Characterization of Korean Colorectal Cancer Reveals Novel Driver Gene and Clinically Relevant Mutations

Abstract

Colorectal cancer (CRC) ranks as the third leading cause of cancer-related deaths worldwide, characterized by genomic heterogeneity arising from ethnic and interindividual differences. Producing region-specific data to characterize ethnic-specific somatic mutations is essential for advancing CRC research. Additionally, accurate somatic mutation detection requires paired tissue analyses to account for interindividual diversity. This study aims to highlight the importance of ethnic diversity in shaping CRC's genomic landscape and emphasize the necessity for region-specific data to refine diagnostic and therapeutic approaches. This study emphasizes the need for region-specific data by analyzing an unprecedented 197 paired samples from the Korean CRC cohort through whole-genome sequencing. We identified 78 potential driver genes. Notably, CBWD5, LRRIQ3, TRIM64B, SPINK5, and ZNRF2 were linked to recurrence, presenting potential therapeutic targets. Our analysis revealed 30 mutational hotspots, with significant variants in KRAS (25%, G12A, G12D, G12V), MAP1A (12%, V2300G), and TP53 (8%, R175H). We identified a significant co-occurrence between KRAS 12 mutation and PIK3CA 545 mutation. Our findings demonstrate potential driver genes and mutational hotspots associated with CRC patient, characterizing the mutational landscape related to clinical characteristics. Significantly advancing our understanding of CRC's heterogeneous nature, this study lays a solid foundation for devising more efficacious management strategies.

Keywords: colorectal cancer; microsatellite instability; mutational hotspots; tumor mutational burden; whole‐genome sequencing.

FIGURE 1

Alterations in 10 oncogenic pathways of CRC in 197 Koreans. Each pathway diagram shows the proportion of patients harboring mutations in the indicated genes, with red shading representing higher mutation frequencies. Percentages beside each pathway denote the proportion of patients with at least one mutation in that pathway. Solid and dotted lines indicate the cell and nuclear membranes, respectively; arrows indicate activation, bars indicate inhibition, and dotted arrows denote indirect regulatory effects.

FIGURE 2

Somatic CNVs in Korean CRC patients. (A) Circos plot showing genome‐wide CNV patterns. The outermost ring represents chromosomes; red and blue dots indicate amplification and deletion degrees, respectively. Inner heatmaps display amplification in red and deletion in blue, with red‐labeled genes showing the highest CNV levels. (B) Boxplots showing CNV frequencies across clinical variables: sex, age, MSI status, KRAS mutation, relapse status, and tumor location. Gains (upper, red) and losses (lower, blue) are indicated. (C) CNV distribution across 10 oncogenic pathways. Red and blue bars represent gain and loss frequencies in key genes of major signaling pathways.

FIGURE 3

Characteristics of somatic mutations in hypermutated and nonhypermutated CRC. (A) Pathway diagrams showing mutation frequencies in individual genes. Red indicates a higher proportion of mutated patients; solid and dotted lines represent the cell and nuclear membranes, respectively. Arrows indicate activation, bars indicate inhibition, and dotted arrows denote indirect effects. Each cell represents one patient, and red shading indicates at least one alteration in the pathway. (B) Oncoplot of somatic mutations by group. Genes are ordered by pathway (Hippo, PI3K–AKT, RTK–RAS, and dMMR‐related genes). Mutation types are color coded, and corresponding clinical characteristics (TMB, MSI status, age, sex, tumor location, metastasis, relapse) are shown below. (C) Somatic interactions between genes. Cyan indicates co‐occurrence, and brown indicates mutual exclusivity. (D) Mutational signatures identified in the hypermutated group. The Y‐axis shows the fraction of mutation types, and the X‐axis indicates 96 base substitutions. (E) Mutational signatures identified in the nonhypermutated group. The Y‐axis shows the fraction of mutation types, and the X‐axis indicates 96 base substitutions.

FIGURE 4

Somatic mutational characteristics of KRAS‐mutated and non‐KRAS CRCs. (A) Pathway diagrams illustrating mutation frequencies in individual genes. Red shading indicates a higher proportion of patients with mutations. Solid and dotted lines represent the cell and nuclear membranes, respectively; arrows indicate activation, bars indicate inhibition, and dotted arrows denote indirect effects. Each cell represents one patient, and red indicates the presence of at least one alteration in the pathway. (B) Oncoplot showing somatic mutation profiles for each group. Genes are ordered by pathway: WNT, PI3K–AKT, and RTK–RAS. Mutation types are color coded, and the lower panel displays clinical characteristics, including age, MSI status, sex, tumor location, metastasis, and relapse status. (C) Somatic interaction matrix of recurrently altered genes. Cyan indicates co‐occurrence, whereas brown indicates mutual exclusivity. (D) Mutational signatures identified in the KRAS‐mutated CRC group. The Y‐axis shows the fraction of mutation types, and the X‐axis represents 96 base substitution contexts. (E) Mutational signatures identified in the non‐KRAS CRC group. The Y‐axis shows the fraction of mutation types, and the X‐axis represents 96 base substitution contexts.

FIGURE 5

Sex‐based landscape of somatic mutations in CRC patients: comparison of 127 males versus 70 females. (A) The oncoplot depicts the top 10 most frequent variant genes in each group. Each color is indicated according to the mutation type. (B) Somatic interactions between groups. Mutually exclusive or co‐occurring sets of genes were detected using a pairwise Fisher's exact test to detect significant gene pairs. (C) Mutational signatures plots known as SBS in 127 male patients with CRC. (D) Mutational signatures plots known as SBS in 70 female patients with CRC.

FIGURE 6

Early‐onset‐based landscape of somatic mutations in CRC patients: comparison of 186 over 40 years old versus 11 under 40 years old. (A) The oncoplot depicts the top 10 most frequent variant genes in each group. Each color is indicated according to the mutation type. (B) Somatic interactions between groups. Mutually exclusive or co‐occurring sets of genes were detected using a pairwise Fisher's exact test to detect significant gene pairs. (C) Mutational signatures plots known as SBS in 186 patients with CRC over 40 years old. (D) Mutational signatures plots known as SBS in 11 patients with CRC under 40 years old.