Abstract 3356: Construction of a 23-gene risk scoring system to predict disease recurrence and chemotherapy response in mucinous colorectal cancer

Abstract Backgrounds: Mucinous adenocarcinoma (MuC), a unique subtype of colorectal cancer (CRC), exhibits distinct molecular features and is associated with poorer outcomes compared to non-mucinous adenocarcinoma (NMuC). Standard CRC treatments often overlook the molecular diversity within CRC, especially in stage II, where the benefit of adjuvant chemotherapy remains debated. Biomarkers are needed to improve risk assessment and guide tailored treatments. Methods: We conducted molecular profiling of 259 CRC samples, including MuC and NMuC subtypes, to identify MuC-specific gene signatures. A 23-gene risk score (MuC-RS) was developed and validated in four independent cohorts (n = 1, 157). Prognostic significance of MuC-RS was assessed using various statistical methods, including Kaplan-Meier survival analysis, log-rank tests, and Cox regression. Results: In the discovery cohort of fresh-frozen tissues, unsupervised clustering revealed distinct gene expression patterns differentiating MuC from NMuC. MuC showed notable enrichment in fibroblast-associated genes, EMT pathways, and mucin glycosylation processes. The MuC-RS effectively stratified patients into high-risk (MuC-H) and low-risk (MuC-L) categories. Multivariate analysis confirmed MuC-RS as an independent prognostic factor. Importantly, stage II MuC-L patients showed limited response to standard chemotherapy but increased responsiveness to immune checkpoint inhibitors (ICIs), associated with higher tumor mutation burden and immune activation. MuC-RS's prognostic utility was further validated through comparisons with consensus molecular subtypes. Conclusion: MuC-RS offers a promising tool for predicting recurrence and informing personalized treatment strategies in CRC, potentially enhancing risk stratification and supporting novel therapies, such as ICI, particularly for stage II MuC patients. Grant Support: This research was supported by the Bio & Medical Technology Development Program of the National Research Foundation (NRF) funded by the Korean government (MSIT) (2022R1A2C2008480 to S.K. Kim, 2018R1C1B6008571 and 2022R1F1A1074317 to J.L. Lee). This study was also supported by grants from the Korea Research Institute of Bioscience and Biotechnology (KRIBB) Research Initiative Program (KGM5192423). Citation Format: Jee-Woo Seo, Jae-Yoon Kim, Ye-Jin Ha, Seung-Woo Baek, Jong-Lyul Park, Sun-Woo Lee, Jong-Lyul Lee, Seon-Young Kim, Seon-Kyu Kim. Construction of a 23-gene risk scoring system to predict disease recurrence and chemotherapy response in mucinous colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3356.