FIG 3.

Meta-analyzed association between CH and risk of incident lung cancer, UKBB and MGBB. Meta-analyzed ORs were bolded. CH-negative individuals were considered as the referent group in all models. All models were adjusted for age and year at blood draw, sex (female/male), race (White/others), smoking status (never/past/current smoker), pack-years of smoking (continuous), family history of lung cancer (no/yes), and the first 10 principal components of genetic ancestry. P for heterogeneity was estimated using Cochran's Q test. CH, clonal hematopoiesis; MGBB, Mass General Brigham Biobank; OR, odds ratio; UKBB, UK Biobank; VAF, variant allele frequency.