1059P - Association between tumor longevity and immune-checkpoint inhibitor outcomes: A retrospective study in head and neck, lung, renal/urothelial cancers

Background

The impact of the duration of tumor persistence within the host – tumor longevity (TL) – on patient (pt) outcomes under immune checkpoint inhibitors (ICIs) is unexplored. We hypothesized that this time parameter may influence ICIs efficacy. We investigated this association in head and neck squamous cell carcinoma (HNSCC), non-small cell lung cancer (NSCLC), and renal/urothelial cancer (R/UC) pts.

Methods

Retrospective, observational study including adult pts with primary/recurrent HNSCC, NSCLC, R/UC receiving ICIs (≥first line) in 2 Italian centers. Endpoints were the association between TL and: 1) progressive disease (PD) vs. complete response (CR)/ partial response (PR)/ stable disease (SD); 2) Common Terminology Criteria for Adverse Events (CTCAE) G≥3 toxicity; 3) overall survival (OS). Univariable/multivariable logistic regression model (for PD, toxicity) and Cox proportional hazards model (for OS) were used. TL, defined as time from cancer histological diagnosis to ICIs start, was included as continuous variable using 3 knots restricted cubic spline, comparing third (Q3) vs. first (Q1) quartiles.

Results

304 pts, diagnosed in 2002-2022, started ICIs in 2014-2022. TL had a right-skewed distribution [median: 15.3 (Q1-Q3, 5.5; 33.1) months]. CR, PR, SD, PD occurred in 16 (5.3%), 90 (29.6%), 65 (21.4%), 133 (43.7%) pts, respectively; G≥3 toxicity was reported in 16 (5.3%) pts and 231 (76.0%) pts died. Multivariable models adjusted for relevant variables (Table) suggested that longer TL was associated with lower PD odds (Odds Ratio [OR] Q3 vs. Q1 = 0.31; 95% CI 0.12; 0.81, p = 0.004), higher G≥3 toxicity odds (OR = 1.27; 95% CI 0.39; 4.12), lower death risk (Hazard Ratio = 0.79; 95% CI 0.49; 1.28). Table: 1059P