Nimptsch Katharina, Song Mingyang, Aleksandrova Krasimira, Katsoulis Michail, Freisling Heinz, Jenab Mazda, Gunter Marc J, Tsilidis Konstantinos K, Weiderpass Elisabete, Bueno-De-Mesquita H Bas, Chong Dawn Q, Jensen Majken K, Wu Chunsen, Overvad Kim, Kühn Tilman, Barrdahl Myrto, Melander Olle, Jirström Karin, Peeters Petra H, Sieri Sabina, Panico Salvatore, Cross Amanda J, Riboli Elio, Van Guelpen Bethany, Myte Robin, Huerta José María, Rodriguez-Barranco Miguel, Quirós José Ramón, Dorronsoro Miren, Tjønneland Anne, Olsen Anja, Travis Ruth, Boutron-Ruault Marie-Christine, Carbonnel Franck, Severi Gianluca, Bonet Catalina, Palli Domenico, Janke Jürgen, Lee Young-Ae, Boeing Heiner, Giovannucci Edward L, Ogino Shuji, Fuchs Charles S, Rimm Eric, Wu Kana, Chan Andrew T, Pischon Tobias
Molecular Epidemiology Research Group, Max Delbrück Center for Molecular Medicine (MDC), Robert-Rössle-Straße 10, 13125, Berlin, Germany.
Department of Nutrition, Harvard T.H. Chan School of Public Health, Boston, MA, USA.
Eur J Epidemiol. 2017 May;32(5):419-430. doi: 10.1007/s10654-017-0262-y. Epub 2017 May 26.
Higher levels of circulating adiponectin have been related to lower risk of colorectal cancer in several prospective cohort studies, but it remains unclear whether this association may be causal. We aimed to improve causal inference in a Mendelian Randomization meta-analysis using nested case-control studies of the European Prospective Investigation into Cancer and Nutrition (EPIC, 623 cases, 623 matched controls), the Health Professionals Follow-up Study (HPFS, 231 cases, 230 controls) and the Nurses' Health Study (NHS, 399 cases, 774 controls) with available data on pre-diagnostic adiponectin concentrations and selected single nucleotide polymorphisms in the ADIPOQ gene. We created an ADIPOQ allele score that explained approximately 3% of the interindividual variation in adiponectin concentrations. The ADIPOQ allele score was not associated with risk of colorectal cancer in logistic regression analyses (pooled OR per score-unit unit 0.97, 95% CI 0.91, 1.04). Genetically determined twofold higher adiponectin was not significantly associated with risk of colorectal cancer using the ADIPOQ allele score as instrumental variable (pooled OR 0.73, 95% CI 0.40, 1.34). In a summary instrumental variable analysis (based on previously published data) with higher statistical power, no association between genetically determined twofold higher adiponectin and risk of colorectal cancer was observed (0.99, 95% CI 0.93, 1.06 in women and 0.94, 95% CI 0.88, 1.01 in men). Thus, our study does not support a causal effect of circulating adiponectin on colorectal cancer risk. Due to the limited genetic determination of adiponectin, larger Mendelian Randomization studies are necessary to clarify whether adiponectin is causally related to lower risk of colorectal cancer.
在多项前瞻性队列研究中，循环脂联素水平较高与结直肠癌风险较低相关，但这种关联是否为因果关系仍不清楚。我们旨在通过孟德尔随机化荟萃分析改进因果推断，该分析使用了欧洲癌症与营养前瞻性调查（EPIC，623例病例，623例匹配对照）、健康专业人员随访研究（HPFS，231例病例，230例对照）和护士健康研究（NHS，399例病例，774例对照）的巢式病例对照研究，这些研究有诊断前脂联素浓度和ADIPOQ基因中选定单核苷酸多态性的可用数据。我们创建了一个ADIPOQ等位基因评分，该评分解释了脂联素浓度个体间变异的约3%。在逻辑回归分析中，ADIPOQ等位基因评分与结直肠癌风险无关（每评分单位的合并比值比为0.97，95%可信区间为0.91，1.04）。使用ADIPOQ等位基因评分作为工具变量，遗传决定的脂联素水平升高两倍与结直肠癌风险无显著关联（合并比值比为0.73，95%可信区间为0.40，1.34）。在一项基于先前发表数据且具有更高统计效力的汇总工具变量分析中，未观察到遗传决定的脂联素水平升高两倍与结直肠癌风险之间的关联（女性为0.99，95%可信区间为0.93，1.06；男性为0.94，95%可信区间为0.88，1.01）。因此，我们的研究不支持循环脂联素对结直肠癌风险有因果效应。由于脂联素的遗传决定因素有限，需要更大规模的孟德尔随机化研究来阐明脂联素是否与结直肠癌风险降低存在因果关系。
