Genetic variants associated with longer telomere length are associated with increased lung cancer risk among never-smoking women in Asia: a report from the female lung cancer consortium in Asia

Abstract

Recent evidence from several relatively small nested case-control studies in prospective cohorts shows an association between longer telomere length measured phenotypically in peripheral white blood cell (WBC) DNA and increased lung cancer risk. We sought to further explore this relationship by examining a panel of seven telomere-length associated genetic variants in a large study of 5,457 never-smoking female Asian lung cancer cases and 4,493 never-smoking female Asian controls using data from a previously reported genome-wide association study. Using a group of 1,536 individuals with phenotypically measured telomere length in WBCs in the prospective Shanghai Women's Health study, we demonstrated the utility of a genetic risk score (GRS) of seven telomere-length associated variants to predict telomere length in an Asian population. We then found that GRSs used as instrumental variables to predict longer telomere length were associated with increased lung cancer risk (OR = 1.51 (95% CI = 1.34-1.69) for upper vs. lower quartile of the weighted GRS, p value = 4.54 × 10(-14) ) even after removing rs2736100 (p value = 4.81 × 10(-3) ), a SNP in the TERT locus robustly associated with lung cancer risk in prior association studies. Stratified analyses suggested the effect of the telomere-associated GRS is strongest among younger individuals. We found no difference in GRS effect between adenocarcinoma and squamous cell subtypes. Our results indicate that a genetic background that favors longer telomere length may increase lung cancer risk, which is consistent with earlier prospective studies relating longer telomere length with increased lung cancer risk.

Keywords: association study; genetic risk score; genetics; lung cancer; telomere length.

FIGURE 1. Relation of telomere-length associated variants with measured telomere length in peripheral white blood cell DNA from 1,536 women included in previous nested case-control studies of various cancers in the Shanghai Women’s Health Study

A best-fit line (solid gray line) is drawn for the relationship of measured log-transformed telomere length with telomere-length associated weighted genetic risk score for (A) cancer cases and controls (R²=0.01, P-value=0.001) and (B) controls (N = 533) only (R²=0.01, P-value=0.04).

FIGURE 2. Adjusted odds ratios for risk of lung cancer among never-smoking females in Asia comparing upper quartile to lower quartile of weighted telomere length genetic risk scores, by study

Lung cancer risk was positively associated with increasing weighted GRS (P-value=1.53×10⁻¹¹) with no significant evidence for heterogeneity of effect (P-value=0.34).

FIGURE 3. Adjusted odds ratios of weighted telomere length genetic risk scores with lung cancer risk among never-smoking females in Asia, by decile

Bars are lung cancer association odds ratios for each weighted GRS decile and error bars represent 95% confidence intervals around the odds ratios. The first decile is used as the reference group with an odds ratio of one. As compared to individuals with a telomere-length associated GRS in the first decile, individuals with a GRS in the tenth decile have a 61% (95% CI=34–94%, P-value=2.83×10⁻⁷) increased odds of developing lung cancer.