ESTRO 36 Abstract Book

S599 ESTRO 36 2017 _______________________________________________________________________________________________

K. Latifi 1 , A. Rishi 1 , J. Caudell 1 1 H. Lee Moffit Cancer Center, Radiation Oncology, Tampa, USA Purpose or Objective We hypothesized that the nodal response at the midpoint of radiotherapy for squamous cell carcinoma of the head and neck would correlate with outcome. Material and Methods After IRB approval, 37 patients with non-metastatic squamous cell carcinomas of the head and neck treated with definitive radiotherapy (RT) with cone beam computed tomography (CBCT) or CT on rails during treatment were identified. Nodal volumes were contoured on CT simulation films as well as weekly CBCT or CT on rails. Volume change between the first week of treatment and fourth week of treatment were calculated. Outcomes, including locoregional control (LRC) and disease free survival (DFS) were calculated from the end of RT and estimated via Kaplan-Meier method, with comparisons made via log-rank test. Results Median follow-up of all patients was 12 months. Primary sites included oropharynx (n=34) and oral cavity (n=3). Median dose was 70 Gy (range 54 – 70 Gy). Systemic therapy was used in 78.4% of patients (n = 29). Median response was 32% (range -39.3% - 78%) comparing week 4 and week 1 nodal volumes. Patients with a response greater than 32% at week 4 had a 1 year locoregional control rate of 100% compared with 75.6% for patients with a ≤ 32% response (p = 0.03). Similarly, patients with a nodal response > 32% had a 1 year DFS rate of 100% vs. 67.1% for ≤ 32% (p = 0.01). Conclusion Nodal response at four weeks may be associated with ultimate outcome of patients with squamous cell carcinoma of the head and neck treated with definitive radiotherapy. While further work is needed, monitoring of nodal response may allow for both volume and dose adaptation of therapy. EP-1101 Leptomeningeal spread after stereotactic radiation for brain metastases from breast cancer O. Kaidar-Person 1 , A. Deal 2 , C. Anders 3 , M. Ewend 4 , L. Carey 3 , E. Dees 3 , J. Camporeale 1 , J. Ramirez 5 , J. Benbow 5 , L. Marks 1 , T. Zagar 1 1 University of North Carolina- Chapel Hill- North Carolina- USA-, Department of Radiation Oncology, Chapel Hill, USA 2 UNC Lineberger Comprehensive Cancer Center- Chapel Hill- NC, Statistics, Chapel Hill, USA 3 University of North Carolina- Chapel Hill- North Carolina- USA-, Medicine, Chapel Hill, USA 4 University of North Carolina- Chapel Hill- North Carolina- USA-, Neurosurgery, Chapel Hill, USA 5 University of North Carolina- Chapel Hill- North Carolina- USA-, UNC Lineberger Comprehensive Cancer Center- Chapel Hill- NC, Chapel Hill, USA Purpose or Objective Our objective was to explore the incidence and predictive factors for subsequent development of leptomeningeal disease in women with breast cancer who received stereotactic radiation (SRT, 1 to 5 fractions) for brain metastases. Material and Methods We conducted a retrospective analysis from a prospective collected metastatic breast cancer database of all patients with brain metastases seen between 2012 to 2016. Results Electronic Poster: Clinical track: CNS

A total of 98 patients with breast cancer brain metastases were included. Twenty-one (21%) patients developed leptomeningeal spread (initial enhancement on MRI, with/without symptoms). The median time to development of leptomeningeal spread from the diagnosis of primary breast cancer was 3.7 years (95% CI 1.3 to 15.3) and the median time to development of leptomeningeal spread from the diagnosis of brain metastases was 1.3 years (95% CI 7 days to 4.3 years). All 21 patients developed symptoms due to leptomeningeal disease. Age, primary tumor receptor status, Karnofsky performance status, craniotomy prior to SRT, whole brain irradiation (WBRT), prior to SRT were not associated with increased or reduced risk of leptomeningeal spread (all p> 0.05). Median overall survival from initial diagnosis of leptomeningeal spread was 2.7 years (95% CI 1.4 to 3.7), with 2 patients surviving 5 years (5 and 5.3 years). Conclusion In this study, we did not identify clinically significant factors that were associated with an increased risk of leptomeningeal dissemination. The relatively long overall survival for breast cancer patients with brain metastases suggests that brain SRT remains a valid option for breast cancer brain metastases. Several exceptional responders with brain metastases and leptomeningeal disease were identified. A better understanding of this unique population of patients is needed. EP-1102 Primary and secondary gliosarcomas: clinical, molecular, and survival characteristics H.J. Kim 1 , S.H. Kim 2 , J.H. Chang 3 , I.J. Lee 1 , C.O. Suh 1 , J. Cho 1 1 Yonsei University, Radiation Oncology, SEOUL, Korea Republic of 2 Yonsei University, Pathology, SEOUL, Korea Republic of 3 Yonsei University, Neurosurgery, SEOUL, Korea Republic of Purpose or Objective Gliosarcoma is an extremely rare disease, which is a variant of glioblastoma exhibiting a biphasic histologic characteristic of both glial and mesenchymal components. We investigated to identify prognostic or therapeutic factors impacting on survival outcomes in patients with gliosarcoma treated in a single institution. Material and Methods Patients who had been treated with a pathology- confirmed diagnosis of gliosarcoma at Yonsei University Medical Center between 1991 and 2015 were retrospectively analyzed. Patients who were ＜ 20 years old at diagnosis or have not been followed up after treatment were excluded. Primary gliosarcoma (PGS) was defined as a tumor developed de novo, whereas those diagnosed subsequent to glioblastoma that have been treated with surgery and adjuvant radiotherapy were termed secondary gliosarcoma (SGS). Molecular analysis performed on 9 patients including IDH1, TP53, Ki-67 and EGFR. Results A total of 19 patients were identified, including 17 PGS and 2 SGS patients. All patients received surgery followed by adjuvant radiotherapy with a median dose of 60 Gy. Gross total resection was performed in 6 patients, while subtotal resection was performed in 13 patients. The concurrent chemotherapy with temozolomide was performed in 10 patients. The Median overall survival (OS) for all patients was 12.9 months from the diagnosis of gliosarcoma, with a progression free survival (PFS) of 5.5 months. The median OSs were 12.9 and 5.2 months for PGS and SGS, respectively (P = 0.035) and the median PFSs were 6.1 and 0.8 months (P = 0.048). In the two cases of SGS, SGS developed 10 and 18 months after the diagnosis of GBM for each. The univariate analysis revealed that good performance status (KPS ≥ 80), PGS and salvage treatment after recurrence were significant prognostic