Patterns of Failure, Volume of Disease Progression, and Subsequent Ablative Management in Locally Advanced Non-Small Cell Lung Cancer (LA-NSCLC) Treated with Definitive Chemoradiation and Consolidation Immune Checkpoint Inhibitors (ICI)

Abstract

For patients (pts) with LA-NSCLC treated with chemoradiation and consolidation ICI (CRT+ICI), the patterns of failure (POF) and volume of disease progression (PD) are not well characterized. The primary objective of this study was to classify POFs, the frequency of low volume relapse (LVR), and identify pts eligible for further ablative therapy. We retrospectively identified pts with unresectable stage III NSCLC treated with CRT+ICI between October 2017 and December 2021 at a single institution. Site of first failure was classified as locoregional (LRF), distant (DF), or synchronous LRF + DF. Any LRF was subclassified as in field (IFF; PD within 90% isodose line), marginal (MF; within 50% isodose line) or out of field (OOF; outside of 50% isodose line). LVR was defined as < 3 discrete sites of PD in any number or location of organs. Pts with distant LVR were considered to have oligometastatic relapse. Ablative candidates were defined as pts with < 3 discrete sites of PD amenable to further RT or surgery. Cumulative incidence of PD was calculated with death as a competing risk. Progression free survival (PFS) and overall survival (OS) were calculated from the end of RT and assessed via Kaplan Meier. Multivariable Cox modeling was used to assess correlation of pt characteristics and time-to-event outcomes. Logistic regression was used to predict variables associated with LVR. A total of 229 pts received CRT+ICI. Median follow up was 39 months and 119 pts experienced PD. Median PFS and OS were 18.4 and 34.5 months, respectively. Of pts with PD, 71 (60%) had DF, 28 (24%) had LRF+DF, and 20 (17%) had LRF. Of pts with any LRF, 28 (57%) had IFF, 10 (21%) had MF, and 10 (21%) had OOF. Estimated 1-year cumulative incidence of LRF, DF, and LRF+DF were 9.3% (95% CI 4.5-16), 39% (95% CI 31-48), and 19% (95% CI 12-27), respectively. A total of 63 (53%) pts had LVR. In pts with LVR, 19 (30%) had isolated thoracic relapse and 44 (69%) had oligometastatic relapse. Most oligometastatic disease was intracranial (22 metastases, 44%). Pts with LVR had a longer median OS vs pts with high volume relapse (37.4 vs 15.2 months, p<0.001). At time of PD, 56 (47%) pts were candidates for further ablative therapies. Subsequent anticancer therapies were local therapy alone (35%), local and systemic therapy (16%), systemic therapy alone (36%), or no therapy (13%). On multivariable analysis, LVR (HR 0.39; 95% CI 0.21-0.73, p = 0.003) and longer receipt of ICI (HR 0.96; 95% CI 0.95-0.98; p<0.001) were associated with improved survival while squamous histology (HR 2.26; 95% CI 1.18-4.32; p = 0.039) was associated with worse survival. Longer receipt of ICI was the only variable predictive for the development of LVR (OR 1.03; 95% CI 1.01-1.05; p = 0.004). This is the largest real-world series reporting POF after CRT+ICI for stage III NSCLC. Approximately half of pts experience LVR and are candidates for further ablative therapy. Further data are needed to define optimal treatment strategies for pts with LVR after CRT+ICI.