Our study primarily focused on European ancestry, but we also included two non-European studies but these non-European studies lacked statistical power on their own to identify ancestry-specific effects. Therefore, we did not perform ancestry-specific meta-analyses. Nevertheless, our results offered cross ancestry replication. One of the associations identified in the conditional analyses, rs8102683 (near CYP2A6), confirmed an association with CPD that was previously identified by Kumasaka et al. in a Japanese population [70] but this is the first time it was associated in Europeans (rs8102683 is also correlated with rs56113850 (r2 = 0.43), a SNV identified previously by Loukola et al. [59] in a genetic association study of nicotine metabolite ratio in Europeans). As more non-European studies become available, it would be of great interest to perform non-European ancestry studies, in order to fine-map causal variants for smoking-related traits.
