Deciphering RTK-RAS and MAPK Pathway Dependencies in Gemcitabine-Treated Pancreatic Ductal Adenocarcinoma Through Conversational Artificial Intelligence

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal malignancy marked by substantial molecular heterogeneity and variable response to gemcitabine-based therapy. While KRAS mutations are nearly universal, the broader RTK-RAS and MAPK signaling architecture and its relationship to treatment response remain incompletely defined. We conducted an integrative clinical-genomic analysis of 184 PDAC tumors stratified by age at diagnosis and gemcitabine exposure, interrogating somatic alterations across curated RTK-RAS/MAPK gene sets. Conversational artificial intelligence agents (AI-HOPE-RTK-RAS and AI-HOPE-MAPK) enabled dynamic cohort construction and pathway-level analyses, with findings validated using standard statistical methods. In late-onset PDAC, ERBB2 and RET mutations were significantly enriched in gemcitabine-treated tumors. Early-onset cases demonstrated differential enrichment of CACNA2D family alterations in non-treated tumors and higher frequencies of FLNB and TP53 mutations in treated disease. Importantly, late-onset patients not treated with gemcitabine who lacked RTK-RAS or MAPK alterations exhibited significantly improved overall survival. These findings reveal age- and treatment-dependent pathway dependencies beyond canonical KRAS status and support a precision oncology framework in PDAC. Conversational AI facilitated rapid, multidimensional clinical–genomic integration to uncover clinically relevant signaling substructures.

Ähnliche Arbeiten