ESTRO 2020 Abstract book

S555 ESTRO 2020

vs Her2 enriched vs Triple negative vs no-basel type), pathologic T stage (T1 vs T2) and positive lymph node numbers (1 vs 2 vs 3) were independent factors for LRFS (all of P <0.05). A nomogram was then developed to estimate the probability of 5-year LRFS based on these five important variables. The C-index of the nomogram for LRFS prediction was 0.705 and 0.688 in the primary and validation cohorts separately. The calibration curve showed that the nomogram was able to predict 5-year LRFS accurately in both cohorts. According to the total nomogram risk scores of the patients, the cohort was then classified into low-risk group (≤100, 37.9%;) moderate-risk group (101~199, 40.3%;), high-risk group (≥200, 21.8%). The 5-year outcomes were significantly different among these three groups. In low-risk group, patients who received PMRT or not both achieved a favorable LRFS (100% vs 97.1%, P=.067), DFS (94.3% vs 92.9%, P =.229) and OS (98.9% vs 97.2%, P=.308). In moderate-risk group, no differences in LRFS (93.8% vs 93.7%, P=.977) , OS (90.6% vs 96.1%, P=.133), but difference in DFS (81.6% vs 88.1%, P=.024) were observed between PMRT and no PMRT patients. In high-risk group, compared with no PMRT, PMRT resulted in similar DFS (81.9% vs 79.5%, P=.579) and OS (88.4% vs. 89.7%, P=.675), but had a tendency to improve LRFS (92.2% vs 85.1%, P=.051). After PSM adjustment, there were no significant differences in LRFS, DFS and OS in low-risk and moderate-risk groups. However, in the high-risk group, PMRT still resulted in similar DFS (83.1% vs 74.0%, P=.136) and OS (89.4% vs. 90.5%, P=.728) but significantly better LRFS (93.3% vs 82.1%, P=.016). Conclusion The proposed nomogram provides an individualized risk estimate of LRFS in patients with pT1-2N1 breast cancer. Risk-adapted PMRT for high-risk patients is a viable effective strategy for pT1-2N1 breast cancer. Better survival can be achieved in these patients. PO-0954 Occult Primary Breast Carcinoma with axillary involvement: two different treatment approaches. F. Sousa 1 , I. Azevedo 1 , D. Moreira 1 , B. Castro 1 , A. Purpose or Objective Occult breast cancer (OBC) is defined as clinically recognizable axillary metastatic carcinoma from an undetectable primary breast tumor and represents only 0.3 to 1% of breast cancers. The majority of the literature on OBC consists of small, single-institution series and case reports, lacking any randomized trial. Hence there is no clear consensus on the optimal locorregional treatment. This study aims to compare the treatment outcomes in patients treated with mastectomy and axillary lymph node dissection (ALND) or ALND without breast surgery followed by whole breast irradiation (WBI). As secondary objective we looked for clinicopathologic factors of OBC to determine its prognostic significance. Material and Methods We performed a retrospective cohort study using our Institution database to collect clinical and pathological data of female patients diagnosed with OBC between November 2006 and January 2017 treated either with mastectomy with ALND or ALND followed by WBI. Patient characteristics, DFS and OS were compared between those two groups. The statistic analysis was calculated with SPSS 22.0 using Kaplan and Meier method for survival curves and log rank test for group comparison. Results 30 women with median age of 55 years (range:37-79) at diagnosis were identified and included. 9 (30%) patients were treated with mastectomy and ALND followed by radiotherapy to thoracic chest wall (50Gy/25f) and 21 (70%) were treated with ALND without breast surgery Laranja 1 , J. Conde 1 , V. Castro 1 , H. Pereira 1 1 Instituto Português de Oncologia do Porto, Radioncology, Porto, Portugal

between toxicity and the dosimetric/anatomical variables. Multivariate logistic regression analysed, as covariates, the variables with p<0.10 on univariate analysis. P values <0.05 were considered significant. Results From August 2010 to March 2017, a total cohort of 1160 patients have been treated with VMAT and SIB in 3 weeks treatment. Of those, 852 patients had at least 2 year follow-up, and were included in the analysis. Skin area larger than 400 cm 2 and the volume larger than 5 cc receiving more than 105% of the boost prescription dose were significant predictors for acute skin toxicity. Multivariate analysis for late skin toxicity at 2 years was significant for boost volume larger than 70 cc, skin area greater than 400 cm 2 and breast size larger than 1500 cc. At 5 years, none of the analyzed variables were significant. For cosmetic outcome, the multivariate analysis confirmed the significance only for the breast size (>1000 cc) and the boost size larger than 70 cc at 2 and 5 years, respectively. Regarding breast pain, multivariate analysis did not evidence significant predictors at 2 and 5 years. Conclusion Breast size is not a significant predictor for long term toxicity and cosmesis. These data suggest that hypofractionated radiation therapy using VMAT is a viable therapeutic modality in large-breasted patients. PO-0953 Risk-adapted Postmastectomy Radiotherapy based on Prognostic Nomogram for pT1-2N1M0 Breast Cancer M. LI 1 , J. Yue 2 , X. Wan 3 , B. Hua 4 , Q. Yang 5 , P. Yang 6 , Z. Zhang 7 , Q. Pei 8 1 Beijing Hospital/ National Center of Gerontology, Department of Radiation Oncology, Beijing, China ; 2 Shandong Cancer Hospital Affiliated to Shandong University/ Shandong Academy of Medical Sciences, Department of Radiation Oncology, Jinan, China ; 3 The Sixth Affiliated Hospital- Sun Yat-sen University, Department of Radiation Oncology, Guangzhou, China ; 4 Beijing Hospital/ National Center of Gerontology, Department of Breast Cancer Surgery, Beijing, China ; 5 Qilu Hospital of Shandong University, Department of Radiation Oncology, Jinan, China ; 6 Hunan Cancer Hospital/ The Affiliated Cancer Hospital of Xiangya School of Medicine- Central South University, Department of Radiation Oncology, Changsha, China ; 7 Xiangya Hospital- Central South University, Department of Radiation Oncology, Changsha, China ; 8 Xiangya Hospital- Central South University, Department of General Surgery, Changsha, China Purpose or Objective The aim of this study was to develop a widely accepted prognostic nomogram and establish a risk-adapted PMRT strategy based on locoregional recurrence for pT1-2N1M0 breast cancer. Material and Methods A total of 1637 patients with pT1-2N1M0 breast cancer treated at 6 participating institutions were retrospectively reviewed between 2000 and 2016. Of them, 389 (23.8%) patients received PMRT, while 1248 (76.2%) patients did not. A nomogram was developed to predicted local recurrence free survival (LRFS) based on the Cox proportional hazards model. The concordance index (C- index) and a calibration curve were used to determine its predictive and discriminatory capacity. A propensity score- matched (PSM) analyses was performed in risk-adapted The median follow-up of 60.0 months. There was no significant difference between PMRT and no PMRT groups for entire patients. A multivariable analysis of the primary cohort showed that age (≤45 years vs >45 years), tumor site (inner region vs central region vs lateral region), molecular subtype (Luminal A vs Luminal B vs Luminal Her2 model. Results