PO.PS01.07 · 人群科学
身体质量指数多基因风险评分(PRS BMI)与西班牙裔社区健康研究/拉丁裔研究中女性肥胖相关癌症风险
A body mass index polygenic risk score (PRS BMI ) and obesity-related cancer risk among women in the Hispanic Community Health Study/Study of Latinos
该海报暂无可下载的资料
AACR 官方页面
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
背景。肥胖与包括女性结肠癌、直肠癌、子宫内膜癌和乳腺癌在内的癌症风险增加相关。多基因风险评分(PRS)通过汇总风险变异的效应,估计个体对复杂表型性状(如肥胖)的遗传易感性。我们构建了身体质量指数的PRS(PRS BMI),并考察了PRS BMI是否与西班牙裔/拉丁裔女性的肥胖相关癌症(ORC)风险相关。我们还考察了BMI是否调节或中介了PRS BMI与ORC风险之间的关联。
方法。这项前瞻性队列研究纳入了8,380名西班牙裔/拉丁裔女性(基线时平均年龄=41.8±0.3岁),来自西班牙裔社区健康研究/拉丁裔研究(HCHS/SOL)。在2008-2011年基线时,测量身高和体重并用于计算BMI,参与者提供了血样用于基因分型。PRS BMI使用从多基因评分(PGS)目录下载的既往发表权重构建。通过与四个州癌症登记处的链接,确定了从基线至2021年诊断的13种肥胖相关癌症的发病情况(在平均10.8年的随访期间共诊断出n=223例ORC事件)。调查多变量Cox回归模型估计了PRS BMI(每标准差,SD)与ORC风险之间的关联(风险比,HR,及95%置信区间,CI),包括总体关联以及按BMI(<25 vs. ≥25 kg/m2)分层的关联。调查多变量Cox回归和线性回归模型检验了BMI(连续变量,kg/m2)对PRS BMI(每SD)与ORC风险之间关联的中介效应。
结果。在分层分析中,在BMI<30 kg/m2的女性中,PRS BMI每增加1个SD与ORC的HR为0.74(95%CI=0.56-0.97)相关,但在BMI≥30 kg/m2的女性中则无此关联(HR=0.95;95%CI=0.68-1.33;P交互=0.15)。在中介分析中,PRS BMI与BMI显著相关(beta a=1.86;95%CI=1.66-2.06);在校正PRS BMI后,BMI与ORC风险显著相关(beta b=0.07;SE=0.01;HR=1.07;95%CI=1.05-0.19);且BMI显著中介了PRS BMI与ORC风险之间的关联(间接效应,beta a*beta b=0.13;95%CI=0.07-0.19)。
结论。使女性易于具有较高BMI的基因可能主要通过BMI影响ORC风险。这些结果凸显了体重管理对西班牙裔/拉丁裔女性癌症预防的重要性。
查看英文原文 English abstract
Background. Obesity is associated with an increased risk of cancer including cancers of the colon and rectum, endometrium, and breast among women. Polygenic risk scores (PRS), which sum the effects of risk variants, estimate an individual's genetic predisposition to complex phenotypic traits, such as obesity. We constructed a PRS for body mass index (PRS BMI ) and examined whether the PRS BMI was associated with the risk of obesity-related cancers (ORCs) among Hispanic/Latina women. We also examined whether BMI moderated or mediated the association between PRS BMI and ORC risk.
Methods . This prospective cohort study included 8,380 Hispanic/Latina women (mean age at baseline=41.8±0.3 years) from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL). At baseline in 2008-2011, height and weight were measured and used to compute BMI, and participants provided blood samples for genotyping. The PRS BMI was constructed using previously published weights downloaded from the polygenic score (PGS) catalog. The incidence of 13 obesity-related cancers diagnosed from baseline through 2021 was ascertained through linkages with four state cancer registries ( n =223 incident ORCs were diagnosed over a mean follow-up of 10.8 years). Survey multivariable Cox regression models estimated the associations (hazard ratios, HRs, and 95% confidence intervals, CIs) between the PRS BMI (per standard deviation, SD) and ORC risk overall and by BMI (<25 vs. ≥25 kg/m 2 ). Survey multivariable Cox regression and linear regression models tested the mediating effect of BMI (continuous, kg/m 2 ) on the association between PRS BMI (per SD) and ORC risk.
Results. In stratified analyses, a 1-SD increase in the PRS BMI was associated with an ORC HR of 0.74 (95%CI=0.56-0.97) among women with BMI<30 kg/m 2 , but not among women with a BMI≥30 kg/m 2 (HR=0.95; 95%CI=0.68-1.33; P Interaction =0.15). In mediation analyses, PRS BMI was significantly associated with BMI (beta a =1.86; 95%CI=1.66-2.06); BMI was significantly associated with ORC risk, adjusting for PRS BMI (beta b =0.07; SE=0.01; HR=1.07; 95%CI=1.05-0.19); and BMI significantly mediated the association between PRS BMI and ORC risk (indirect effect, beta a *beta b =0.13; 95%CI=0.07-0.19).
Conclusion . Genes predisposing women to a higher BMI may primarily impact ORC risk through BMI. These results highlight the importance of weight management for cancer prevention among Hispanic/Latina women.
利益披露 Disclosure
H. Parada, None..
T. Sofer, None..
M. Pichardo, None..
K. M. Perreira, None..
L. Zhou, None..
F. J. Penedo, None..
A. Pirzada, None..
M. Daviglus, None..
G. A. Talavera, None..
L. C. Gallo, None.