Mendelian Randomization Analysis Of 5p15.33, Telomere Length And Lung Cancer Risk: Results From 14,324 Cases And 10,783 Controls In The Transdisciplinary Research in Cancer of the Lung (TRICL) Group Of The International Lung Cancer Consortium (ILCCO)
Linda KACHURI, Mount Sinai Hospital & Dalla Lana School of Public Health at the University of Toronto, Canada
DAVEY SMITH G. 3
, LIU G. 2,4
, LANDI M. 5
, CHRISTIANI D. 6
, CAPORASO N. 5
, MCKAY J. 7
, TRICL AND ILCCO P. 8,9
, BRENNAN P. 7
, AMOS C. 10
, HUNG R. 1,2
1 Lunenfeld-Tanenbaum Research Institute of Mount Sinai Hospital, Toronto, Canada
2 Dalla Lana School of Public Health, University of Toronto, Toronto, Canada
3 School of Social and Community Medicine, University of Bristol, Bristol, UK
4 Ontario Cancer Institute, Princess Margaret Cancer Center, Toronto, Canada
5 Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, USA
6 Departments of Environmental Health and Epidemiology, Harvard School of Public Health, Boston, USA
7 Section of Genetics, International Agency for Research on Cancer, Lyon, France
8 Xifeng Wu, Melinda C. Aldrich, Gad Rennert, Dawn Teare, Chu Chen, Gary E. Goodman, Jennifer A. Doherty, John K. Field, Lambertus A. Kiemeney, Adonina Tardón, Aage Haugen, Stephen Lam
9 Loic Le Marchand, Matthew B. Schabath, Angeline S. Andrew, Mattias Johansson, Jonas Manjer, Philip Lazarus, Susanne Arnold, Stig Egil Bojesen
10 Geisel School of Medicine, Dartmouth College, Lebanon, USA
Purpose: Telomere length (TL) has been consistently associated with lung cancer risk, but the direction of this effect remains controversial. To elucidate the causal relationship between TL and lung cancer, we compared results from an observational study and a Mendelian Randomization (MR) analysis, where we developed novel genetic instruments for TL and tested their association with lung cancer in 20 pooled TRICL/ILCCO studies.
Methods: The observational analysis estimated odds ratios (OR) for TL, measured using qPCR, in 1128 cases and 928 controls. To develop novel TL instrumental variables (IVs), variants identified through deep-sequencing of 5p15.33 (TERT/CLPTM1L) were genotyped in 900 controls. Variants meeting MR criteria were combined into a single IV. Six SNPs from previously identified TL genes (ACYP2, TERC, NAF1, OBFC1, DHX35, RTEL1) were used as additional IVs. Associations with lung cancer for all IVs were estimated using data from 14324 cases and 10783 controls, and applied in a likelihood-based MR model to estimate the causal effect of TL.
Results: The observational analysis suggested that longer TL was inversely associated with lung cancer risk (OR=0.94, p=0.04), especially squamous carcinoma (OR=0.77, p=1.1×10-4). In the first MR stage, we identified 8 5p15 SNPs associated with TL (p<5×10-3), including 6 novel rare variants, not previously reported. The combined 5p15.33 IV was reliably associated with TL (β=0.15, p=1.8×10-7) and explained 2.3% of TL variance. Using this instrument and 6 other SNPs as IVs, our MR analysis showed that longer TL predisposes to increased lung cancer risk (OR=1.77, 95% CI=1.32-1.54), especially for adenocarcinoma (OR=2.02, 95% CI=1.29-3.71).
Conclusions: This largest MR analysis of TL and lung cancer, using individual-level data and novel genetic instruments, demonstrates that longer TL is associated with increased lung cancer risk. Previously reported inverse associations for long TL were likely due to residual confounding and reverse causation.
Funding Source: CIHR (CGS-137441)