Abstract 7412: The relationship between lifestyle factors and polygenic risk score with prostate cancer risk in the Multiethnic Cohort

Abstract Introduction: Behavior modifications for men at high genetic risk of prostate cancer (PCa) have the potential to prevent or delay disease onset. In this study, we evaluated whether the association of a previously developed healthy behavior score with PCa risk varies across levels of genetic susceptibility, as measured by a multi-ancestry polygenic risk score (PRS) in the Multiethnic Cohort (MEC). Methods: We conducted a nested case-control analysis of 6, 096 incident PCa cases and 24, 684 controls among African American (13%), Latino (23%), Japanese American (35%), Native Hawaiian (7%), and White (22%) men in the MEC. Among the PCa cases, 2, 049 were aggressive cases and 776 were lethal cases. We calculated the 2015 Prostate Cancer Behavior Score from the baseline questionnaire, which comprised six modifiable factors (smoking, BMI, physical activity, intake of tomatoes, processed meat, and fatty fish), with each factor scored as 1 for meeting healthy thresholds and 0 otherwise. This score ranged from 0 to 6 with a higher score indicating a healthier behavior pattern. The PRS of 451 risk variants was constructed with weights from the largest multi-ancestry genome-wide association study for PCa. In the overall study and within each racial/ethnic group, we evaluated the association of the behavior score (categorized as 0-2, 3, 4, 5-6) with risk of overall, aggressive, and lethal PCa in logistic regression analyses and assessed interactions between the behavior score and PRS (categorized as 0-50%, 50-75%, and 75-100%) on PCa risk using the likelihood ratio test (LRT). A P < 0.05 was considered statistically significant. Results: Comparing the highest to the lowest category of the behavior score (5-6 vs. 0-2), there was a suggestive 29% (95% CI: 0.47-1.04, P=0.09) lower risk of lethal PCa in the overall study. Although not statistically significant, a similar inverse association was observed in Whites (OR=0.36), Japanese Americans (OR=0.26), and Latinos (OR=0.26) but not in African Americans (OR=1.62). When stratified by PRS category in the overall study, this inverse association appeared to be stronger for men in the 75-100% PRS category (OR=0.55, 95% CI: 0.27-1.04, P=0.07) than those in the lower PRS categories (ORs 0.72-0.97, P >0.9). These differences in the association of behavior score with risk of lethal PCa across PRS categories were not statistically significant (PLRT = 0.56). No interaction was observed between the behavior score and PRS on risk of overall and aggressive PCa. Conclusion: Findings from this multiethnic study provided further support that the 2015 Prostate Cancer Behavior Score was associated with a reduced risk of lethal PCa. However, there was limited evidence that this association was modified by PCa PRS, which could be due to the relatively small number of lethal PCa cases in our study. Further investigations in larger diverse populations are warranted to confirm our findings. Citation Format: Raymond W. Hughley, Anqi Wang, Peggy Wan, Xin Xiong, Wei Xiong, David Bogumil, Song Yi Park, Lynne Wilkens, Loic Le Marchand, David Conti, Christopher Haiman, Fei Chen. The relationship between lifestyle factors and polygenic risk score with prostate cancer risk in the Multiethnic Cohort [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 7412.