Abstract Details

We analyzed data from the UK Biobank, a large cohort study that enrolled over 500,000 Britons aged 40-69. We included participants of European ancestry. First, we used logistic regression to test the association between self-reported/ICD-defined lung cancer and risk of IS. Then, we constructed a polygenic risk score (PRS) using 31 independent genetic variants known to associate with lung cancer, fitted logistic regression to assess the relationship between this PRS and risk of IS, and implemented the inverse variance weighted method of Mendelian randomization (MR). We tested for horizontal pleiotropy using the MR-Egger and MR Pleiotropy Residual Sum and Outlier (MR-PRESSO) approaches.

Out of 409,629 participants of European ancestry in the UK Biobank, there were 5,060 IS cases (mean age, 61.6 [standard deviation 6.5]; female, 1,813 [35.8%]). The prevalence of lung cancer was 1.9% (n=94) and 0.5% (n=1,961) among persons with and without IS, respectively (unadjusted p<0.001). In multivariate analysis, self-reported/ICD-defined lung cancer was associated with 54% higher odds of IS (OR, 1.54; 95% CI, 1.30-1.81; p<0.001). Genetic analyses supported a causal link between lung cancer and IS. In PRS analysis, each additional standard deviation of the lung cancer-related PRS was associated with 3% higher odds of IS (adjusted OR, 1.03; 95% CI, 1.01-1.06; p=0.018). In MR analysis, genetically determined lung cancer was associated with 5% higher odds of IS (OR, 1.05; 95% CI, 1.01-1.09; p=0.022). There was no significant horizontal pleiotropy (MR-Egger intercept p=0.663; MR-PRESSO global test p=0.993).