Abstract C098: Integrative multi-omics profiling reveals genomic ancestry-associated molecular features of early-onset colorectal cancer

Abstract

Colorectal cancer (CRC) remains the second leading cause of cancer-related mortality in the United States, with early-onset colorectal cancer (EOCRC) increasing at an alarming rate. Populations underrepresented in molecular oncology studies continue to experience disparities in CRC incidence and clinical outcomes. We integrated genomic, transcriptomic, spatial transcriptomic, and microbiome profiling to identify molecular features associated with genomic ancestry and EOCRC biology. Clinical and molecular data from 205 primary CRC tumor-normal pairs were analyzed using whole-exome sequencing and RNA sequencing. Comparative analyses were performed against 3,578 publicly available non-Hispanic White CRC cases. Visium HD spatial transcriptomics was performed on 14 paired CRC tumors to characterize compartment-specific gene expression across cancer, immune, and stromal regions. Microbiome profiling was conducted on stool samples from 16 patients using 16S rRNA sequencing, and tumor-associated microbial DNA was analyzed from captured tumor DNA. Multi-modal datasets were integrated using a conversational artificial intelligence framework. EOCRC exhibited a higher median Peruvian-in-Lima-like (1KG-PEL-like) genomic ancestry proportion, distinct somatic copy number alteration profiles, recurrent MYC amplification, WNT pathway alterations, and unique transcriptional programs compared with LOCRC and publicly available non-Hispanic White cohorts. Spatial transcriptomics identified compartment-specific activation of CRC-associated pathways across cancer, immune, and stromal regions, linking molecular alterations to the spatial organization of the tumor microenvironment. Microbiome analyses demonstrated differences between stool-derived and tumor-associated microbial communities. AI-assisted integration identified molecular features distinguishing EOCRC from LOCRC and enhanced interpretation across genomic, transcriptomic, spatial, and microbial datasets. Integrative multi-omics profiling provides a comprehensive characterization of EOCRC across multiple molecular scales. These findings identify genomic ancestry-associated molecular features, reveal spatial and microbial heterogeneity within the tumor microenvironment, and demonstrate the value of AI-enabled multi-modal integration for advancing precision oncology and reducing disparities in colorectal cancer outcomes. Francisco G. Carranza, Brigette Waldrup, Sophia Manjarrez, PE-CGS Network. Integrative multi-omics profiling reveals genomic ancestry-associated molecular features of early-onset colorectal cancer [abstract]. In: Proceedings of the 19th AACR Conference on the Science of Cancer Health Disparities; 2026 Oct 18-21; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2026;35(10 Suppl):Abstract nr C098.