ESTRO 2021 Abstract Book

S1000

ESTRO 2021

Purpose or Objective Concurrent chemoradiotherapy is the standard non-surgical treatment for stage 3 NSCLC. There is evidence for adjuvant immunotherapy in patients who express more than 1% PDL1. An audit was performed to analyse the uptake and tolerance of adjuvant immunotherapy in eligible patients, after chemoradiotherapy treatments. Materials and Methods Patients were included if they were diagnosed with stage 3 NSCLC with PDL1 of >1%, in West Yorkshire and treated at Leeds Cancer Centre from 01/01/2018 to 01/06/2020. Baseline tumour and patients demographics were analysed. Information on use of immunotherapy, including duration and toxicity were collected. Overall and progression free survival are calculated. Results A total of 66 patients have been included in the study. Patient and tumour demographics are summarised in table 1. Durvalumab was commenced in 38 (58%) of patients. Majority of patients who could not commence Durvalumab was noted to have poor performance status or unresolved toxicity after chemoradiotherapy. 17 patients terminated Durvalumab treatment prior to completion, 7 due to toxicity of treatment (4 pneumonitis, 2 colitis and 1 hepatitis), the remainder due to disease progression. Overall survival was not reached in the study population. One year overall survival is >90%. Patients who received Durvalumab appear to have improved progression free survival (PFS) compared to patients who did not receive Durvalumab (graph 2a), although this is not statistically significant. PFS of patients who received Durvalumab, stratified by PD1 expression, and patients who did not receive Durvalumab, are displayed in graph 2b. Conclusion Chemoradiotherapy followed by adjuvant Durvalumab produces good overall and progression free survival in stage 3 NSCLC patients. Toxicity from Durvalumab appears worse than those demonstrated in the PACIFIC study. PO-1205 Factors predicting pathological response after neoadjuvant chemoradiation in oesophageal cancer K.S. Chufal 1 , I. Ahmad 1 , R. Bajpai 2 , A.A. Miller 3 , R.L. Chowdhary 4 , K. Bhatia 4 , M. Gairola 1 1 Rajiv Gandhi Cancer Institute & Research Centre, Radiation Oncology, New Delhi, India; 2 Keele University , School of Medicine, Staffordshire, United Kingdom; 3 Illawarra Cancer Care Centre , Radiation Oncology, Wollongong, Australia; 4 Rajiv Gandhi Cancer Institute & Research Centre, Radiation Oncology , New Delhi, India Purpose or Objective To develop a model for pathological complete response (pCR) prediction based on pre-operative patient and treatment-related variables, in patients receiving neoadjuvant chemoradiation for oesophagal carcinomas. Materials and Methods We queried our cancer registry to retrieve patients who underwent an esophagectomy between January 2010 and December 2020 and applied the following inclusion criteria: 1. Squamous Cell Carcinoma/AdenoCarcinoma of the thoracic oesophagus and oesophagogastric junction, who received NA-CRT and esophagectomy. 2. All oncological interventions were delivered at our institution. All patients were re-staged as per the AJCC 8th edition and were retrospectively classified as CROSS- Eligible/Ineligible based on the CROSS investigators' criteria. All patients were treated on a 6MV Linear Accelerator, and elective nodal irradiation was not performed. 125 patients received concurrent chemotherapy with weekly Paclitaxel & Carboplatin and others received weekly platinum (n = 105) or weekly Cisplatin & 5-FU (n = 24). We performed a multivariable binary logistic regression with backward elimination on individual patient data to predict pCR following NACRT, based on pre-operative variables. Categorical variables which were entered into the model were: Sex; Clinical T stage (T 0-3 , vs T 4 ); Clinical N stage (N 0 vs N 1 vs N 2-3 ); RT Technique (Conventional vs 3DCRT/IMRT); post hoc assessment of eligibility for CROSS protocol (Yes vs No), and; Chemotherapy Protocol (Paclitaxel-Carboplatin vs All other agents). Continuous variables which were entered into the model were: BMI (Kg/m 2 ); Cranio-Caudal length of Primary disease (cm); EQD 2 ( α / β = 4.9)(Gy 2 ); RT duration (days); Interval between completion of RT and Surgery (days); Package time (interval between starting RT and Surgery)(days). Results A total of 254 patients were included in this analysis. The model's overall accuracy in predicting pathological response was 67.3% (Hosmer & Lemeshow test of significance = 0.268). Out of the 13 variables entered into the model, only three were significantly associated with the probability of achieving a complete pathological response. Male sex (OR = 0.51, 95% CI = 0.30-0.87, p = 0.014) and increasing cranio-caudal length of primary disease (OR = 0.86, 95% CI = 0.76-0.98, p = 0.024) were negatively associated with the probability of achieving pCR, whereas concurrent delivery of weekly Paclitaxel-Carboplatin (OR = 2.94, 95% CI = 1.73-4.97, p < 0.001) was associated with a higher probability of achieving pCR. Digital Poster: Upper GI (oesophagus, stomach, pancreas, liver)