ESTRO 38 Abstract book

S390 ESTRO 38

resulted in more extensive RT than recommended by the SG. Future analyses will be aimed at evaluating the outcome of these patients in relation to risk factors and the actual RT given. PO-0760 Heterogeneity of Radiosensitivity, Recurrence, and PD-L1 in Breast Tumor Single Cell RNA-Seq Data B. Jang 1 , W. Han 2 , I.A. Kim 3 1 Seoul National University Hospital, Radiation Oncology, Seoul, Korea Republic of ; 2 Seoul National University- College of Medicine, Surgery, Seoul, Korea Republic of ; 3 Seoul National University- College of Medicine, Radiation Oncology, Seoul, Korea Republic of Purpose or Objective Tumor heterogeneity has been recently revealed in an advent of single-cell RNA sequencing. At a single cell level, heterogeneity of intrinsic tumor radiosensitivity, recurrence-risk, and their relationship remain unknown in breast cancer. In this study, we investigate the heterogeneity of intrinsic tumor radiosensitivity and recurrence risk in breast cancer patients. Material and Methods We investigated single cell mRNA sequencing data from GEO database (GSE75688) to profile 281 primary breast tumor cells. Subtypes of each tumor cell were classified by using PAM50 gene signature. OncoType DX and Endopredict gene signatures were used to estimate enrichment score and to assess the recurrence risk for each cell. To explore radiosensitivity, Gene Set Variation Analysis (GSVA) were performed across cells using radiosensitivity gene signatures including both RSI (Radiosensitivity Index) and 31-gene. PD-L1 expressed tumor cells were profiled as well. Results GSVA analysis showed that mean recurrence risk score of cells were different according to PAM50 breast cancer subtype, and cells were classified as basal subtype demonstrated higher risk recurrence score compared to those of other subtypes (P<0.001). Heterogeneity of recurrence risk was observed more frequently in luminal A subtype compared to basal/HER2 positive subtypes. Radioresistant tumor cells, representing high enrichment score of RSI, were more commonly found in basal type compared to those of other subtypes (P=0.029). RSI score was varied among cells from patient to patient, indicating the heterogeneity of intrinsic tumor radiosensitivity. There were significant correlations in a single cell resolution between RSI and 31-gene radiosensitivity (Pearson's r=0.25), between RSI and Endopredict risk score (r=0.26), and between OncoType DX and Endopredict enrichment scores (r=0.61), respectively. Most of the PD- L1 expressed cells were classified into high-risk recurrence group (92.8 % for Endopredict and 100% for OncoTypeDX). Conclusion Tumor cells having various subtype or recurrent-risk coexisted regardless of individual pathologic subtype in a single cell level. Most radioresistant tumor cells showed basal subtype or PD-L1 expression. PD-L1 expressed cells were enriched in the high-risk or basal/HER2/luminal B subtype. Our results demonstrated the heterogeneity of intrinsic tumor radiosensitivity and recurrence risks in breast cancer patients. PO-0761 Hypofractionated whole breast irradiation safety after breast-conserving surgery for young patients I. Meattini 1 , Y. Kirova 2 , C. Saieva 3 , L. Visani 1 , E. Olmetto 1 , V. Salvestrini 1 , J. Kim 4 , W. Jung 4 , I. Desideri 1 , A. Fourquet 2 , P. Poortmans 2 , L. Livi 1 , K. Kim 4 1 University of Florence - Azienda Ospedaliero Universitaria Careggi, Radiation Oncology Unit - Oncology Department, Florence, Italy ; 2 Institut Curie, Department of Radiation Oncology, Paris, France ;

II breast cancer (BC) are unclear. Therefore, we conducted a prospective cohort study from 2011-2014 in the Netherlands (RAPCHEM: NCT01279304), to evaluate the 5-year locoregional recurrence rate for BC patients treated with PST, surgery, and PORT given according to strict study-guidelines (SG). The aim of the current analysis is to evaluate adherence to these SG. Material and Methods Between January 2011 and January 2015, all patients with cT1-2N1 (four or more suspicious nodes at imaging excluded) BC, treated with PST in the referral area of 18 Dutch RT centres were included. Trained clerks of the Dutch Cancer Registry registered tumour and treatment characteristics. Surgery consisted of breast conserving therapy (BCT) or mastectomy with a sentinel node (SN) biopsy and/or an axillary lymph node dissection (ALND). SG recommended whole breast RT for patients treated with BCT and adapted chest wall/nodal RT based on three risk categories, mainly based on the ypN status: 1. Low-risk (ypN0): breast RT in case of BCT/no chest wall RT in case of mastectomy; 2. Intermediate-risk (ypN1): chest wall/breast RT; 3. High-risk (ypN2): chest wall/breast RT with regional RT (level I-IV). In 2013, an amendment was developed taking into account new insights in axillary treatment: in case of a positive SN, additional ALND could be replaced by axillary RT. Results From 2011-2014 we included 853 patients: 291 in the low- , 378 in the intermediate-, and 184 in the high-risk groups, respectively. 160 (19%) patients did not undergo an ALND. Overall, 63% of the patients were treated according to the SG (66% after BCT, 60% after mastectomy). Almost 10% received RT to fewer target volumes than recommended in the SG, and almost 25% received RT to more extensive target volume (Table). These percentages remained stable over the inclusion period. The largest variation was seen in the intermediate risk group (ypN1), where only 54% were treated according to the SG and where 16% received less and 27% received more RT than the SG. In half of the patients not treated according to the SG, the reason for deviation was unclear; in the other half, 60% deviated on an individual basis, whereas 40% were treated according to the local institutional protocol, since the radiation- oncologists did not agree with the SG (Figure).

Conclusion The SG of the RAPCHEM study were followed in only 63% of patients. The largest variation in applied post-operative RT was observed in patients with ypN1 disease (intermediate risk group); deviation of the SG more often

Made with FlippingBook - Online catalogs