Gut dysbiosis as a potential guide for immunotherapy (dis)continuation after 2 years in non-small cell lung cancer: A mono-institutional, multi-omic assessment.

Abstract

2654 Background: Although most phase II and III clinical trials have set the duration of immune checkpoint blockers (ICB) for advanced non-small cell lung cancer (NSCLC) at two years, there remains uncertainty regarding the feasibility and safety of discontinuing treatment after this period. Of note, gut microbial taxonomic profiling prior to starting immunotherapy shows promise as biomarker for predicting ICB response. Here, we recommend integrating multi-omics approaches over time (24 months -mo-) to inform clinical decision-making and guide personalized treatment strategies. Methods: Pts completing 18 to 24 mo of ICB treatment between July 2016 and January 2023 were identified and enrolled (NCT04567446) at Gustave Roussy. Clinical factors influencing treatment (dis)continuation were assessed at 24 mo. Multi-omic analyses, including gut-based biomarkers (TOPOSCORE by whole genome sequencing), PET-FDG imaging, and ctDNA, were proposed at this timepoint. Key outcomes, including overall survival (OS) and progression-free survival (PFS) rates, were analyzed. Results: Among 123 advanced NSCLC pts treated for ≥18 mo, 35 (28,5%) completed 24 mo, with 31 included in the analysis (4 excluded due to PD). Of these, 68% continued ICB, while 32% stopped between 23.5 and 29.7 mo, mainly based on the physician or the patient decision. Clinical characteristics were comparable between the 2 groups (Table 1). After a median follow-up of 59.1 mo, no significant OS and PFS differences were observed between pts who discontinued and those who continued (OS p=0.9012, PFS p=0.3715). Among the multi-omic assessments performed at 24 mo, only gut-based biomarkers appeared to be conditionally associated with PFS24 rates. The proportion of long-responders (progression-free at 24 mo) was higher among those with a favorable gut composition compared to those with harmful composition (81% vs 44%, respectively, p=0.0870). Conclusions: Our results suggest that multi-omics approaches may help safely discontinue ICB after two years of treatment. In this context, multi-institutional validation and the implementation of a translational multi-omic algorithm, including gut-based biomarkers, could provide insight into the optimal duration of ICB therapy beyond the predefined 24-mo period. Clinical trial information: NCT04567446 . Clinical characteristics of the cohort (n=31). Characteristics Cessation group Pursuit group p-value* N = 10 N = 21 Gender - no. (%) Male 5 (50) 12 (57) 0.7366 Female 5 (50) 9 (43) Age years - median (range) 61 (39-68) 62 (43-77) ECOG performance status - no. (%) 0-1 9 (90) 15 (71) 0.2044 2 1 (10) 6 (29) PD-L1 expression - no. (%) <1% 0 2 (13) 0.5749 ≥1%-<50% 3 (30) 2 (13) ≥50% 7 (70) 11 (73) unknown - 6 Treatment regimen - no. (%) Chemoimmunotherapy 3 (30) 2 (10) 0.1907 Monoimmunotherapy 7 (70) 19 (90) Line of treatment - no. (%) First 7 (70) 10 (48) 0.7332 ≥ Second 3 (30) 11 (52) *Chi-Square test.