ESTRO 2020 Abstract book

S510 ESTRO 2020

fractions,histopathology, GPA score and systemic treatment were analyzed. Univariate analyses were performed. Survival analysis was performed based on Kaplan-Meier curves. The logistic regression (stepwise forward logistic regression) is performed to find variables which are the most important in predicting 3-month survival. ROC analysis was used to find the optimal cut-off point for the GPA scale, the sum of specific clinical characteristics and the Zubrod scale.The survival function percentiles and the probability of 3-month survival were calculated.The 2 survival curves were compared using the log-rank test and Gehan's Wilcoxon test. Results Median survival of the total cohort of patients was 6 months. Patients with severe neurological symptoms presented worse survival. Among unfavorable risk factors more then 3 tumors of at least 3 cm concomitant with neurological deterioration or Zubrod 3 concomitant with neurological deterioration correlates with poor outcome, also GPA score of less then 2 was related to risk of early death in logistic regression (p<0,03) . Death within 3 months was predicted by sum of factors lower then 4 (AUC=0,684), GPA less then 2 (AUC=0,687) and Zubrod 3 (AUC=0,737) . Conclusion SRS alone is an effective treatment even in patients with unfavorable risk factors. However, severe neurological deterioration concomitant with Zubrod 3 or extensive extracranial disease are connected with high risk of death within 3 months. More specifically, the sum of at least 4 of analyzed factors may justified qualification to SRS. PO-0865 Survival outcomes and prognosis in patients with triple-negative breast cancer and brain metastases R. An 1 , Y. Wang 1 , C. Gao 2 , A.S. Raghavendra 2 , D. Amaya 1 , N. Ibrahim 2 , J. Li 1 1 MD Anderson Cancer Center- University of Texas, Radiation Oncology, Houston, USA ; 2 MD Anderson Cancer Center- University of Texas, Breast Medical Oncology, Houston, USA Purpose or Objective Triple-negative breast cancer (TNBC) has a high propensity for brain metastasis (BM) and a poor prognosis. We evaluated survival outcomes and prognostic factors among patients with TNBC and BMs. Material and Methods We retrospectively reviewed 88 patients with BM from historically confirmed TNBC treated at MD Anderson Cancer Center in May 2008 ‒ April 2018 . For the initial treatment to BM, 64 patients received upfront stereotactic radiosurgery (SRS); 21 patients received whole brain radiotherapy (WBRT), and 3 patients received surgery. Endpoints were brain metastasis progression free survival (BMPFS) from the initiation of BM treatment, times to local and distant failure from the start of BM treatment, and overall survival (OS) from BM diagnosis. Both intracranial- local and -distant failure are considered as BM progression. Local failures is defined as increased in size of any treated lesions per radiology report or assessment of treating physicians. Kaplan-Meier analysis and Cox proportional hazard regression models were used to estimate survival curves and identify prognostic factors. Results The median interval between the diagnoses of primary breast cancer and BM was 24.6 months (mo) (95% confidence interval 9.4 ‒ 110.7). The median OS time for all patients was 13.3 months, and the cumulative survival rates were 54.6% at 1 year and 29.8% at 2 years. Factors independently associated with increased risk of death in multivariate analysis were Karnofsky performance score (KPS) <70 (p=0.003) and uncontrolled extracranial metastasis at BM diagnosis (p=0.03). No significant

conventional fractionation (1.8 – 2 Gy/die), after initial postoperative radiotherapy or combined radiochemotherapy. Median age at the recurrence was 53 years (range 21-75 years). Twelve patients received chemotherapy (Temozolomide) concomitant to re- irradiation and adjuvant, 8 patients received re-irradiation and then adjuvant chemotherapy (CCNU), 10 patients received re-irradiation alone. Overall survival was calculated with Kaplan-Meier method. We carried out a neurocognitive evaluation with psycho-oncologist collaboration and patients followed a neurocognitive rehabilitation therapy. Results Mean time between radiation therapies was 36 months (range 6-176 months). All patients carried out re- irradiation, with no cases of Grade ≥3 toxicity, particularly in the twenty patients subjected to concomitant and/or adjuvant chemotherapy. After median follow up of 15 months, median overall survival results 8 months (range 1- 95 months) and was 44% after 1 years and 29% after 2 years of follow up. The patients’ group treated with concomitant chemoradiotherapy (Temozolomide) presents a better median overall survival compared to group treated with re-irradiation alone (16 months vs. 7 months); overall survival at 1 year was 57.1% vs 35.7% and at 2 years was 47.6% vs. 26.8%. From neurocognitive evaluation (Minimental test and quality of life evaluation) we report a good feasibility of re-irradiation, with good compliance to neurocognitive rehabilitation therapy Conclusion In our experience re-irradiation associated with concomitant chemotherapy (Temozolomide) for recurrence high grade gliomas represents a good treatment option, with a better OS after 1 and 2 years respect re-irradiation alone or with adjuvant chemotherapy and presents a good neurocognitive tolerance. Patients selection is important to determinate whom could benefit from these approach. Prospective trials are required to confirm these preliminary findings PO-0864 SRS alone of brain metastases with unfavorable prognostic factors M. Harat 1,2 , M. Blok 3 1 Franciszek Lukaszczyk Memorial Oncology Center, Radiosurgery and Radiotherapy of CNS, Bydgoszcz, Poland ; 2 Collegium Medicum- Nicolaus Copernicus University, Oncology and Brachytherapy, Bydgoszcz, Poland ; 3 Franciszek Lukaszczyk Memorial Oncology Center, Radiotherapy, Bydgoszcz, Poland Purpose or Objective Several studies have shown that the SRS alone can improve the control of the treated lesions and improve quality of life in patients with up to 10 brain metastases. However, treatment of patients with metastatic disease and unfavorable risk factors is still controversial and not well studied. We analyzed brain metastases patients with unfavorable risk factors to optimize qulification to SRS alone. Material and Methods Data prospectively collected and introduced into a registry of patients treated with SRS in our department were analyzed. At least one unfavorable prognostic factor were defined for 85 patients treated with SRS only for brain metastases between 04.2018-03.2019. The unfavorable clinical factors were defined according to extracerebral disease, molecular results, age, line of systemic therapy, mass effect and neurological symptoms. Sum of factors was 1- when one factor was favorable, was 6 - when all six factors were favorable (such patient was excluded form analysis). Additionaly Zubrod performance scale, number of brain metastases, intracranial disease volume, dose,