Abstract 6165: Identification of biomarkers of early onset colorectal cancer from transciptome and whole genome data

Abstract Colorectal cancer (CRC) is not only the third most commonly occurred cancer worldwide, but it is the leading cause of cancer deaths in 2018. However, the overall incidence of CRC and death from CRC has declined since 1998, the case of early-onset CRC (EOCRC) defined as CRC diagnosed under age 50 has been increasing. Including the etiology of EOCRC, the reasons underlying this upward trend are not well understood. Moreover, some suggested that EOCRC patients have difficulty in early diagnosis compared to late-onset colorectal cancer (LOCRC). For this reason, EOCRC patients tend to be diagnosed at advanced stages. In addition to the clinical feature, EOCRC tumors have different features in pathology and molecularity compared with LOCRC tumors. With these distinct characteristics of EOCRC, there is necessary to characterize EOCRC compared to LOCRC patients. In this study, patients under 50 years (from hereon referred to as ‘‘early-onset’’) and over 70 (‘‘late-onset’’) were considered for comparison. Whole transcriptome sequencing (WTS) for profiling gene expression and whole genome sequencing for mutations were carried out with tumor tissues of early-onset (n=49) and late-onset (n=50) CRC patients. Subsequently, we performed bioinformatic analysis to identify genomic and transcriptomic features. For the precise selection of differentially expressed genes, we selected constantly differentially expressed genes (DEG) in TCGA COAD and READ data. Genes up-regulated in both data set were CPS1, CCL19, CHGB, GLDC, WASF3, RPL39L and MAP9. The down-regulated were PRSS33, ANPEP, DKK4 and NTS. To identify whether CMS classification had different feature in each type if age onset was different, we classified CMS subtypes using transcriptome data. The ratio of each CMS subtypes both early and late-onset group had no significant difference. In terms of somatic mutations reported for driving colorectal cancer, three genes (ATM, BCL9L, NF1) had significantly mutated in early-onset group. Moreover, Non-synonymous mutation frequency had no significant difference while total mutation frequency had significantly difference between early and late-onset group. Lastly, Mutational signature had difference in both groups, assuming that the early-onset group had been through different event attributing to mutational changes. Although these findings should be further validated by detailed functional experiments, our genomic and transcriptomic analysis provide additional insights into understanding of young CRC patients. Citation Format: Yejin Ha, Yun-Jae Shin, Ka Hee Tak, Jong-Lyul Park, Jeong-Hwan Kim, Jong Lyul Lee, Seon-Young Kim, Chan Wook Kim. Identification of biomarkers of early onset colorectal cancer from transciptome and whole genome data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6165.