Purpose/Objective(s): Stereotactic radiosurgery (SRS) for benign intracranial neurogenic tumors is an established standard-of-care. The widespread implementation of SRS for benign spinal tumors has been limited by lack of long-term data. We update our institutional experience of outcomes after SRS for benign spinal tumors. Materials/Methods: In this IRB-approved, retrospective review, we identified 132 patients with 161 benign intradural extramedullary spinal tumors treated with SRS between 1999 and 2016. Of this group, 120 patients with 149 tumors (93%), including 39 meningiomas, 26 neurofibromas, and 84 schwannomas, had follow-up magnetic resonance imaging available for review. Tumor volume on follow-up imaging was calculated as an idealized ellipsoid using 3 orthogonal measurements and was compared to initial pretreatment volume. The cumulative incidence of local failure (LF) was calculated with death as the competing risk. Other endpoints included tumor shrinkage, symptom response, toxicity, and secondary malignancy. Results: The most common fractionation regimens were 16 Gy (range 12 e 21 Gy) in 1 fraction and 20 Gy (range 18 e 30 Gy) in 2 fractions, with a median single fraction equivalent dose of 15.1 Gy 4 (range 11.4 e 22.0 Gy 4) for all patients. Median follow-up was 41.6 months (interquartile range: 23.8 e 95.9 months), including 53 and 20 treatment courses with greater than 5 and 10 years, respectively. There were 9 local failures (6%), for a cumulative incidence of LF of 3%, 5%, and 13% at 3, 5, and 10 years respectively. Of those that failed, the median time to LF was 51.3 months (range 15.6 e 95.1 months). Excluding the 10 tumors that were previously irradiated, including a single patient with 4 LF events in multiple radiationinduced meningiomas, the 3, 5, and 10-year rates of LF were 1%, 2%, and 9%, respectively for these radiation-naïve tumors (n Z 139). At last follow-up, 35% (n Z 52) of all lesions treated were decreased in size, of which 42% (n Z 22) had >50% decrease in estimated volume compared to baseline. Of the 82 treatment courses preceded by pain, 66 had clinical follow-up of symptoms. Of these, pain improved in 36% (n Z 24), was stable in 53% (n Z 35) and worsened in 11% (n Z 7). Median time to pain improvement was 12 months. Following SRS, 11 patients required surgical resection as salvage for tumor growth or persistent symptoms. A single patient, treated in 2003 and previously reported, had transient myelopathy 9 months after treatment. No additional late toxicity or secondary malignancies were observed. Conclusion: Our experience suggests that SRS provides longer-term local control of benign neurogenic spinal tumors, similar to outcomes of SRS for benign intracranial tumors. SRS provides a treatment alternative to surgical resection. Given the risk of late recurrence, we await longer follow-up.

Purpose/Objective(s): For NSCLC patients treated with SBRT, we investigated if proximity to the proximal bronchial tree is associated with non-cancer death. Materials/Methods: From 2006-2013 patients with a single early stage NSCLC tumors were irradiated with CBCT guided SBRT (median dose 54 Gy in 3 fractions) in 5 institutes. Patients with progressive disease, metastases, or second cancers at the time of death were scored as death from cancer; all other deceased patients were scored as non-cancer death. Treatment plans were collected, and the main and lobar bronchi were automatically delineated using atlas based segmentation. For each patient the shortest distance from the edge of the GTV to the proximal bronchial tree (PBT) was determined. Patients were stratified into 3 groups; GTV2 cm from the PBT (peripheral (A); RTOG 0236), GTV 1 cm and < 2 cm from the PBT (B), and GTV <1 cm from the PBT (C). Actuarial non-cancer survival at 1y, 2y and 5y were determined (i.e., death from cancer was censored). Association between the stratified distance of the GTV to the PBT and non-cancer death were evaluated using univariate Cox regression (at the P<0.05 level), and compared to the association with cause specific survival to test for competing mortality risk. Finally, the stratified distance was included in a multivariate Cox analysis, also including; age, gender, performance status, comorbidity index, lung-function FEV1, Mean Lung Dose, tumor diameter, and smoking history. Results: Seven hundred sixty-nine patients were identified with a median age of 75 y. Median Biologically Equivalent Dose (a/b Z 10 Gy) was 126 Gy, 180 Gy and 227 Gy for group C, B, and A, respectively, with 33, 71 and 665 patients per group. Median GTV diameter was 4.1 cm (1.1-7.0), 2.7 cm (0.9-5.7) and 2.2 cm (0.7-6.5) for groups C, B, and A, respectively. Survival rates were lower for patients in group C (Table 1) with a Hazard ratio (HR) of 2.91 (P < 0.001). Patients in group B had a lower, non-significant HR of 0.87 (P Z 0.554). The association with cause specific survival showed a significantly higher HR (2.45, P Z 0.036) for patients in group C with respect to A, but not B. In the multivariate Cox analysis, the stratified distance from the GTV to PBT was significantly associated with non-cancer death (group C: P<0.001, HR Z 3.56, group B: P Z 0.319, HR Z 0.79), as well as age (P Z 0.001, HR Z 1.03), performance status (P<0.001, HR Z 0.30) and lung-function FEV1 (P Z 0.004, HR Z 0.99). Conclusion: Patients with a tumor < 1 cm from the proximal bronchial tree had 3.56 fold higher risk of non-cancer death than patients with a peripheral tumor. Delivered dose distributions will be further studied.

cytotoxic T-cells and reduces immunosuppression by regulatory cells, encouraging abscopal effects for enhanced systemic tumor control. Given the promising finding that RT potentiates the effects of immunotherapy at distant sites, we tested the hypothesis that this combination would improve the time to progression-of-disease (POD) of the primary tumor and/or metastases in patients with Stage III or IV lung cancer. Materials/Methods: We retrospectively analyzed imaging data and radiology reports of 44 patients with Stage III or IV non-small cell (NSCLC) or small cell (SCLC) lung cancers treated with immunotherapy and either concurrent chest RT, prior chest RT, or no prior RT at our institution between March 2015 and November 2016. Immunotherapy was defined as 1 cycle of monoclonal antibodies against cytotoxic T-lymphocyte antigen-4 (CTLA-4) or against programmed death-1 (PD-1). Combined therapy was defined as overlapping time periods of the two treatment modalities, and sequential therapy was defined as chest RT completed prior to onset of immunotherapy. POD was determined by RECIST-defined tumor progression at any site including lung, and/or radiology reports that subsequently led to a change in management. Results: Of 44 patients, 26 (59%) were female, 35 had NSCLC, and the median age was 68 years (range 39-91). The sequential therapy group had 29 patients, the combined group had 6 patients, and the non-RT group had 9 patients. Our analysis failed to demonstrate a significant association between the treatment modality received and time to POD (pZ0.68 on one-way ANOVA). Overall median time to POD was 70 days, and median follow-up time was 379 days. Log rank test of Kaplan-Meier curves also failed to demonstrate a statistically significant difference in time to POD between the three patient groups (pZ0.48). Linear correlation based on age also did not detect a difference in outcome (rZ0.03, pZ0.84 on Pearson's). Conclusion: Numerous prospective and retrospective analyses of cancer treatment have provided intriguing yet mixed results in time to POD and overall survival following combined immunotherapy and RT. While this study failed to detect a difference in outcomes between concurrent and sequential therapies at one institution, further evaluation within larger patient populations allowing for more sophisticated stratification is an appropriate next step.

No abstract

protocol and 31 (15%) patients had maD. Patients with maD had an inferior overall survival (OS) (HR 2.9, 95% CI 1.8−4.4, P < .0001) as well as a higher risk of loco-regional progression (HR 5.7, 95% CI 2.7−11.1, P < .0001). Especially, patients with maD concerning normal tissue delineation and/or dose constraints had a worse OS (HR 3.6, 95% CI 1.5-7.3, P = 0.006) although there was no impact on the incidence of toxicities. Adjusting for the UICC stadium and the gross tumor volume as prognostic factors, the effects remained similar. Conclusion: Non-adherence to the RT protocol was associated with an inferior OS and loco-regional control. These results underline the importance of RTQA.