ESTRO 2020 Abstract Book

S141

ESTRO 2020

Conclusion QA results of the LungART trial demonstrated high compliance with the protocol. Future LungART data will provide insights on the relationship between RT QA and clinical outcomes and better definition of the clinical target volume (CTV) after analysis of patterns of locoregional recurrence. PH-0281 Outcomes of re-irradiation & repeat radiotherapy in NSCLC: A propensity matched analysis L. Sandhu 1 , A. McWilliam 1 , H. Mistry 2 , D. Woolf 3 , C. Faivre- Finn 1 , C. Golby 4 , A. Abravan 1 , M. Van Herk 1 , G. Price 1 , A. Salem 1 1 University of Manchester, Division of Cancer Sciences, Manchester, United Kingdom ; 2 University of Manchester, School of Pharmacy, Manchester, United Kingdom ; 3 Christie NHS Foundation Trust, Clinical Oncology, Manchester, United Kingdom ; 4 Christie NHS Foundation Trust, Radiotherapy Physics, Manchester, United Kingdom Purpose or Objective Non-small cell lung cancer (NSCLC) locoregional recurrence and second primary are common after (chemo)radiotherapy could be treated with re-irradiation (same area) and repeat radiotherapy (different area), respectively. Our aims were to investigate the outcomes of NSCLC patients treated with re-irradiation and repeat radiotherapy and compare to a matched cohort. Material and Methods NSCLC patients who received two curative-intent thoracic radiotherapy courses in the Christie NHS Foundation Trust (Manchester, UK) [FC(NC1] were retrospectively analysed (2013-2019). Radiotherapy plans were rigidly co- registered, if available, to quantify treatment field overlap and differentiate re-irradiation from repeat radiotherapy. Re-irradiation was designated if >0.5cm 3 of the thoracic region received a radiotherapy dose higher than the maximum dose delivered in either of the two radiotherapy plans. A NSCLC patient group who received a single course of curative-intent thoracic radiotherapy was utilized to generate a matched cohort based on age, gender, radiotherapy type (stereotactic ablative radiotherapy (SABR) vs fractionated intensity modulated radiotherapy (IMRT)), TNM stage and ECOG performance score. Logistic regression was used to create a propensity score model. The primary endpoint was overall survival (OS). Results 79 NSCLC patients (median age: 75) underwent two curative-intent thoracic radiotherapy courses (retreatment cohort). The second radiotherapy treatment (RT2) was SABR in 27 and fractionated IMRT in 52 patients. The tumour stage at RT2 was: stage I (n=47), stage II (n=7), stage III (n=7), stage IV (n=3) and unknown (n=15). The median prescribed dose (peaked for SABR) for the RT1 and RT2 was 57Gy (range: 45-88) and 57Gy (range: 54-60), respectively. Table 1 shows characteristics for the retreatment and matched cohort (n=77). The median OS (measured from RT1) for the retreatment cohort was 51m (95% CI: 45-58); fig1A . The median OS was 51m (95%CI: 49- NA) for patients treated with re-irradiation (n=25) and 58m (95%CI: 51-NA) for patients treated with repeat radiotherapy (n=28). After adjusting for lead time bias (OS measured from RT2), patients who underwent retreatment within 1y of the first radiotherapy had worse OS (median: 20m, 95% CI: 12-36) compared to patients where the inter- radiotherapy gap was >1y (median: 25m, 95% CI: 15-55); p =0.013 ( fig1B ). The type of radiotherapy and

Nodal dissection was performed according to lobar specific recommendations in most pts (see figure). Thirty-six pts (25%) had sampling 36 (25%), 32 (23%) a selective dissection and 73 (51%) a systematic dissection (1 patient was not evaluated). Even if all pts had a theoretical R0 resection, after revision, R0 resection was confirmed in 41 pts (29%), but the resection was uncertain in 58 (40.8%), microscopically incomplete (R1) in 42 (29.5%), and macroscopically incomplete (R2) in 1 pts (0.7%). For the RT QA, 87% of the radiotherapy treatment plans were adequate, and 13% (19) had major deviations including dose to the target volume (53%) and dose constraints to the heart (37%). The recommended target volume in the LungART trial depended on the primary tumor location and the involved LN stations. Ninety-seven (68%) of treatment plans had adequate mediastinal coverage, 34 (24%) too large coverage and 6 (8%) was marginally covered. Among the 142 pts analyzed, ipsilateral stations 4 and 10, and station 7 were included in 134 (94%) of cases. The involved and irradiated stations are demonstrated in the following figure: