Abstract 1004: Genome-wide analysis of the interaction between physical activity and genetic susceptibility on colorectal cancer incidence: A nested case-control study in the UK Biobank

Abstract Background: Colorectal cancer (CRC) risk is influenced by a complex interplay between genetic predisposition and lifestyle factors, such as physical activity (PA). To investigate potential gene-environment interactions, we conducted a genome-wide interaction study (GWIS) using data from the UK Biobank (proposal #94695). Methods: Participants with prior cancer diagnoses, missing genotyping data, missing PA data, or missing covariates (e.g., deprivation index, BMI, smoking, alcohol) were excluded. Genotyping data underwent stringent quality control with thresholds set for minor allele frequency (MAF > 0.03), genotyping call rate (> 99%), and Hardy-Weinberg equilibrium (p > 1 × 10-6). After exclusions, 325, 858 participants with the required covariates were included as the base population. From this cohort, 3, 607 CRC cases were matched to 14, 155 controls using a 1:4 incidence density matching approach based on sex, age at recruitment (±5 years), and follow-up duration (±6 months). PA was categorized based on whether participants met the World Health Organization guidelines: engaging in at least 150 minutes of moderate-intensity activity or 75 minutes of vigorous-intensity activity per week. Conditional logistic regression models were used to evaluate SNP × PA interactions, adjusting for BMI, smoking, alcohol consumption, deprivation index, the first ten genetic principal components, and batch effects. Interaction p-values (Pinteraction) were derived from the statistical significance of the SNP × PA interaction term in the model. Results: After quality control, 402, 069 SNPs were retained for analysis. We could not find any SNPs that met the genome-wide significance threshold of p < 1 × 10-7. However, seven SNPs met the suggestive significance threshold of p < 1 × 10-5, including rs6791309 (Pinteraction = 1.20 × 10-6), which had the lowest interaction p-value, and rs6770805 (3.65 × 10-6), which is located within the ADCY5 gene. Other suggestive variants were identified in genes such as KCNIP4, CDH18 and LOC124904472. Conclusions: This GWIS identified SNPs with suggestive evidence of interaction with PA in relation to CRC risk, although no variants achieved genome-wide significance. Variants in genes such as ADCY5, KCNIP4, CDH18 and LOC124904472 highlight loci that may contribute to CRC risk through gene-environment interactions. Further replication studies and functional analyses are necessary to validate these findings and uncover the underlying biological mechanisms. Citation Format: Sooyoung Cho, Aesun Shin. Genome-wide analysis of the interaction between physical activity and genetic susceptibility on colorectal cancer incidence: A nested case-control study in the UK Biobank [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 1004.