Neoadjuvant chemoradiation (CRT) for locally advanced rectal cancer (LARC) with or without oxaliplatin (OX): Individual patient data (IPD) meta-analysis of three randomized controlled trials (RCTs) with subgroup analyses of age cohorts.

4074 Background: Neoadjuvant CRT with fluoropyrimidine (FP) is standard treatment for LARC, which is increasing in younger patients (pts). RCTs examining the addition of OX are still controversial. A post hoc analysis of the CAO/ARO/AIO-04 trial showed significant benefit in pts < 60y. We hypothesised that younger pts with LARC might have improved outcomes with OX-CRT. Methods: Systematic review and IPD meta-analysis were performed. Data from 3 RCTs (CAO/ARO/AIO-04, ACCORD-12, PETACC-6) testing the addition of OX to standard FP-based CRT in LARC were available (of 9 RCTs identified). Primary endpoint: disease-free survival (DFS), secondary endpoints: pathologic complete response (pCR), overall survival (OS). Analyses were by intention to treat (ITT), stratified by trial. Age cut-offs were 60y and 50y. Given the focus on young age a multivariate analysis evaluating all possible confounders was not intended in the current work. Results: IPD from 2914 pts were included (48.5% of available literature). Median age was 63; 70% were male; 79% had a performance status = 0; 72% were stage ≥III. In ITT (Hazard Ratio [HR] 0.88, 95%CI 0.77-1.01, p = 0.06), DFS was not significantly improved by the addition of OX (Table). In < 60y (n = 1166, 40% total), DFS was significantly improved by OX (HR 0.77, 95%CI 0.62-0.96, p = 0.02). In < 50y (n = 350, 12% total) there was a numerically better DFS, although not significant (HR 0.73, 95%CI 0.49-1.08, p = 0.12). Interaction test between age and DFS was non-significant (60y p = 0.11; 50y p = 0.44). In ITT, OX increased pCR from 13% to 16% (Odds Ratio [OR] 1.28, 95%CI 1.04-1.57, p = 0.024 [stratified by trial]), without significant interaction with age (60y p = 0.11, 50y p = 0.74). No OS benefit was demonstrated (HR 0.97, 95%CI 0.82-1.15, p = 0.75). Conclusions: This first IPD meta-analysis of three RCTs evaluating the addition of OX to CRT did not show significant interaction of OX with age. However, we confirm a signal for DFS benefit in pts < 60y and a non-significant increment in DFS in < 50 y although this analysis may be underpowered. Stage-stratified analyses and feasibility/toxicity data in age cohorts will be presented. [Table: see text]