ESTRO 38 Abstract book

S283 ESTRO 38

immunosuppressive barriers. It is still unclear how each fractionation protocol can modulate the immune microenvironment. Nevertheless, it is essential to determine the best RT fractionation protocol to associate with checkpoint inhibitors (CPI). Currently, in the setting of localized rectum cancer, RT can be delivered before surgery according to two validated schemes with a different dose per fraction. Clinical studies are ongoing to evaluate CPI (such as anti-PD-L1) in association with RT. However, only few trials aim to optimize the RT fractionation to improve efficacy of these associations. Here we wanted to characterize different fractionation protocols effect on immune response and associate them with CPI. Material and Methods Mice bearing subcutaneous CT26 colon tumors were irradiated using SARRP device according to different radiation schemes (1, 3 or 18 fractions) with a same biologically effective dose. The mice were monitored for tumor growth. The radiation immune response (lymphoid, myeloid cells, lymphoid cytokines and immune checkpoints) was monitored by flow cytometry at different time points after treatment . The same fractionation protocols were performed with or without CPI against immune checkpoints modulated by RT. Results In absence of CPI, we showed that 18x2Gy and 3x8Gy induced the longest tumor growth delay compared to 1x16.4Gy. While 3x8Gy and 1x16.4Gy induced a lymphoid response (CD8 + T cells, Regulators T cells (Treg) and natural killer cells); 18x2Gy induced a myeloid response (MDSC, neutrophils, tumor-associated macrophages 2). The secretion of granzyme B by CD8 + T cells and interferon g by CD4 + T cells was more importantly increased with 3x8Gy. The expression of PD-L1 by tumor cells was moderately increased by RT but the most durably with 18x2Gy. While TIGIT expression by CD4 + , CD8 + T and NK cells was increased or maintained by 3x8Gy compared to control group, it was decreased or maintained by 18x2Gy and 1x16.4Gy. There was a synergistic effect of any CPI with RT whatever the dose per fraction compared to RT and IgG. The 18x2Gy protocol did not seem to benefit from anti-TIGIT when associated with anti-PD-L1. RT was dramatically more effective with 3x8Gy compared to all the other treatments schemes when associated with anti-TIGIT and anti-PD-L1 with more than 90% of the tumor in complete response. Conclusion Each fractionation scheme induced different lymphoid and myeloid responses as well as various modulation of PD-L1 and TIGIT expression. Furthermore, 3x8Gy was the most effective protocol when associated with anti-PD-L1 and anti-TIGIT. This is the first study combining RT and anti- TIGIT with promising results; further studies are warranted. In order to describe mechanisms, which may explain differential effects of the different fractionation protocols on immune response, RNA sequence-analysis is still ongoing. PV-0538 Prostaglandin related distinct regenerative activities in hair follicles following radiation injury S. Lai 1,2 , W. Huang 3 , S. Chen 1 , S. Lin 1,4,5 1 National Taiwan University, Institute of Biomedical Engineering- College of Medicine and College of Engineering, Taipei, Taiwan ; 2 National Taiwan University Hospital and College of Medicine, Division of Radiation oncology- Department of Oncology, Taipei, Taiwan ; 3 National Taiwan University, Institute of Biomedical Engineering- College of Medicine and College of Engineering-, Taipei, Taiwan; 4 National Taiwan University, Research Center for Developmental Biology and Regenerative Medicine, Taipei, Taiwan; 5 National Taiwan University Hospital and College of Medicine, Department of Dermatology, Taipei, Taiwan

HPV negative. In a propensity-matched analysis adjusted for age, sex, Charlson-Deyo score, systemic therapy use, treatment facility, income, ethnicity and race; HPV positive patients had significantly better OS, but only in patients with locally advanced disease (T3-4 or node positive; HR = 0.77 (95% CI: 0.64-0.93), p=0.005), whereas for patients with early stage disease (T1-2 and node negative) there was no association with OS (HR = 1.08 (95% CI: 0.83-1.42), p=0.56). Similar results were found in unmatched multivariable analysis (Locally advanced disease: HR: 0.79; p=0.005 and early-stage disease HR: 1.12; p=0.34). The table shows the results from the propensity-matched multivariable analysis and the figure the propensity-matched unadjusted Kaplan-Meier curves, illustrating HPV infection as a prognostic marker in locally advanced disease. The multivariable model for radiation dose showed a clear dose-response for all patients with decreasing HRs for increasing total RT dose up to 55 Gy (compared to total RT doses less than 40 Gy, 40-45 Gy- HR: 0.71; p<0.001; >45-50 Gy- HR: 0.74; p<0.001; >50-55 Gy- HR: 0.60, p<0.001; >55-60 Gy- HR: 0.69; p<0.001; >60-65 Gy- HR: 0.78; p=0.003; >65 Gy- HR: 0.84; p=0.11). A stronger dose-response was seen for locally advanced patients who were HPV negative (HR: 0.43 for >50-55 Gy), compared to those that were HPV positive (HR: 0.72 for >50-55 Gy).

Conclusion We found that HPV infection is a positive prognostic marker for patients with locally advanced anal cancer, but not for those with early stage disease. We further found that higher radiation dose up to 55 Gy was associated with better survival, especially for locally advanced disease in HPV negative patients, and less so in those with HPV positive disease. PV-0537 Optimized fractionated radiotherapy with anti-PD-L1 and anti-TIGIT: a promising combination M. Grapin 1 , E. Limagne 2 , C. Richard 2 , R. Boidot 2 , M. Thibaudin 2 , V. Morgand 1 , G. Créhange 1 , F. Ghiringhelli 2 , C. Mirjolet 1 1 Centre Georges-François Leclerc, Radiation Oncology, Dijon, France ; 2 Centre Georges-François Leclerc, Platform of Transfer in Biological Oncology, Dijon, France Purpose or Objective Radiotherapy (RT) is able to induce an immunogenic antitumor response, but also to increase some