Abstract A015: Impact of healthy lifestyles and polygenic risk scores on pancreatic cancer risk: The Multiethnic Cohort Study
Brian Huang,Adelynn Paik,David Bogumil,Fei Chen,Peggy Wan,Lynne Wilkens,Loic Le Marchand,David Conti,Christopher Haiman,Veronica Wendy Setiawan
DOI: https://doi.org/10.1158/1538-7445.panca2023-a015
IF: 11.2
2024-01-18
Cancer Research
Abstract:Background: Healthy lifestyles have been observed to reduce the risk of pancreatic cancer. However, it is unknown whether this protective impact varies among individuals with different levels of genetic susceptibility for pancreatic cancer. Methods: We investigated the association between healthy lifestyles, genetic risk, and pancreatic cancer incidence in 70,780 participants (32,783 males, 37,907 females) from the Multiethnic Cohort Study (MEC), a prospective cohort of White, African American, Native Hawaiian, Japanese American, and Latino men and women. We calculated a healthy lifestyle factor score (HLFS) based on smoking status, alcohol intake, body mass index and physical activity. We also estimated genetic risk using an established polygenic risk score (PRS) of 49 pancreatic cancer risk variants. Cox regression was used to assess the independent and joint associations of the HLFS (analyzed as quintiles) and PRS (analyzed as quintiles and high/low categories) with pancreatic cancer risk. Models were adjusted for age at blood draw, sex, race, family history of pancreatic cancer, and in the genetic analyses, the first ten principal components to account for population stratification. Results: The analytical cohort was comprised of 23,643 (33.4%) Japanese Americans, 16,816 (23.8%) Latinos, 14,305 (20.2%) Whites, 10,645 (15.0%) African Americans, and 5,371 (7.6%) Native Hawaiians. There were 721 incident cases of pancreatic cancer from an average follow-up period of 13.7 (SD 4.8) years. Higher levels of the PRS were associated with a 25-62% increase in pancreatic cancer risk, compared to the first PRS quintile. These elevated risks were borderline significant for quintile 2 (HR 1.25, 95% CI 0.97-1.61) and statistically significant for quintile 3 (HR 1.35, 95% CI 1.05-1.73), quintile 4 (HR 1.46, 95% CI 1.15-1.87), and quintile 5 (HR 1.62, 95% CI 1.27-2.06). For the HLFS, higher quintiles were associated with a 10-27% decreased risk of pancreatic cancer, compared to the lowest quintile, but this was only statistically significant for the highest quintile (HR 0.73, 95% CI 0.56-0.94). When evaluating the HLFS by levels of the PRS (low 0-50% vs. high 50-100%), the association between the highest HLFS quintile and pancreatic cancer was more pronounced among those in the high PRS group (HR 0.64, 95% CI 0.45-0.91), compared to those in the low PRS group (HR 0.80, 95% CI 0.55-1.18). However, this difference in effect by PRS levels was not statistically significant (p-interaction=0.43). Conclusions: Our study demonstrates that including information on healthy lifestyles and genetic risk may improve the risk stratification of pancreatic cancer for multiethnic populations. In addition, it is possible that healthy lifestyles could provide more benefit to individuals at higher genetic risk. Citation Format: Brian Huang, Adelynn Paik, David Bogumil, Fei Chen, Peggy Wan, Lynne Wilkens, Loic Le Marchand, David Conti, Christopher Haiman, Veronica Wendy Setiawan. Impact of healthy lifestyles and polygenic risk scores on pancreatic cancer risk: The Multiethnic Cohort Study [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Pancreatic Cancer; 2023 Sep 27-30; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(2 Suppl) nr A015.
oncology