Artificial Intelligence Reveals Prognostic TP53 Pathway Alterations in FOLFOX-Treated Early-Onset Colorectal Cancer Among Populations at Risk

Abstract

BackgroundThe incidence of early-onset colorectal cancer (EOCRC; <50 years) continues to rise, with the most rapid increases observed among Hispanic/Latino (H/L) populations. Although the TP53 signaling pathway is a critical driver of colorectal tumorigenesis, its prognostic significance in FOLFOX-treated EOCRC--particularly among H/L patients-- remains poorly defined. MethodsWe analyzed 2,515 colorectal cancer (CRC) cases (H/L = 266; non-Hispanic White [NHW] = 2,249), stratified by ancestry, age at onset, and FOLFOX treatment status. Statistical comparisons were performed using Fishers exact, chi-square, and Kaplan-Meier analyses. To enhance integrative data exploration, we applied AI-HOPE and AI-HOPE-TP53, novel conversational artificial intelligence (AI) platforms that enable natural language-driven analysis of multi-dimensional clinical, genomic, and therapeutic data. ResultsTP53 pathway alterations were detected in 85% of H/L and 83% of NHW patients treated with FOLFOX. Among late-onset (LO) cases, CHEK2 mutations were significantly lower in H/L patients compared with NHW (0.8% vs. 2.6%, p = 0.02), while TP53 mutations were more frequent in LO NHW patients receiving FOLFOX (78.9% vs. 73.4%, p = 0.04). Within LO NHW cases, FOLFOX-treated patients exhibited an enrichment of TP53 mutations (73.4% vs. 68.1%, p = 0.02) and lower frequencies of ATM (7.2% vs. 12.1%, p = 0.001) and CDKN2A (1.0% vs. 2.5%, p = 0.03) mutations compared to non-FOLFOX-treated counterparts. Across all cohorts, missense mutations represented the predominant alteration type. Remarkably, in EO H/L patients treated with FOLFOX, TP53 pathway alterations were associated with significantly improved overall survival (p = 0.014), an effect not observed in other subgroups. ConclusionsTP53 pathway alterations may serve as an ancestry- and treatment-specific biomarker of favorable prognosis in FOLFOX-treated EOCRC H/L patients. The application of AI-driven integrative analytics through AI-HOPE platforms accelerated biomarker discovery and underscores the potential of conversational AI to advance precision oncology and health equity.