Association between telomere length SNPs and five cancer types: a Mendelian randomization study from the GAME-ON post-GWAS consortium. C. Zhang1, S. Burgess4, P. Kraft5, S. Lindstrom5, R. Hung6, U. Peters7, H. Ahsan1,2,3, T. Sellers8, A. Monteiro8, G. Trench9, J. Doherty10, B. Pierce1,2 on behalf of the CORECT, DRIVE, ELLIPSE, OCAC, and TRICL studies and the GAME-ON Network 1) Department of Health Studies, The University of Chicago, Chicago, IL; 2) Comprehensive Cancer Center, The University of Chicago, Chicago, IL; 3) Departments of Medicine and Human Genetics, The University of Chicago, Chicago, IL; 4) Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK; 5) Program in Genetic Epidemiology and Statistical Genetics, Department of Epidemiology, Harvard School of Public Health, Boston, MA; 6) Lunenfeld-Tanenbaum Research Institute of Mount Sinai Hospital, Toronto; 7) Department of Epidemiology, School of Public Health, University of Washington, Seattle, WA; 8) Department of Cancer Epidemiology, H. Lee Moffitt Cancer Center, Tampa, FL; 9) Department of Genetics, Queensland Institute of Medical Research, Brisbane, Queensland, Australia; 10) Section of Biostatistics & Epidemiology, The Geisel School of Medicine at Dartmouth, Lebanon, NH.

   Epidemiological studies have reported associations between telomere length (measured in peripheral blood cells) and risk for various cancer types, but results have been inconsistent. These inconsistencies have been attributed in part to methodological issues related to the use of cross-sectional or retrospective study designs, which may be biased due to reverse causality or the effects of cancer progression and treatment. One way to address this issue is to indirectly estimate the association between telomere length and cancer risk using single nucleotide polymorphisms (SNPs) associated with telomere length. We estimated the association between a multi-SNP score consisting of nine telomere length-associated SNPs and risk for five cancer types (breast, lung, colorectal, ovarian and prostate cancer) including various subtypes, using data from the Genetic Association Mechanisms in Oncology genome-wide association study consortium (GAME-ON). The association estimates correspond to the effects of telomere length on the cancer risk under Mendelian randomization assumptions. We found the multi-SNP score for short telomeres to be significantly associated with decreased risk of lung adenocarcinoma (OR = 0.31, 95% CI: 0.23-0.41, P = 6.1x10-15), and suggestively associated with decreased risk of colorectal (OR = 0.73, 95% CI: 0.52-1.03, P = 0.08) and prostate (OR = 0.82, 95% CI: 0.65-1.03, P = 0.09) cancer. These estimates can be interpreted as the reduction in odds of cancer risk per 1000 base pair reduction in telomere length. Our results suggest that short telomere length is associated with decreased risk of lung adenocarcinoma, possibly associated with decreased risk of prostate and colorectal cancer, and not associated with breast or ovarian cancer. Findings from this study provide additional information to help interpret relationships between telomere length and specific cancer types, and may be used in the development of cancer prediction strategies.

You may contact the first author (during and after the meeting) at