expression below 10%, compared with 0.7 cm for patients with PD-L1 of at least 10%. NLR was positively associated with metastasis size (pZ0.081). PD-L1 expression, NLR, and receipt of immunotherapy before or after SRS were not associated with response to SRS. Conclusion: Tumor and patient immune parameters may influence the presentation of limited brain metastases in NSCLC patients. We did not find that immune parameters or administration of immunotherapy influ- ence radiographic response rates following SRS. Additional studies are required to determine if local management of brain metastases should be altered with the growing role of immunotherapy in the treatment of NSCLC. Moreover, immunotherapy delivered in conjunction with stereo- tactic radiosurgery appeared to be well tolerated. Author Disclosure: A. Orisamolu: None. N. Ohri: None. C. Guha: RTOG. M.K. Garg: Speaker’s Bureau; Varian, Covidien. MO_31_2761 Clinical Outcome of Brainstem Glioma Treated with or Without Concurrent Chemoradiotherapy: An Institutional Retrospective Analysis of 71 Patients G.K. Rath, 1 A.K. Gandhi, 2 S. Mallick, 1 and D.N. Sharma 1 ; 1 All India Institute of Medical Sciences, New Delhi, India, 2 Dr. Ram Manohar Lohia Institute of Medical Sciences, Lucknow, India Purpose/Objective(s): Brainstem glioma (BSG) is an aggressive and rare tumor of childhood and adults. Treatment outcomes are dismal and role of concurrent chemoradiotherapy (CTRT) is not established in these patients .We intended to study the clinical characteristics along with outcome of BSG patients treated with CTRT at our institution. Materials/Methods: We retrospectively evaluated 71 patients of all age groups with BSG treated at our department in the period January 2015 to October 2017. Demographic and disease characteristics in this patient cohort were recorded, and their progression free survival (PFS) was analyzed with respect to age (</> 18 years), gender, grade, use of CTRT and adjuvant chemotherapy. Results: Median age at presentation was 15 years (range 3-66 years), with a male: female ratio of 1.5. 31 patients presented with gait ataxia and 29 presented with cranial nerve palsies. 34 patients were diagnosed radio- logically as high grade, 32 as low grade and grade was equivocal in 5 patients. None of the patient underwent surgery. Radiotherapy dose was 56-60 Gray over 5.5 -6 weeks at 1.8-2 gray/fraction. All patients completed their radiotherapy except 7 patients.28 patients received con- current Temozolomide (75 mg/m 2 ) ,17 patients received adjuvant Temo- zolomide (150-200 mg/m 2 D1-5 every 4 weeks for 3-6 cycles) and 9 patient received both concurrent and adjuvant Temozolomide. Median follow up duration was 23.2 months (range 8-65.3 months).At last follow up, 21 patients had progressive disease. Median PFS for the entire group was 16.7 months and PFS at 24 months and 36 months was 75% and 65% respectively. On univariate analysis, patients less than 18 years had poorer outcome as compared to older (Median PFS 9.5 vs. 41.32 months; p 0.004) and PFS was significantly poorer in the patients who received concurrent Temozolomide than those who did not (Median PFS 9.8 vs. 21.5 months; p 0.05).Gender, grade and adjuvant chemotherapy did not statistically alter treatment outcomes. Age less than 18 years continued to be statistically significant on multivariate Analysis (p .01) Conclusion: Younger age appears to be a bad prognostic factor for BSG patients. Temozolomide has a questionable role to play in the treatment of these patients. Author Disclosure: G.K. Rath: None. A.K. Gandhi: None. S. Mallick: None. D. Sharma: None. MO_31_2762 Prospective Phase II Study of Scanned Beam Proton Therapy for Spine Sarcomas D.J. Konieczkowski, 1 , 2 Y.L.E. Chen, 2 K.D. Bernstein, 2 B.Y. Yeap, 3 M.C. DiMaria, 4 W. Jiang, 5 N. Thomas, 2 S.L. McGovern, 6 J.T. Mullen, 7 J.H. Schwab, 8 F.J. Hornicek, 9 and T.F. DeLaney 2 ; 1 Harvard Radiation Oncology Program, Boston, MA, 2 Department of Radiation Oncology, Massachusetts General Hospital, Boston, MA, 3 Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, MA, 4 Cancer Center Protocol Office, Massachusetts General Hospital, Boston, MA, 5 Biostatistics Center, Massachusetts General Hospital, Boston, MA, 6 The University of Texas MD Anderson Cancer Center, Houston, TX, 7 Department of Surgical Oncology, Massachusetts General Hospital, Boston, MA, 8 Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, MA, 9 Department of Orthopaedic Surgery, University of California Los Angeles, Los Angeles, CA Purpose/Objective(s): Pre- and/or post-operative radiation therapy (RT) may improve local control in patients undergoing surgery for spine sar- comas. In addition, definitive-intent RT may have a role in patients who do not undergo surgery. However, delivery of adequate RT doses is chal- lenging due to proximity of normal structures. We previously reported a Phase II trial of passively scattered proton therapy for spine sarcoma. Here, we hypothesized that scanned beam proton therapy would allow high local control with acceptable toxicity. Materials/Methods: We conducted a Phase II, NCI-sponsored clinical trial of scanned beam proton therapy for primary or recurrent chordomas and chondrosarcomas of the spine or sacrum. In patients undergoing surgery, CTV1 (encompassing pre-operative gross tumor plus a micro- scopic margin) was treated pre-or post-operatively to 45-50.4 GyE, CTV2 (encompassing pre-operative gross tumor) post-operatively to a total of 64.8-70.2 GyE, and GTV (only if residual gross disease) post- operatively to a total of 77.4-79.2 GyE. Surgery consisted of en bloc resection with the goal of R0 margins, although in some cases a microscopic positive margin (R1) was planned with the goal of preser- ving nerve function. In patients not receiving surgery, the same CTV1 and GTV dose levels were used. Results: A total of 60 patients (51 chordoma, 9 chondrosarcoma) were enrolled between February 2011 and March 2017. Sixty-three percent were male, with a median age of 59 (range 18-89). All but one had primary tumors. Tumors were predominantly sacral/coccygeal (nZ35, 58.3%). Fifty-one patients (85%) underwent surgery, of whom 65% had R0, 29% R1, and 6% R2 resections (all at outside institutions prior to trial enrollment). In surgical patients, RT was most often delivered both pre- and post-operatively (32 patients, 63%; median total dose 68.7 GyE), and less often pre-operatively only (7 patients, 12%; median 50.4 GyE) or post-operatively only (12 patients, 20%; median 70.8 GyE). Nine patients (15%) underwent definitive RT without surgery (median 77.4 GyE). With a median follow-up of 28.3 months among 55 patients still alive, two-year local control was 92%, recurrence-free survival (RFS) 85%, and overall survival 94%. On univariate analysis, post- operative only RT (vs both pre- and post-operative) was associated with shorter RFS (62.9% vs 90.8% at two years, HR 5.86 [1.75-19.7], pZ0.005); there was also a trend towards shorter RFS in non-R0 re- sections (71.6% vs 89.1% at two years, HR 2.65 [0.87-8.01], pZ0.074). No local or distant recurrences were observed in the definitive RT arm. Acute and chronic toxicities were mild (3.4% grade 3-4, each); one patient experienced a late grade 2 sacral insufficiency fracture, and two patients experienced late grade 1 hardware failures. No RT-associated myelopathies were observed. Volume 102  Number 3S  Supplement 2018 Poster Viewing Q&A Sessions E325