors was collected through the International Gamma Knife Research Foundation (IGKRF). The incidence of malignant transformation and separate radiation-associated intracranial neoplasia was calculated in patient-years. Follow-up duration was defined as the time from radiosurgery to the time of death or last-follow up.

Results

To date, we have compiled data on 11 320 patients with a total of 59 200 patient-years of follow-up who under radiosurgery for meningioma (n=3141), AVM (n=2854), trigeminal neuralgia (n=1967), vestibular schwannoma (n=1914), pituitary adenoma (n=1182), other schwannoma (n=184) and hemangioblastoma (n=78). The overall median follow-up time was 3.97 years (0-24 years). Follow-up durations included 3928 patients with 2 to 5 years, 2018 patients with 5 to 10 years, 1462 patients with 10 to 15 years and 508 patients with greater than 15 years of follow-up.

Two cases of malignant transformation of vestibular schwannomas were reported at 8.7 and 11.8 years after radiosurgery, pathologically verified as malignant schwannoma. Two cases of presumed WHO grade1 meningioma transformed to an atypical meningioma and malignant meningioma at 3.5 years and 5.3 years respectively. No other cases of malignant transformation were reported.  Three new malignant brain tumors were reported including one AVM patient and one meningioma patient, who developed radiographic features of distant intracranial malignancy at 4.3 and 8.7 years respectively. One patient with a pituitary adenoma developed pathologically verified osteosarcoma locally 12.8 years after radiosurgery.

Conclusion

The present analysis indicates that the incidence of malignant transformation after radiosurgery for benign tumors is approximately 1 in 14 800 patient-years. The incidence of a new malignancy after radiosurgery, either locally or distant, is 1 in 19 733 patient-years. These risks are not substantially higher than the CBTRUS (2009-2013) derived annual incidence rate of all primary malignant CNS tumors of 7.18 per 100 000.  Patients can safely be counseled that the risk of malignancy after radiosurgery remains extremely low even at long-term follow-up of greater than 10 years.

Spinal Cord is the most critical organ in spine radiosurgery. Once the spinal cord has been damaged, it can cause a serious neurological consequence leading to sensory and motor deficit and paralysis. Various mechanisms of radiation tissue reaction have been proposed, but with no therapeutic targets for mitigation of complication. To find a putative target of radiation-induced spinal cord damage, we explored neuroinflammation at the site of radiation to the spinal cord. Fisher 344 rats were irradiated to the spinal cord C4-T2 with a single radiosurgical doses of 23-33 Gy. The rats were randomized to sham treatment and Ramipril, ACE inhibitor, 1.5mg/kg/day until paralysis occurred. The spinal cord was then harvested and tested for neuroinflammation with immunohistochemical studies. There was a sharp curve causing no paralysis at 23 Gy to 100% paralysis above 28 Gy radiation at 125±4 days post-radiation. Ramipril reduced the paralysis rate to 60 % at the high doses, and significantly delayed the onset of paralysis to 135±4 days (p<0.05). The number of microglia (by anti-iba1 stain) on 40X filed by confocal microscopy was 2.5 ± 0.57 in control, 6.9 ± 0.89 in radiation group, and 4.12 ± 1.20 in radiation plus Ramipril-treated group. This was statistically significant (p<0.05) particularly in dorsal funiculus area. VEGF expression were increased in the sham-treated spinal cord, in contrast to the Ramipril-treated group. The finding indicates that Ramipril decreased the paralysis rate and delayed the onset of paralysis. Ramipril reduced the radiation-induced neuroinflammation, and it may be a potential agent of mitigating radiation complication.

Objectives: In principle there are two reasons why dose rate on Leksell Gamma Knife (LGK) is being reduced during patient irradiation: 1) Co-60 sources decay with half-life of 5.26 years and 2) overall irradiation time is being extended by using multiple isocenters and conformal treatment plans (e.g. with blocked beams). It is thus important to study and evaluate effect of dose rate in LGK clinical conditions. This in vitro study is a pilot experimental work performed with meduloblastoma DAOY cells.    

Methods and materials: Multiple experiments were performed with meduloblastoma DAOY cells irradiated on LGK by various dose rate (0.35 – 1.60 Gy/min). Currently, after Co-60 sources reloading in our center, experiments continue with dose rate up to 3.50 Gy/min. Irradiation was performed in a spherical Elekta ABS plastic phantom which was adapted to accommodate micro centrifuge tube (Eppendorf tube) containing cells. Leksell GammaPlan treatment planning software was used to plan cell irradiation. To produce different dose rate, sector blocking (0, 4 or 6 sectors blocked) was used together with 16 mm collimator to ensure cells homogenous irradiation. Plating efficiency and surviving fraction was determined for each experimental cell sample. Nine different doses in the range 0 – 6 Gy were used to have enough experimental points to obtain surviving curve. Linear quadratic model was used to fit experimental data. Surviving curves for different dose rates were plotted and compared.       

Results: This is an initial pilot study with very preliminary data. However, based on so far obtained data it could be observed higher cell survival for dose rates lower than 0.40 Gy/min compare to higher dose rates over 0.75 Gy/min. Currently, experiments continue with dose rates up to 3.50 Gy/min.

Conclusion: Very preliminary data from this study do show different cell survival for studied meduloblastoma DAOY cells based on a dose rate that was used for an irradiation. Higher survival is observed for a lower dose rate. However, to confirm this hypothesis and initial observation, more experimental work is needed.

This study was supported by Ministry of Health, Czech Republic - conceptual development of research organization (Nemocnice Na Homolce - NNH, 00023884)

Objectives: The use of appropriate radiobiological models to estimate the tumor control probability (TCP) and the normal tissue complication probability NTCP is an important step to rank and compare proton vs photon plans for cost effectiveness. The objectives of this study are to asses and quantify the uncertainties resulting from the choice of radiobiological models on the medical decision and to propose a new approach to estimate the real benefit from proton therapy.

Material and methods: clinical cases of cranio-spinal irradiations for pediatric patients were studied. The treatment plans were generated and calculated with photon and proton dose calculation algorithms to deliver the same prescription dose. The DVH metrics are the base of TCP/NTCP calculation. Two radiobiological models were used for TCP calculations: Poisson and equivalent uniform dose (EUD); and two NTCP models were used: Lyman-Kutcher-Burman (LKB) and EUD.

Results and discussion: proton and photon achieved close TCP values with both models. Regarding NTCP reduction, the choice of the NTCP model can deeply influence the medical decision since NTCP-LKB were higher than NTCP-EUD for most of the organs. Nevertheless, the proton plans offer NTCP reductions for most of the OARs. However, the magnitude of absolute NTCP reduction is sometimes very similar due to high TD50/5 values, initially proposed for grade > 2 and late toxicity from photon treatments. This gives the misleading impression that the proton does not show a real benefit in terms of NTCP reduction. Conversely, if comparing EUD values in Gy, it is obvious that proton offers the best and a significant dose reduction (EUD proton << EUD photon), especially for organs in thorax region as lung, heart, esophagus, with p < 0.05. Thus, our findings suggest, that it would be more consistent to refer and fitting TD50/5 to lower grade toxicity as "grade ≤ 2" to better estimate the NTCP reduction and have relevant reasons to select the optimal plan predicting lower toxicity.

Conclusions: The considerable impact of radiobiological model on the radiotherapy outcomes urges to renew the reference toxicities to tune NTCP parameters’.  The use of the recommended mean doses as TD50/5 in NTCP model could be a realistic approach to estimate low grade NTCP. In addition, EUD values translate the DVH data is a robust indicator be better estimate the dosimetric benefit. Moreover, further improvements of DVH, including secondary electrons, RBE variation and secondary neutrons are necessary.

Background

Secondary malignancy is a known complication of any radiation exposure.  Though this risk is highest in the high dose treatment volume, it is clearly established that low dose radiation also results in a lifetime increased risk of cancer.  Intracranial radiosurgery results in a small extracranial dose of radiation which therefore will increase a patient's risk of cancer within the body. Different treatment platforms’ differing physical qualities result in small differences in this body dose delivered. This is of particular importance to young patients treated for benign conditions who have an otherwise normal life expectancy.  The aim of this study is to compare the risk of extracranial secondary cancer after stereotactic radiosurgery (SRS) using different treatment platforms.

Methods

For an average sized 5 year old and adult the dose to the body to 14 female and 12 male organs received during treatment with SRS at 12.5Gy was calculated for four different radiosurgical treatment platforms using doses interpolated from literature and measurements.  Lifetime risk of secondary cancer per 100,000 exposed was calculated for males and females exposed aged 5/15/25/35/45 years old utilising the National Cancer Institute RadRAT (v4.1.1) calculator.  Chi squared statistical analysis.

Results

The platforms investigated were Gamma Knife Perfexion (Elekta, Stockholm), Linac (micro-multileaf collimator (mMLC) (Philips SL75-5 (Elekta) and cones (Radionics, Burlington, MA)), and Cyberknife (Accuracy, Sunnyvale).

The estimated excess number of extracranial cancers after treatment was:

Gamma Knife Perfexion 151-22 (0.43-0.06%) female 5-45yr/ 53-11 (0.14-0.03%) male 5-45yr per 100,000;  Linac mMLC 1840-261(5.3-0.78%) female 5-45yr/ 649-139 (1.7-0.36%) male 5-45yr per 100,000; Linac cones 3010-437 (8.6-1.3%) female 5-45yr/ 1080-235 (2.9-0.61%) male 5-45yr per 100,000; Cyberknife 6680-1270 (19.1-3.8%) female 5-45yr/ 2740-837 (7.3-2.2%) male 5-45yr per 100,000.  At all ages/sex there was a statistically significant (p<0.001) difference in risk of secondary malignancy between Gamma Knife Perfexion, Linac and Cyberknife.

Conclusion

The extracranial dose of radiation that the body receives in radiosurgery results in an excess lifetime risk of secondary cancer which can only be estimated from extrapolated data from other ionising radiation exposure data.   However these risks are potentially substantial for younger patients with normal life expectancy, so patients should be offered treatment utilising a radiosurgery platform which minimises this lifelong risk.  Gamma Knife Perfexion results in the lowest estimated lifelong risk of secondary malignancy, with linac mMLC, linac cones and cyberknife progressively increasing risk.

Purpose: To asses by MRI the dynamic changes of deep brain large metastases after hypofractionation stereotactic radiation therapy.
Methods and Materials: In retrospective analysis, 146 patients (mean age 55 y.o.) with 215 brain metastases, treated by hypofractioning radiotherapy were included (dose 8-10 Gy, mean metastasis volume 1.5 cm³). 1.5 and 3.0T scanners were used with Т13D sequences 1mm before and after contrast enhancing, Т2 WI tra 2mm, flair tra 1-3mm, T2 WI cor 2mm, DWI. In 21 patients SWI was added. In 30 patients СT or MRI PWI and PET with 11C methionine were added. Tumour volume control was performed by GammaPlan 10.1 station. Patients were assessed before treatment, after 1 month, then, every 3 months.
Results: We have detected 6 radiologic patterns: tumour dimensions changes (68%), structural necrosis (28%), metastasis contrasting decreasing (64%), contour changes (83%), perifocal swelling area reduction (97%), intratumour haemorrhage (14%). The most variable were dimensions and volume changes: volume reduction or stabilisation, volume increasing by necrosis or disease progression, volume increasing on 1st control, then decreasing on 2nd control (necrosis), then increasing by haemorrhage. According PET with 11C methionine and PWI continued tumor growth was detected in 4% of cases.
Conclusion: Brain metastases reaction after radiotherapy characterised by heterogeneity. Imaging results (volume and contour changes, haemorrhage) are non-specific and may be signs of local disease progression or postradiation reactions. To process these data in the future and to avoid misinterpretation of the results when pseudo-progression and to determine the optimal dose/volume/fractionation need well-designed prospective, multicenter clinical trials with strict inclusion criteria of patients in the study with a standardized MRI Protocol, timely re-examination as MRI and PET and PWI.

Background

Patients with stage IV non-small cell lung cancer (NSCLC) are considered incurable and mainly treated palliatively.  In contrast, a radical treatment approach is increasingly recommended in patients with oligometastatic stage IV disease. The purpose of this study was to investigate progression free survival (PFS) and overall survival (OS) of NSCLC patients diagnosed with synchronous oligometastatic disease who underwent radical treatment.

Methods: Patients with NSCLC and oligometastatic disease at diagnosis, who were treated with radical intent between 2008 and 2016, were included.  Treatment consisted of systemic treatment and radical/stereotactic radiotherapy or resection of the intrathoracic disease. Treatment of the metastases consisted of radical/stereotactic radiotherapy, surgical resection or radiofrequency ablation (RFA).

Results: Ninety-one patients (52% men, mean age 60 years), were included with a median follow-up of 35 months.  Almost all patients (98%) were in good condition ( WHO=0-1). The intrathoracic tumor stage, ignoring M-status, was; IA ( (11%)), IB (3%), IIA (12%), IIB (11%), IIIA (36%) and IIIB (26%). Seventy-seven patients (85%) presented with a solitary metastasis, mostly occurring in the brain (32%), bone (25%) or adrenal gland (14%). Fourteen patients presented with 2 or more metastases (9 patients (10%) with 2 metastases, 2 patients (2%) with 3 metastases and 3 patients (3%) with 4 metastases, respectively).

Eighty-one patients (89%) were treated with radiotherapy for the primary tumor of whom 75 patients (93%) received either concurrent (N=9 (12%)) or sequential (N=66 (88%)) systemic therapy. Eight patients (9%) underwent surgery for the primary tumor; 2 patients (2%) received only systemic treatment.

The metastases were treated with ablative/stereotactic radiotherapy (72 (79%)), surgical intervention (5 (6%)), only systemic treatment (6 (6%)), a combination of surgical intervention and radiotherapy (6(6%)), gamma knife (2(2%) and RFA (1(1%))

Thirty-eight patients (42%) died during follow-up. The cause of dead was lung cancer in all patients, except one. Sixty-three (69%) patients developed recurrent disease. Eleven recurrences (17%) occurred within the irradiated area. Most recurrences where brain (16(25%)) and pulmonary metastases (13(21%)).

For the whole group, the median PFS was 14 months (range 2-89, 95% CI 12-16) and the median OS was 32 months (range 3-89, 95% CI 25-39). The 1- and 2-year OS rates were 85% and 58% and  the 1- and 2-year PFS rates were 55% and 27%, respectively.

Conclusion: Radical local treatment of a selected group of NSCLC patients in good condition presenting with synchronous oligometastatic stage IV disease resulted in favorable long-term PFS and OS.

Purpose or Objective: To evaluate the effects of Normal Tissue Objective (NTO) and Stereotactic Radiosurgery Normal Tissue Objective (SRSNTO) in radiotherapy treatment planning to reduce dose to normal brain and surrounding OARS whilst maintaining PTV coverage.

Materials and methods: Twenty patients, previously treated, for multiple cranial metastases in a single fraction were retrospectively planned using the Eclipse Treatment Planning System [Varian Medical Systems, Palo Alto, CA, USA]. In each case, NTO and SRSNTO were compared as means of controlling dose fall off from the PTV. The cohort of patients included one 4 lesion case, three 3 lesions cases and sixteen 2 lesions cases. PTV volumes varied from 0.5cc to 15.4cc. Plans included 5 non-coplanar arcs with a single isocentre placed at the centre of mass of the total target volume. All plans were optimized using standard objectives with both NTO and SRSNTO. Replans were also done with no upper constraint on target dose. All plans were normalized such that 99% of the target volume received 80% of the prescribed dose. Plan quality was evaluated by; volume of brain receiving 4Gy, volume of brain receiving 12Gy, mean dose to brain, dose to OARS and PTV coverage.

Results: PTV coverage was comparable amongst the techniques. D2% to PTV increased by 1% when planning with SRSNTO compared to NTO. SRSNTO with no upper objectives results in increased D2% by 13% compared to with standard objectives.

V12Gy to brain was not significantly different (p = 0.1) between NTO and SRSNTO but was significantly reduced (p = 0.01) with SRSNTO with no upper objectives. Mean dose to brain was significantly reduced when optimized with SRSNTO (p = 0.01). Plans optimized with SRSNTO and no upper objectives saw a further decrease in mean dose to brain (p=0.01).

V4Gy to brain was reduced by an average of 38cc (range 5cc-259cc) representing an average 28.7% reduction when planning with SRSNTO compared to NTO (p = 0.01). SRSNTO with no upper objectives resulted in a further significant decrease of V4Gy to brain (p=0.01).

OAR doses were not significantly changed when planning with NTO or SRSNTO and were all clinically acceptable.

Conclusion:  Multiple target radiosurgery planning with SRSNTO can result in significantly decreased V4Gy and mean dose to normal brain with no reduction in PTV coverage. SRSNTO optimised with no upper objectives and no OAR constraints offers an uncomplicated planning solution to reduce brain dose further with little optimizer interaction.

Purpose: Little is known about the results of post-stereotactic radiosurgery (SRS) salvage whole brain radiotherapy (WBRT) in brain metastasis (BM) patients.

Methods: Using our prospectively accumulated database including 1194 consecutive patients undergoing SRS alone for initially-diagnosed BMs during the 2009-2012 period (JLGK0901 Study, Lancet Oncol 2014;15: 387-95, UMIN ID; 000001812), we studied the 127 patients (10.6%, 55 females, 72 males, mean age; 62 [range; 36-83] years) who underwent salvage WBRT. Competing risk analysis was applied, as appropriate.

Results: Cumulative incidences of WBRT were 6.3%, 8.8%, 10.2%, 10.8% and 11.0% at the 12th, 24th, 36th, 48th and 60th post-SRS month. Patient age <65 years, SCLC and 2-4 tumors correlated significantly with WBRT. Post-WBRT MST was 4.2 (95% CI; 3.0-5.0, IQR; 2.1-9.5) months. Actuarial survival rates were 34.0%, 17.3%, 6.9%, 2.6%, 1.3% and 1.3% at the 6th, 24th 19th, 24th, 30th and 36th post-WBRT month. The crude incidence of neurological death was 31% (37 patients) and actuarial neurological death rates were 19.8%, 26.5%, 27.4%, 30.0%, 30.0% and 30.0% at the 6th, 24th 19th, 24th, 30th and 36th post-WBRT month. MRI-confirmed leuko-encephalopathy occurred in 11 patients (9%) and WBRT was a significant factor impacting a higher incidence of leuko-encephalopathy (HR; 0.008, 95% CI; 0.001-0.081, p<0.0001). SRS-related complications occurred in 25 patients (18%). Age >65 years, neurological symptoms and WBRT (HR; 0.560, 95% CI; 0.365-0.861, p=0.0081) correlated significantly with higher complication rates.

Conclusions: To our knowledge, this is the first prospective study demonstrating the results of post-SRS salvage WBRT in BM patients.

Objectives:  To evaluate freedom from progression (FFP) and risk of adverse radiation effect (ARE) among brain metastases re-treated with stereotactic radiosurgery (SRS) after prior SRS.

Methods:  Brain metastases retreated with SRS were identified within a cohort of 4365 brain metastases with available follow-up imaging treated with single-fraction Gamma Knife SRS at our institution from September 1998-December 2013. FFP and ARE were measured from the date of repeat SRS with censoring at last follow-up imaging.

Results:  A total of 136 brain metastases in 74 patients had repeat SRS at a median of 13.3 months (interquartile range, 7.1-22.4 mo) after prior SRS with or without history of prior whole brain radiotherapy.  The most common primary sites were breast (50), lung (25), and melanoma (42).  The median imaging follow-up after repeat SRS was 16.3 months among these lesions with at least one follow-up scan.  The median quadratic mean diameter (QMD) was 1.4 cm (range, 0.3-4.4 cm; interquartile range, 0.9-1.9 cm) and median target volume 1.1 ml (range, 0.03-26.9 ml; interquartile range, 0.3-2.9 ml).  The median dose was 18.0 Gy (range, 12.0-20.0 Gy; interquartile range, 17.5-18.5 Gy).  Overall, the 1-year FFP probability was 81% (95% confidence interval, 72-87%).  The 1-year probability of ARE by imaging was 31% (24-43%) with a 13% (8-22%) 1-year probability of symptomatic ARE.  The 1-year FFP probabilities were 91% (76-97%), 83% (69-91%), and 58% (36-74%) for QMD ≤1 cm, 1.01-2 cm, and >2 cm, respectively, with corresponding 1-year ARE probabilities of 0%, 18% (9-34%), and 23% (10-48%).  On Cox proportional hazards multivariate analysis, both longer interval from prior SRS to repeat SRS by quartile and smaller QMD (≤1 cm vs. 1.01-2 cm vs. >2 cm) were significantly associated with higher FFP probability (p = 0.003 with HR 0.616 and p < 0.001 with HR = 2.71, respectively).  Only QMD was significantly associated with risk of symptomatic ARE (p = 0.008).

Conclusion:  Repeat SRS for brain metastases with QMD ≤2 cm yielded good local control with acceptable risk of symptomatic ARE.  Longer interval from prior SRS was associated with higher control probability.

Purpose / Objective: Symptomatic radionecrosis is the most common side effect of brain metastases stereotactic radiosurgery (SRS), occuring in around 10% of the patients.  Brain volume receiving 12 Gy or more (V12) is the most published identified risk factor.  Different cut-off values are reported but the risk may be modulated by different factors like e.g. smoking history, diabetes, involved lobe, previous whole brain radiotherapy (WBRT),...

Brain metastases SRS is performed since 2008 in CHU-UCL-Namur.  In this retrospective study, we aimed to identify risk factors of symptomatic brain radionecrosis after SRS for metastases and to develop a specific predictive model to calculate prospectively the individual risk probability.

Materials and Methods: We identified all patients treated with single-fraction SRS for brain metastases at the CHU-UCL-Namur between 2008 and 2016. All patients eligible for retrospective analysis had to be followed-up by regular clinical examination and imaging for a minimum of six months.  We recorded the following potential variables: V12, history of WBRT, localization of the metastases, post-operative status, DS-GPA index (diagnosis-specific estimate of survival for patients with brain metastases), smoking history and diabetes.  Univariate and multivariate analyses were performed to assess the predictive value of multiple variables.  Logistic regression was used to design a predictive formula of individual risk.

Results: 131 eligible patients were identified with a total of 226 different SRS treatments, some patients being treated several times for different metastases.  Symptomatic radionecrosis developed in 21 patients (9.29% of the 226 irradiated targets, 16.03% of the whole population surviving more than 6 months).

Multivariate logistic regression analysis identified V12 as the only risk factor (P = 0.004) and history of previous whole-brain irradiation as a protective factor (P = 0.026). These were the only independent variables that correlated significantly with the occurrence of symptomatic radionecrosis. None of the other tested factors did.

We used logistic regression to design a risk prediction model for symptomatic radionecrosis, integrating V12 and WBRT history. This formula estimates the probability of occurrence : Probability = e^ß/(1+e^ß) with ß=-2,50+0,66(V12)-1,2(WBRT).

Conclusion: Symptomatic radionecrosis from brain metastases radiosurgery can be individually predicted by a statistical model according to the 12 Gy treatment volume.  The apparent protective effect of previous whole-brain irradiation is a suprise and should be further investigated.

Objectives:

To evaluate the tolerance and effectiveness of SABR applied in the treatment of Low and Intermediate Risk (LR&IR) Prostate Cancer Patients (PCP).

To provide an evaluation of the level of risk group impact on the treatment results.

To assess Androgen Deprivation Therapy (ADT) usage on PSA decline after SABR.

Material and Methods:

400 PCP (213 LR and 187 IR, including T_2c) irradiated with CyberKnife using the fd 7.25 Gy to the TD 36.25 Gy. At the start of treatment, 60.3% of patients used ADT and this percentage gradually decreased to 0% after 38 months. Follow-up median 15.0 months. Patients were monitored on SABR completion and subsequently 1, 4, 8 months later and then every 6 months. GI and GU acute and late adverse effects, PSA and ADT usage were evaluated.

Results:

9 patients (2.25%) failed (5 in LR and 4 in IR group) - 4 relapses and 4 nodal metastases. No G3/4 late adverse effects (EORTC/RTOG) were observed. 0.5% G3 GU and 0.3% G3 GI acute reactions, on SABR completion day and one month later, were noted respectively. The median of PSA declined 1.5 ng/ml during the firs month and 0.6 ng/ml during the next three months. No impact of risk groups on treatment results was found. the only impact og ADT on PSA decline was confirmed for ADT&time points interaction.

Conclusions:

SABR of LR and IR PCP is safe and effective treatment.

The inclusion of T_2c patients and the low percentage of IR patient failure permit us to form the assumption that this procedure could be utilized in the treatment of more advanced cases.

The results do not let us define clearly the impact of ADT on radioablation results of LR and IR+T_2c prostate cancer patients.

 Purpose: Stereotactic body radiotherapy (SBRT) is an established technique in early stage lung cancer. Both volumetric modulated arc (VMAT) and conformal arc (3DCA) techniques can be used as a treatment method. Previously, we have shown that VMAT is superior to 3DCA technique in terms of plan evaluation parameters.  In this study, we aimed to analyze whether deformable margin delineation technique (DMD) improves the quality of the 3DCA technique and to compare it with VMAT plans.

Methods: Twenty stage I non-small cell lung cancer patients were included. VMAT and 3DCA non-coplanar plans were generated with 6MV FFF photons to conform planning target volume (PTV) according to RTOG 0915. As conventionally 3DCA plans have inferior target coverage compared to VMAT plans, we deformed PTV contour in order to get a better isodose coverage using DMD technique. Briefly, DMD technique is adaptation of isodose levels to the PTV by deforming margins. All techniques were compared in terms of dosimetric parameters; Ratio of prescription isodose volume (IV) to PTV (conformity index - CI), ratio of 50% prescription IV to PTV (Intermediate dose spillage volume - IDSV), maximum dose in % of dose prescribed at 2 cm from PTV (Intermediate dose spillage location - IDSL), and percentage of lung receiving 20 Gy (V20) respectively.

 Results: All plans were acceptable and no deviation was observed according to RTOG criteria’s. CI ranged between 1.00–1.17 (Mean: 1.02); 1.00–1.25 (Mean: 1.06); 1.04–1.29 (Mean 1.15) for 3DCA-DMD-FFF, VMAT-FFF and 3DCA-FFF, respectively. 3DCA-DMD have significantly better CI compared to others (p<0.001, p<0.001). IDSV values ranged between 3.22–4.74 (Mean: 4.00); 3.24–5.92 (Mean: 4.15); 3.27–5.30 (Median: 4.17) for 3DCA-DMD, VMAT-FFF and 3DCA-FFF, respectively. 3DCA-DMD have significantly lower IDSV which indicates superior falloff gradient (p<0.013, p<0.001). IDSL values ranged between 35.7%-67.0% (Mean: 53.2%); 42.1%-79.2% (Mean: 57.8%); 38.8%-75.8% (Median: 57.4%) for 3DCA-DMD, VMAT-FFF and 3DCA-FFF, respectively. 3DCA-DMD have significantly lower IDSL values which shows a better falloff gradient 2 cm away from PTV (p=0.011, p<0.001). V20 for lung ranged between 0.86%–11.9% (Mean: %4.19); 0.80%–14.51% (Mean: 4.31%); %0.80-%10.88 (Mean: %4.06) for 3DCA-DMD, VMAT-FFF and 3DCA-FFF, respectively. No difference was found between 3 techniques for V20 (p=0.881, p=0.079).

Conclusion: Our results have shown that 3DCA plans can be ameliorated by using DMD method and can be even better than VMAT in terms of CI, IDSV and IDSL.  We believe that 3DCA-DMD is a novel, simple and effective technique for a better SBRT  plan.

Purpose 

The aim of the study is to evaluate the feasibility, acute toxicity and symptoms control of CyberKnife (Accuray, Sunnyvale, CA)-based stereotactic radiotherapy (CBK-SRT) on intraorbital metastases.

Materials and Methods 

This retrospective analysis included patients (pts) with symptomatic metastases located wholly within the orbit. Palliative radiation treatment was performed using CyberKnife image-guided technology (using skull-tracking technique). Gross tumor volume (GTV) volume was defined on a pre-radiotherapy magnetic resonance imaging (MRI) with Gadolinium. Treated volumes and dose-volume histograms (DVHs) are discussed. Acute toxicity was recorded according to Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer (RTOG/EORTC) Scale.

Results 

Between April 2012 and July 2016, 24 metastases (21 pts, 3 treated bilaterally) underwent CBK-SRT for intraorbital lesions (10 intraocular, 14 periocular) from different primary tumors (breast in 13 pts, lung in 3 pts, kidney in 2 pts, lymphoma in 1 pts, thyroid in 1 pts, trunk leiomyosarcoma in 1 pts). Patients were treated in 2 to 3 fractions (median 3 fractions), individual sessions of CBK-SRT were separated by 24 hours. Treatment dose was prescribed to a median isodose line of 75% (range 73 - 80%) normalized to an median maximum dose of 24 Gy (range 18.75 - 32 Gy). Median volume of GTV was 1.50 cubic centimetre (range 0.14-17.5), median maximal dose to ipsilateral optic nerve was 16.95 Gy (range 1.55-24.00), median mean dose to optic chiasm was 2.1 Gy (range 0.50-18.00), median mean dose to ipsilateral lens 1.35 Gy (range 0.30-9.00). At the end of the treatment, grade 1 toxicity according to RTOG/EORTC score was observed in 8 cases. No change in visual field or loss of vision was documented. 14 lesions of 24 had undergone post-radiotherapy MRI and after median follow-up of 5.5 months (range 2.0-26.5) and no local progression occurred: 6 complete response,  6 partial remission and 2 stabilization of disease were observed. All of these patients reported decreasing pre-radiotherapy symptoms and improvement in quality of life. Longer follow-up (more than 12 months) is available in 4 lesions with complete radiological response in all cases.

Conclusions

In our experience, CyberKnife radiotherapy is a well-tolerated, safe and efficacious technique for palliative treatment of intraocular and periocular metastases.

Objective: To analyze the patterns of care of local therapies and their impact on overall survival (OS) among elderly patients with early-stage non-small cell lung cancer (NSCLC) in the United States.

Methods: The National Cancer Database was queried for patients at least age 80 years with NSCLC diagnosed between 2004-2013 with clinical stage T1-3N0M0. Local therapy was analyzed over time and by age. Univariable and multivariable (MVA) models were performed to investigate the impact of prognostic factors on OS.

Results: Among 40,561 patients meeting inclusion criteria, 17,418 (43%), 13,008 (32%), and 10,135 (25%) of patients underwent surgical resection, radiotherapy, and observation, respectively, as their initial mode of local therapy. Overtime, while the utilization of surgical managements generally remained stable, the utilization of conventionally fractionated radiotherapy and observation decreased in favor of stereotactic body radiotherapy (SBRT, p < 0.001). Among operable patients (n = 16,377), after MVA several factors were associated with OS including the choice of local therapy favoring resection over conventionally fractionated radiotherapy and observation (HR compared to lobectomy 1.362, and 2.656, respectively, each p < 0.001). In contrast, there was no statistical difference in OS between resection and SBRT among operable patients (HR for SBRT 1.128, p = 0.156).

Conclusions: The utilization of SBRT as the definitive local therapy in elderly patients with early-stage NSCLC is increasing in the U.S. Given its generally favorable toxicity profile, SBRT should be considered in the substantial proportion of elderly patients still not receiving any definitive local therapy. Among medically operable elderly patients, OS was similar between resection and SBRT.

Purpose: To provide a benchmark for seminal vesicle (SV) margin selection to account for intra-fractional motion, and to investigate the effectiveness of two motion surrogates in predicting intra-fractional SV volumetric coverage.
Methods and Materials: 15 prostate Stereotactic Body Radiation Therapy (SBRT) patients were studied. Each patient has five pairs (one patient has four pairs) of pre-treatment and post-treatment cone-beam CTs (CBCTs). Each pair of CBCTs was registered based on fiducial markers in the prostate. All pre-treatment SV volumes were expanded with isotropic margins of 1, 2, 3, 4, 5 and 8 mm to form a series of PTVs, and their corresponding intra-fractional coverage to the post-treatment SV was used to calculate the “ground truth” for exact coverage with different margin recipes. Two commonly used motion surrogates, the center-of-mass (COM) and the border of contour, were evaluated using Pearson product-moment correlation coefficient and exponential fitting for predicting SV underdosage. Action threshold of each surrogate was calculated. For reference, the margin for each surrogate was also calculated based on a traditional margin recipe.
Results: 95% post-treatment SV coverage can be achieved in 9%, 53%, 73%, 86%, 95% and 97% fractions with 1, 2, 3, 4, 5 and 8 mm margin size, respectively. 5 mm margins provided 95% intra-fractional SV coverage in over 90% of the fractions.
The correlation between the COM and border was weak, moderate and strong in the left-right (LR), anterior-posterior (AP) and superior-inferior (SI) directions, respectively. Exponential fitting gave the underdosage threshold of 4.5 and 7.0 mm for the COM and border. The Van Herk’s margin recipe recommended 0, 0.5 and 0.8 mm margins in the LR, AP and SI directions based on the COM. For the border, the corresponding margin was 1.2, 3.9 and 2.5 mm.
Conclusions: 5 mm isotropic margins for the SV is the minimum required to mitigate the intra-fractional SV motion relative to the prostate. Both the COM and border are acceptable predictors for SV underdosage with 4.5 and 7.0 mm action threshold. Traditional margin calculation based on the COM or border underestimates the margin and should be avoided for this application.

PURPOSE:  Aim of the present study is assess efficacy and toxicity of Stereotactic Body Radiotherapy (SBRT) in patients affected by unresectable locally advanced pancreatic cancer.

MATERIALS AND METHODS: All patients received a prescription dose of 45 Gy in 6 consecutive fractions. Delivery was performed with VMAT and flattening filter free beams (FFFs).  Primary end point of the study was freedom from local progression (FFLP) while secondary end points were overall survival (OS), progression-free survival (PFS), and toxicity. Actuarial survival analysis and univariate or multivariate analysis were investigated. Toxicity was recorded according to the common toxicity criteria version 4.0.

RESULTS:  Forty-five patients were enrolled in a phase 2 trial with a median follow-up of 13.5 months. Two-year FFLP was 90% and both on univariate (P < .03) and multivariate analyses (P < .001), lesion size was statistically significant. Median PFS and OS were 8 and 13 months, respectively. On multivariate analysis, tumor size (P < .001) and FFLP (P < .002) were significantly correlated with OS. Chemotherapy before SBRT were administered in thirty-two (71%) patients with locally advanced pancreatic cancer. Median OS from diagnosis was 19 months. Tumor diameter (P < .002), FFLP (P < .035), and computed tomography SBRT (P < .001) were significantly correlated with OS from diagnosis, at multivariate analysis. No grade 3 or greater toxicity was observed.

CONCLUSION: Stereotactic body radiotherapy is a safe approach for patients with locally advanced pancreatic cancer  and it fits efficiently in a multimodal approach.

Introduction: It is believed that arteriovenous malformations(AVMs) are dynamic lesions with ongoing morphological and hemodynamic changes. Enlargement and progression of pediatric cerebral AVMs following Gamma Knife Radiourgery(GKRS) is a rare phenomenon. Herein, we report 5 cases of pediatric AVMs who presented progression of AVM nidus years after GKRS and discuss their treatment strategies.

Methods: We performed a retrospective review of 5 pediatric patients treated with GKRS for cerebral AVMs at our institution from January 2004 up to and including September 2015. Patient demographics, AVM characteristics, GKRS parameters and AVM responses were recorded.

Results: All 5 patients were treated with GKRS at least twice and  followed up both clinically and radiologically. The median age was 7 years (range 5-11 years), with 3 being boys (60%). The median AVM volume were 3.8 cm³. The median Spetzler-Martin (SM) and Pollock grades of the treated AVMs were 2 and 0.86 respectively. The median follow-up was 63 months (range 42-122 months) after first GKRS. The parameters for median and range in first GKRS planning were prescription isodose 55% (50%-59%), prescription dose 16 Gy (15-18 Gy), maximal dose 29.1 Gy (25.4-36.0 Gy), and number of shots 7 (4-14). Dynamic enlargement and progression of AVMs were confirmed based on angiography in 4 patients and magnetic resonance imaging(MRI) in one patient during a median latency period of 5 years(range 3-9 years). 3 of 5(60.0%) AVMs treated demonstrated a response on follow-up MRI and displayed obliteration based on angiography. But recurrent nidi occurred adjacent to the obliterated nidus 3-8 years (median 5years) after GKRS. Enlargements of AVM nidus were seen in the other 2 patients 5 and 9 years after GKRS respectively. After progression 4 patients were treated with second GKRS and the other one received a combined treatment of embolization and GKRS. No patient developed post-GKRS edema or other major complications.

Conclusions: The findings from this study suggest that GKRS is a safe and effective treatment for pediatric AVMs, yielding an acceptable obliteration rate with minimal permanent severe morbidity and no mortality. Sporadic case reports of AVM nidus progression after GKRS in children with a long latency warrant further investigation and call for continuous follow-up of the pediatric AVM patients even after radiographic confirmation of nidus obliteration.

Object:

To determine the efficacy and safety of gamma knife radiosurgery for atereovenoius malformations of paediatric and adolescent patients.

Methods.

Between May 2008 and August 2016 more than 2300 patients were treated using 201 source cobalt 60 Leksell gamma Knife 4c at Pakistan gamma Knife center Karachi. It included 372 patients with AVMs. There were 93 pts which were 18 yrs or younger. Seventy one patients (72.4%) had hemorrhage at the time of presentation. Fifteen patients had multiple hemorrhages (15.3%).Mean target volume was 3.7cc (range 0.32-31.8cc).Mean prescription margin dose used was 18.9 Gy (range 14-22 Gy).

Results:

Out 0f 93 patients, radiological follow up for more than 03 years was available for 52 patients. Complete Obliteration on angiography/MRI images was found in 32 patients (61.5 %). Partial or ongoing obliteration is described in 20 patients. No acute morbidity is noted within 48 hours of treatment. Two  patient (post embolisation) with partial obliteration with evidence of rebleed was retreated at 02 years. The incidence of hemorrhage at 03 years after gamma knife was 4%. Post gamma knife odema was noted around the obliterating AVM in 5% cases without new neurological deficits. One patient with SM grade IV in Rt. thalamic region had increase in left hemiparesis. One patient with left occipital AVM developed visual field defects.

Conclusions:

Gamma Knife Radiosurgery for paediatric AVMs offers a safe and effective treatment option, with low permanent complication rates during early follow up.

Objective:We analyzed the outcomes of 65 patients with intracranial AVM treated by multi-session cyberknife radiosurgery（CKR）and compared the rate of obliteration with outlining the content of the AVM nidus. 

Methods: Between January 2008 and February 2011, sixty-five patients underwent multi-session CKR for cerebral AVMs. Among 65 patients, 20 had prior embolization, 4 patients underwent embolization combined with gamma knife, 2 had prior gamma knife. Thirty-nine patients underwent cyberknife as their initial treatment. Delineation of the AVM targets were as follows: AVM with prior embolization, the radiosurgical targets include AVM nidus and embolization areas and some draining veins, but low radiation dose was delivered to embolization parts and veins. According to Spetzler-Martin grading, 13 patients was classified as grade I, 24 patients as grade II, 15 patients as grade III, 8 patients as grade IV and 5 patients as grade V. The mean target volume was 8.8 cm³ (range, 1.2-27.0 cm³).  Seven patients with small volume AVM were irradiated by cyberknife in a single session, the rest patients had 2 sessions (n=35) or 3 sessions (n=23).The mean marginal dose was 23Gy (range 15-28Gy). 

Results:AVM obliteration was confirmed by MRI or angiography in 51 patients at a mean follow-up of 46 months (range 36-70 months). The rates of total obliteration were 78% at 3 years. Marked reduction of the size of AVM has been obtained in 14 patients who were not obliterated completely. Among thess14 patients, 3 had the second CKR, 1 had embolization, 1 had gamma knife, the rest were followed up further. The patients were treated by embolization combined with CKR, the rate of AVM obliteration was 83% (20/24). In Spetzler-Martin grade I and II, 34 of 37 (92%) AVMs was obliterated. The higher rate of obliteration of AVM was related to small volume of AVM located in non-critical areas, prior embolization, in which the radiation target including the embolization area and high dose was also associated with a higher obliteration. Eighteen patients had brain edema in the follow-up MRI, 10 of them needs medication and resolved later. Three patients had a hemorrhage during the follow-up period and recovered partially. A permanent neurological deficit due to adverse radiation effects developed in 2 patients. 

Conclusions: CKR proved to be most effective for patients with smaller AVMs located in non-critical areas. Delineation of AVM target including the embolization area in patients who had prior embolization was key point to higher rate of obliteration

OBJECTIVE: To describe our experience in treating large brain arteriovenous malformations (AVMs) with dose-staged (DS) stereotactic radiosurgery (SRS).

METHODS: We treated 28 patients with large AVMs from September 2008 to January 2014 with a minimum follow up of three years, median of 61 months. Patients mean age was 26 years, with no difference in number of patients for each gender, 14 males and 14 females. Seven were classified with Spetzler-Martin grade V, 13 with Spetzler-Martin grade IV and 7 with Spetzler-Martin modified by Oliveira as grade 3A. All patients were treated with 5 fractions from Monday through Friday. Five patients received the dose of 650 cGy a day, 16 received 600 cGy a day, 5 patients received 550 cGy a day and 2 patients received 500 cGy a day.

RESULTS: The mean obliteration rates for DS-SRS was 21,43%. One patient had a post-SRS hemorrhage and died. We could not observe any alterations in angiographies of three patients.  The initial treatment fractions were 5 x 650 cGy, but due to motor complications we considered to decrease to 600 cGy each fraction. Even with lower doses, it was observed alterations in magnetic resonance image (MRI) as hyper signal in T2 and some transient motor deficits but excepting our major bleeding complication, all patients recovered to their pre-treatment functional status.

COCLUSIONS: Large brain AVMs are a challenge to any method of treatment and considering SRS it is needed a long term follow up, but in our small experience DS-SRS seems to be a reasonable treatment approach.

Objectives

Stereotactic radiosurgery represents a consistent therapeutic option in the management of hemorrhagic brainstem cavernomas (BCs). Its long-term efficacy and related morbidity need to be precised.

Methods

We included 28 patients with hemorrhagic brainstem cavernomas treated by Gamma Knife radiosurgery (GKRS) in our University Hospital between 2007 and 2014. We retrospectively analyzed clinical data and imaging follow-up in order to assess the annual haemorrhage rate and patient functional outcomes. 

Results 

The mean age at treatment was 40.6 years. Mean follow-up was 3.32 years (range 1-6). Patients harbored a mean of 2.1 bleeds before GKRS. The BCs’ location was: 7 in the medulla oblongata, 17 in the pons, and 4 in the midbrain. Median target volume was 137 mm³ (IQR, 89-327 mm³). Median dose was 15 Gy at the 50% isodose. No morbidity related to treatment was reported. Two patients rebleeded after GKRS. The overall annual haemorrhage rate (AHR) before GKRS was 34.9% (30 hemorrhagic events reported during a cumulated time of “diagnosis to treatment” of 86 years). After GKRS the AHR was  2.3% (p<0.001). 92% of patients had a modified Rankin Scale ≤2.  

Conclusions

GKRS is an effective treatment for hemorrhagic brainstem cavernomas to significantly reduce the AHR, with no related morbidity.

Since January 2003 until December 2016, in our Institution 500 patients harbouring AVMs have been irradiated with Cyberknife; 220 patients had unruptured AVMs.

Material and methods: 105 patients were females and 115 males; the age ranged from 12 to 81 years (mean 39 yrs; median 38); The nidi were localized in the temporal lobe in 52 cases, 48 in parietal lobe, 38 in occipital, 23 in frontal lobe, 16 in the basal ganglia; 15 in the cerebellum. The assessment of the volume was made on MRI and AGF images. Range of lesion volume was 1-23 ml (mean 2.3 ml, median 1,6 ml). Deep venous drainage was identified in 31 cases and in 35 cases the drainage was both superficial and deep; 116 pts presented Spetzler-Martin grade III, 66 pts grade II, 27 grade IV and 8 grade V; only 3 patients presented grade I. Pollock-Flickinger score range was 0.23-4.63 (mean: 1.35).

Clinical aspect was characterized by epilepsy in 70 patients (32%), headache in 26 patients (12% ) visus deficits, paresthesia etc…in 22 patients (10%). For 42 patients the AVMswas an accidental finding. Before treatment all patients were investigated with 3D cerebral angiography and then image-fusion technique with CT images  was performed. After treatment MRI was scheduled every 6 months and three years after radiosurgery angiography was performed in 152 patients. 47 patients undergone only MRI and 1 patient performed only CT scan.

The peripheral dose ranged from 10.5 Gy to 22.5 Gy (mean: 18.9 Gy);  the maximum dose from 15 Gy to 30 Gy (mean 25.25 Gy). The follow up range was 6-152 months (mean 51 months).

Results:  The nidus was no longer recognizable in 84/200 (42%) patients with FU of at least 36 months. In 91/200 (45.5%) the nidus was reduced and in 25 nidus was unchanged; 20  patients are lost at FU. After the first treatment we observed 12 cerebral bleeding and  three of them were fatal. 51 patients undergone second radiosurgery: among them we  observed 9  cases of cerebral bleeding and no death. Only One bleeding was noted among 12 cases that required a third radiosurgery.

In 97 patients the symptoms present before treatment disappeared during the follow up and in 20 of them the clinical aspect improved.

Conclusion: Preliminary data of our study seem to indicate Cyberknife radiosurgery is helpful in the control of AVMs  without previous bleeding.  

Purpose

The purpose of this study was to define a more effective internal target volume, namely iITV than the classical internal target volume (ITV) for radiation treatment of lung cancer.

Methods

The ITV takes into account only the geometrical aspect but not the temporal nature of the tumor motion, continuous irradiation the whole trajectory of the GTV defined by the ITV will inevitably lead to excessive irradiation of the neighboring healthy tissue. The iITV defined in this study was on the basis of 4D dosimetry which allows the target volume to be determined in both spatial and temporal domains, with the aim to obtain optimal target coverage by completely eliminating excessive irradiation. The 4D dose distribution over the moving target was calculated using deformable image registration (DIR), which tracks the displacement of each CT voxel during the respiratory cycle. The sum of the dose along the trajectory of each voxel presents the accumulated doses receiving by the target. The lowest dose of the GTV therefore indicates the location of the extremity of target’s extension, and the enclosed isodose volume delineates the boundary of the internal target volume. In contrast to the classical ITV which is determined before dose calculation, the iITV was derived as a result of 4D dose calculation. It is therefore denoted as the inverse ITV (iITV).

Results

Our results suggest that 1) iITV reduced the target dose volume by an average of 16.5% compared with that determined by the ITV,  2) optimal dose planning was generally (but not always) achieved with the planning CT performed at the temporal mean tumor position,  3) the degree of target coverage maximization strongly depends on the nature of tumor movement.

Conclusions

By considering both geometric and temporal factors, the 4D dose planning based on DIR allows ITV to be estimated in a more effective approach. The iITV optimizes the internal target volume which can be applied as an indicator for selecting utmost dose plans.

Introduction: Despite advances in treatment, notably in systemic therapy, the prognosis of pancreatic adenocarcinoma (PADC) remains dismal. Stereotactic body radiotherapy (SBRT) is an emerging tool in the complex management of PADC. We review outcomes of SBRT for PADC at our institution.

Methods: We reviewed patients treated with SBRT for either unresectable advanced PADC or locally recurrent PADC after surgery. Eligible patients were treated as part of a prospective trial and received chemotherapy after SBRT. Treatment was delivered using the Cyberknife® tumor tracking system with markers. The median prescribed dose was 30 Gy (30–35 Gy), delivered in 5–6 fractions. Toxicities were reported as per CTCAE v4.0. Survival outcomes were estimated using the Kaplan-Meier method.

Results: Between October 2010 and March 2016, 21 patients were treated at our center, among them 6 were part of a prospective trial. The median follow-up was 7 months (range: 1–28). The 6-month and 1-year local control rates were 94% and 57%, respectively.  The 1 year overall survival was 25% for locally advanced patients and 67% for those with local recurrences. Eighty percent of cancer related deaths were due to metastatic progression. Five patients (24%) had Grade I-II gastrointestinal acute toxicity; one patient had grade V gastrointestinal bleeding 6 months after SBRT.

Conclusion: SBRT for unresectable and recurrent PADC is feasible and provides reasonable local control. Modest dose schedules may be preferable due to the relationship of the pancreas to dose-limiting organs. More work should be done to integrate SBRT with modern systemic therapy in the management of PADC.

Purpose:  Lung SBRT has reported excellent local control rates of over 90%. A 4DCT scan is recommended by RTOG 0915 for intelligent definition of an ITV. At most institutions a single 4DCT scan is acquired at simulation, and slow-scan CBCT’s are acquired for daily alignments. This approach assumes that the ITV volume is relatively consistent over 1-2 weeks of treatment, and/or that that the blurred CBCT representation of the target is sufficient to allow for visualization of ITV changes that might compromise the quality of target coverage. In this study we utilize 4DCT’s acquired prior to each treatment fraction to accurately quantify the consistency of the ITV throughout treatment.

Method: 18 patients who underwent lung SBRT treatment with three fractions were randomly chosen from our institutional database. Each patient received a 4DCT scan at simulation, and prior to each treatment fraction (4 total 4DCTs) using BodyFix immobilization and RPM on a GE RT16. ITVs were semi-automatically defined for each 4DCT dataset (threshold/edge detection plus expert user edit). For each patient, all 4 ITV’s were compared to each other for volumetric and geometric consistency. A PTV=ITV_sim + 3,4,5mm was used to determine what PTV margin was sufficient to cover the ITV/target for all 3 treatment fractions.

Results: ITV_sim  volumes ranged from 2.59 to 54.11 cc (mean=11.56cc), and treatment day ITV volume differences from ITV_sim ranged from –48% to +65%.  5/18 cases showed ITV volume changes ≥ 40%.    For 4/18 patients the ITV_sim volume was the largest (leading to ‘over treatment’ of healthy tissue for all treatment fractions); 3/18 cases the ITV_sim volume was the smallest (leading to ‘under treatment’ of the ITV for all treatment fractions).  For each patient, if we assume a worst case scenario of the smallest ITV volume occurring on simulation day, 5mm and 4mm PTV margins were sufficient to cover the largest ITV. However, a 3mm margin would have been insufficient for full ITV coverage. And this assumes an ideal image registration and perfect application of corrective shifts, which probably does not occur each day.

Conclusion: Non-trivial changes in ITV volume occur during SBRT of lung, predominately due to changes in patient breathing patterns. 5mm and 4mm PTV margins were sufficient for the patients studied here, if we assume a perfect image guided setup correction. A 3mm margin was insufficient for full ITV coverage.

Purpose We investigated the role of integrated [¹¹C]choline PET/CT for target volume selection and delineation in patients with recurrent prostate cancer following External beam radiotherapy (EBRT) for a salvage focal Cyberknife Stereotactic Hypofractionated Radiotherapy (SBRT) treatment.

Methods and Materials From December 2012 to December 2016 a cohort of 33 patients with initial disease category defined as low (7), intermediate (8), high (18), in accordance to NCCN 2008 guidelines, a median age of 74 years (range 62-89) and an history of locally-recurrent prostate cancer following EBRT were referred to our Department for salvage Cyberknife SBRT. The diagnosis of a clinically evident prostate cancer recurrence was based on biochemical progression and imaging studies (CT Scan, Bone Scan and [¹¹C]choline PET/CT). Median iPSA was 19.6 ng.ml (4.9-88 ng.ml), EBRT doses ranged from 74 to 79.2 Gy (median 76 Gy) and the median interval time between diagnosis of relapse and salvage focal Cyberknife treatment of 66 months (range 12-187). The median pre-reirradiation PSA was 4.84 ng/ml (range 2.23-21.04 ng/ml). To reconstruct CTV and organ at risk, CT scan and MRI with T1-T2 sequences were performed and [¹¹C]choline PET/CT images were fused.Nine patients received 3 fractions of 10 Gy (total dose 30 Gy), 24 patients received 3 fractions of 12 Gy (total dose 36 Gy) delivered to the PET positive prostate node (median volume of 14,3 cc - range 5,75-65,04).

Results Salvage Focal Cyberknife treatment was well tolerated without any RTOG grade 3 acute or late toxicity. With a median follow up of 26 months (range 7-49) we observed the following results: no in field recurrence, resulting in a local control of 100%. In 6 pts, a  [¹¹C]choline PET/CT detect the presence of a local recurrence  (median time 15 months; range 8-22 mts) with the evidence of a new positive prostate node outside the irradiated field requiring a second Cyberknife salvage treatment. 3 patients develop lymph nodes o bone metastases 6, 9 and eleven months after Cyberknife treatment.

Conclusions Cyberkife Hypofractionated stereotactic radiotherapy using [¹¹C]choline PET/CT fusion for image guidance is a suitable technique for partial prostate dose escalation.  According to available literature, [¹¹C]choline PET/CT is not clinically recommendable to plan target volume, nevertheless our promising data suggest  a potential role of [¹¹C]choline PET/CT as an image guide tool for the irradiation of focal prostate cancer relapse. Prospective trials are needed to better define the role of differential prostate treatment on imaging defined targets.

Purpose:  To evaluate the utilization of SBRT in the management of primary and metastatic lung cancer and to evaluate factors associated with SBRT utilization, and disparities in SBRT utilization by race.

Materials/Methods: The National Cancer Database (NCDB) is a comprehensive national database that captures approximately 70% of  newly diagnosed cancer patients in the USA.  Data for patients meeting eligibility criteria for our study (receiving radiotherapy to the lung or chest, utilizing a stereotactic treatment modality, and completing treatment in 1-5 fractions) were extracted from the NCDB 2013 PUF data file (encompassing years 2004-2013).   Race was defined as White (Wh), Black (Bl), American Indian (AI), and Asian (As).  Because there were only 6 patients coded as Hawaiian/Polynesian, this group was included with the As group.   Descriptive statistics were used to summarize variables.  Univariate analysis  (UVA) was used to evaluate for disparities in SBRT use by race and ethnicity.    Association between RT use and covariates was assessed using univariate Chi-square test and multiple logistic regression (MVA).

Results:  There were 369,072 patients diagnosed with lung cancer in the PUF database, of whom 22,556 patients (6.1%) met eligibility criteria.  Median age was 75 years,  46.4% were male and 53.6% were female.  Only  4.1% of patients had stage IV disease, and 2.9% had unknown stage.  Median tumor size was 22 mm.   The most frequently used dose regimen was 50Gy in 5fx followed by 48Gy in 4 fractions or 60Gy in 3 fractions.   Overall, SBRT utilization steadily increased from 0.3% in 2004 to 12.9% of all lung cases in 2013 (p < 0.001).  When evaluating by race, SBRT utilization varied from 3.7% for As patients to 6.5% for AI patients (p < 0.001). Asian patients tended to be older and had a higher percentage of stage IV disease.  AI patients tended to be younger and had smaller median tumor size.  Factors that were statistically significant for SBRT use on UVA and MVA included age, facility type, year of diagnosis, analytic stage, crowfly, and lymph node status.  Race was significant on UVA but not on MVA.

Conclusion:  This analysis demonstrates an increased utilization of lung SBRT from 2004 to 2013.  Numerous factors associated with SBRT utilization including facility type, crowfly distance, and lymph node status were identified.  However, race did not appear to be a significant factor for SBRT use on multivariate analysis. 

Stereotactic body radiation therapy (SBRT), has become an integral component of the management of early-stage non–small cell lung cancer (NSCLC). Unfortunately, SBRT is not readily available in developing countries where the cost of advanced radiation equipment prevents widespread availability of this modality. In addition, experience and expertise in modern techniques is scarce. We describe the onsite implementation of a cost-effective complete lung SBRT solution using deep inspiratory breath-hold (DIBH) technique with the Abches System (Apex Medical), delivered by an entry level Varian IX with on-board portal imaging in one of only few linear accelerator equipped hospitals in Myanmar.

Materials and Method:

Abches is a respiration-monitoring device developed by Onishi et al. for facilitating precise irradiation of a target by assuring a stable breath hold. To simulate the breath-hold technique, we used the MOBIUS QUASAR phantom at different amplitudes (from 1-1.5cm) with designed breathing curve to simulate the breath-hold scenario. We designed a lung SBRT protocol using the above equipment taking into account motion-management and image-guidance procedures essential for safe delivery of SBRT. The local team of physicists and radiation oncologists with little or no prior experience in SBRT underwent a 3-day hands-on, onsite course on the basis and application of the protocol, including QA procedures and targets. Planning was done with Varian Eclipse v13.6. After the course, an audit was conducted on the entire SBRT process delivered by the local team on the phantom. Participants were ‘blinded’ to various ‘errors’ and assessed with a checklist. TLDs were used to validate the dose that was delivered within the “tumor”.

Results:

Checklist items were scored 90% and above. The physicist was able to scan and plan the phantom according to the protocol after 3 days of training and manage different scenarios of breath holding. The radiation oncologist was able to make treatment decisions and contour the target effectively. The RTTs were able to monitor and coordinate the machine beam on timing effectively as well as using the onboard imager to localize correctly. The TLD dose results from the delivery of the plan on the MOBIUS QUASAR was within tolerance of 3%, and the gamma analysis 3%/3mm were above 97% using array detectors.

Conclusion:

Our study showed that by adhering closely to a standardized protocol through structured onsite training, SBRT for early lung cancers can be safely delivered in a developing country cost-effectively, using a base model linear accelarator.

Purpose:

To compare the dosimetry of sector-blocked and non-sector-blocked treatment planning techniques for patients with trigeminal neuralgia (TN).

Methods:

Thirteen cases of TN were evaluated in this IRB approved retrospective study. Gamma Knife treatments were performed using a single 4mm shot placed at/or near the DREZ.  Thin-slice MRI images (SPGR or T2) were used for contouring and treatment plans were generated on GammaPlan v10.  All MRI contour sets and prescription doses (range 75-80 Gy) used for sector-blocked (mixing blocked collimators with 4 mm collimator diameters) treatment plans were reused to generate non-sector-blocked (all sectors 4mm) treatment plans. All treatment plans were evaluated by a single radiation oncologist. A paired t-test was used to statistically compare sector-blocked and non-sector-blocked treatment plans.

Results:

In sector-blocked plans,  a median of 3 sectors were blocked (range 1-4 sectors).  The average volumes of normal brain tissue receiving 10, 8, 5, 2, and 1 Gy for sector-blocked vs. non-sector-blocked treatment plans were (0.94±0.25 vs. 0.90±0.28 cc, p=0.07), (1.36±0.32 vs. 1.28±0.40 cc, p=0.02), (2.76±0.67 vs. 2.49±0.79 cc, p<0.01), (12.05±2.89 vs. 10.16±3.15 cc, p<0.01), and (43.58±13.07 vs. 32.57±12.74 cc, p<0.01), respectively. Sector-blocked plans had a slightly higher average maximum dose to the normal brain tissue than non-sector-blocked treatment plans (47.2±16.0 vs. 46.5±16.9 Gy, p=0.65). The average volumes of brainstem receiving 10, 8, 5, 2, and 1 Gy were 20%, 17%, 17%, 25%, and 52% higher in non-sector-blocked treatment plans, when compared to corresponding sector-blocked treatment plans respectively. The average maximum doses to the brainstem for sector-blocked vs. non-sector-blocked treatment plans were (37.0±22.8 vs. 41.0±23.3, p<0.01). Sector-blocked treatment plans had a higher average maximum dose to the cerebellum than non-sector-blocked treatment plans (10.5±6.8 vs. 7.5±4.7 Gy, p<0.01). The average volumes of ipsilateral temporal lobe receiving 10, 8, 5, 2, and 1 Gy were 29%, 4%, 14%, 33%, and 38% higher in sector-blocked treatment plans when compared to non-sector-blocked treatment plans, respectively. The average maximum doses to ipsilateral temporal lobe for sector-blocked vs. non-sector-blocked treatment plans were (26.8±20.1 vs. 27.8±20.8 Gy, p=0.24). The mean beam-on times for sector-blocked vs. non-sector-blocked treatment plans were (57.8±12.8 vs. 37.4±6.5 minutes, p<0.01). 

Conclusions:

Treatment plans utilizing a sector-blocked shot were capable of achieving greater brainstem sparing than those using a non-sector-blocked shot, but at the cost of delivering higher doses to the cerebellum and temporal lobes, and with significantly longer treatment times. 

Objective.The main studies focused on radiosurgery-induced injury to the brainstem are only five. The largest study by Foote et al. analyzed 149 patients and found that the dosimetric factors predictive of cranial nerve palsy included Dmax ≥17.5 Gy, prescribed dose ≥12.5 Gy, length of irradiated cranial nerve ≥16 mm and tumor volume ≥1.7cc. Based on this data, the authors concluded that doses ≥15Gy to the brainstem conferred a significant increase in risk for cranial nerve complications. However, there are not clear clinical evidences and indications about re/irradiation of brainstem.

The objective of the study was to analyze the radiation-related toxicity of the brainstem re-irradiation and its correlation with radiobiological parameters.

Methods.We analyzed 12 patients who underwent re-irradiation for progression or relapse of tumors of the brainstem or close to it. The clinical results were correlated with radiobiological parameters through the linear-quadratic model to express the re-irradiation tolerance in cumulative equivalent total doses when applied in 2Gy fractions (EQD2cumulative). We used α/β values of 2.1 and 3.3Gy.

Results.The histology was high-grade glioma in 4 patients, metastases in 5, meningioma in 2 and unknown in 1 patient. Three patients underwent 5 radiation treatments (1 3Dconformal RT, 4 SRS), 1 patient received 4 RT treatments (1 3DCRT and 3 SRS), 1 patient received 3 RT treatments (3 SRS), 6 patients received one 3DCRT and 1 SRS course, 1 patient received two SRS treatments. The cumulative EQD2 (3,3) ranged 26.5–116.2Gy (mean ± S.D: 73±26.9Gy). The cumulative EQD2 (2,1) ranged 30.5–130Gy (mean ± S.D: 79.5±29.4Gy). The mean time interval between radiotherapy courses was 18.7 ± 20 months (range 0-72 months; median 12 months; n = 23). The mean PTV was  119.9±369.5cc (range 0.1– 1455.6). The mean follow-up was 44 months (range 10-145 months). At the time of analysis 7 patients were alive. No radionecrosis was reported. Only 1 patient developed G1 ataxia and dysphagia and only 1 patient developed a G2 ataxia. Both patients showed a neurological improvement after 1 month of corticosteroid therapy.

Conclusion.The overall outcome in the twelve described patients seems to be encouraging. Modern irradiation systems make it reasonable to administer successive irradiation treatments. Involving only 12 patients, this analysis cannot be expected to provide ground for us to draw definitive conclusions. However, the retrospective EQD2 values reported in this study can be used as starting point for a study focused on dose-reference for safe re-treatments.

Objectives

Multiple platforms are used to treat vestibular schwannomas (VS) with stereotactic radiosurgery/radiotherapy (SRS/SRT). These platforms have different physical qualities which result in different levels of conformity and gradient index, but it is unclear if those differences affect patient outcomes. At University Hospitals Bristol (UHBristol) patients with VS were treated with SRS/SRT using a linear accelerator X-knife [Radionics] cones (linac) between 2006 and 2013.  From 2013 all SRS was delivered using Perfexion/Icon Gamma Knife (GK) [Elekta, Stockholm].   All patients were managed by the same multidisciplinary team.  The aim of this study was to compare these two cohorts of patients regarding the late side effects of treatment to see if the differing platforms' physical qualities equated to different side effect profiles.

Methods

All patients treated at UHBristol with SRS/SRT using linac or GK for VS with a minimum 1 year follow up were included in the 2 cohorts.  Data collection was retrospective review of case notes for linac and prospective data collection for GK cohort. All side effects, excluding those relating to 8th cranial nerve, including severity, date of onset and resolution were recorded, plus any required neuro-surgical intervention.  Side effects relating to trigeminal and facial nerve were categorised according to effect and permanence.  Statistical analysis used chi-squared test.

Results

302 patients with VS were treated with SRS/SRT in UHBristol 2006-2016. 136/198 (69%) linac and 103/104 (99%) GK patients had follow up >=1 year.  Overall total permanent side effects (barring 8th cranial nerve) occurred in 25% (34) linac v 12.6% (13) GK (p=0.029).  Most late side effects occurred in the first 2 years.  New/worsened objective trigeminal neuropathy occurred in 9.6% (13) permanently/2.2% (3) transiently with linac and 1.9% (2) permanently (0 transient) with GK (permanent p=0.0162; all p=0.0044).  New trigeminal neuralgia occurred in 4.4% (6) permanently/1.5% (2) transiently with linac and 1% (1) permanently with GK (all p=0.048).  There was no statistical difference in facial nerve palsy: linac 3.7% (5) permanent /3.7% (5) transient; GK 1% (1) permanent/1.9% (2) transient, but hemi-facial spasm differed: linac 3.7% (5) permanent/1.5% (2) transient versus 2.9% (3) transient with GK (permanent p=0.0492; total p=0.39).

Conclusion

Serious late effects of treatment regarding trigeminal and facial nerve were significantly reduced in patients treated with SRS/SRT with GK compared to linac, likely related to the improved dose conformity with GK reducing radiation dose to these nerves.

Objective

Dizziness is a major factor affecting quality of life in patients with vestibular schwannoma (VS), but treatment of this is unsatisfactory.  Identification of methods to reduce dizziness is required. The aim of this study is to investigate if severity or changes in dizziness symptoms after stereotactic radiosurgery (SRS) relate to vestibular system dosimetry.

Methods

The patient cohort consisted of consecutive patients with VS treated with SRS at the Bristol Gamma Knife Centre between October 2013 and March 2016. Patients who previously underwent surgical treatment were excluded. Dizziness symptoms were measured using the Dizziness Handicap Inventory (DHI) pre-treatment and at 3, 12 and 24 months post SRS.

The semicircular canals (SSC), vestibule (including the areas of the saccule and utricle) and length of vestibular nerve treated were retrospectively contoured on all treatment plans with treatment dose measured. Dosimetric data collected from this was related to change in DHI scores utilising Pearson correlation for statistical analysis.

Results

Follow up data were available for 86 patients who recorded their DHI at pre-treatment and 3 months post SRS and for 69 at 1 year and 38 at 2 years.  The VS treated was mean 1.8cc [0.07-8.9], received a mean 12.3Gy to 50.1% isodose SRS with the Perfexion/Icon Gamma Knife.  Coverage mean 0.99, gradient index mean 2.83 and Paddick Conformity Index mean 0.82.  The mean cochlea dose (all hearing grades) was 4.7Gy; vestibule mean 5Gy [1.3-8.7Gy] and maximum to 1mm³ mean 7.3Gy [1.5-16.1Gy]; SSC mean 3.3Gy [1-6.2Gy] and maximum to 1mm³ mean 5.9Gy [1.8-9.5Gy]; and length of nerve receiving treatment dose mean 18.5mm (6.9-28.2mm). No significant correlation was demonstrated between any of the dose measurement and changes in DHI at 3, 12 or 24 months post SRS.

Conclusion

Unlike the relationship between hearing loss and cochlea dose, there are no dose parameters within the cochlea, SCC, vestibule or length of vestibular nerve treated that relate to changes in dizziness.

Objectives: The most common side effect of stereotactic radiosurgery (SRS) for pituitary tumors is the development of pituitary dysfunction. Secretory adenomas require higher radiation doses to achieve cessation of excess hormone production. The aim of this study was to identify the role of maximum radiation dose delivered to the infundibulum and to find a cutoff value which could predict the development of hypopituitarism in secretory pituitary adenomas. 

Methods: In this retrospective study, patients with secretory pituitary adenomas, who were treated with SRS at All India Institute of Medical Sciences, New Delhi from January 2010 to June 2013 and had minimum radiological and hormonal follow-up of 2 years were enrolled. Radiological follow-up by means of contrast MRI done annually and ophthalmological follow-up by means of 6 monthly visual field charting were also done. Complete hormone analysis was done immediately before SRS and at 6 monthly intervals in follow-up. Good endocrinologic outcome was defined as decrease or normalization of hormone levels following SRS. 

Results. A total of 45 patients satisfied the study criteria and were included in the study. The incidence of new or worsened hypopituitarism was 40% with a mean follow-up in these patients of 41+/- 12 months (range 24 to 66 months). Maximum infundibulum dose delivered was found to be an independent predictor of development of post-SRS hypopituitarism (p=0.001). The average maximum dose received by infundibulum in patients without hypopituitarism was 14.4 ± 6 Gray (Gy) and in those with hypopituitarism was 23.5 ± 9.2 Gy, which was found to be statistically significant (p=0.001). Based on the ROC curves, a cutoff value of ≥18.4 Gy was obtained which could predict development of post-radiosurgical hypopituitarism, with a sensitivity of 77.78% and specificity of 74.07%. Tumor control rate was 95.6% at a mean radiological follow-up of 40.3+/- 11 months (range 24-66). Following SRS, 86.7% patients showed similar or improved vision from before SRS. Good endocrinologic outcome was achieved in 80.6% for acromegaly patients (25/31), 71.4% for patients with prolactinomas (5/7), and 90% for Cushings disease patients (9/10). 

Conclusion. Our study shows that maximum radiation dose delivered to the infundibulum is an independent risk factor for development of postradiosurgical hypopituitarism. The study also shows that keeping the maximum radiation dose received by the infundibulum below 18.4 Gy significantly reduces chances of development of new onset hypopituitarism in secretory pituitary adenomas.

VIM Gamma Knife Perfexion radiosurgery:

dynamic mono-isocentric shielding for internal capsule sparing

Dorenlot A.; Mariani SG; Champoudry J.; Régis J.

Corresponding author: shg.mariani@gmail.com

When treating tremor by VIM (Ventral Intermediate Nucleus) radiosurgery with Gamma Knife Perfexion (GKP), we face the difficulty of improving the gradient to the internal capsule while limiting the number of beam plugging in order to maintain a good overall gradient.

We intend here to evaluate in VIM radiosurgery the efficacy of an original dynamic plug shaping method designed to improve the versatility of sector plugging with GKP.

This method consists in putting two 4 mm shots at the same coordinates but with a different pattern of plugs and weights. Our in-house method is to plug sectors 6 and 7 with a weight of 1 for the first isocenter  and to plug sectors 2 and 3 with a weight of 0,4 for the second.

Pros and cons of this method are compared to a selection of four other ballistic strategies that could represent an alternative as applied to VIM targeting :

-        native 4 mm shot,

-        4 mm shot with four sectors plugged (sectors 2-3-6-7 for a left VIM and 3-4-7-8 for a right VIM),

-        4 mm shot with two “external” sectors plugged (sectors 2-3 for a left VIM and 7-8 for a right VIM),

-        4 mm shot with two “internal” sectors plugged (sectors 3-4 for a left VIM and 6-7 for a right VIM).

The comparison was led retrospectively on 20 patients treated by Gamma Knife for a radiosurgery of the VIM (8 for left VIM, 12 for right VIM).

 For each patient, and for each of the proposed methods, we measured the gradient index, the V90, V12, maximal dose to the internal capsule (D_{CI_10mm3}), volume of 12 Gy of the internal capsule (V12_CI) and mean dose of the internal capsule (D_ci__mean).

The dosimetric comparison between our in-house method and four other alternatives for VIM targeting shows that our in-house method is a good compromise between sparing the internal capsule and keeping a good gradient of dose fall off over the other surrounding tissues.