Polygenic risk scores for nicotine use and family history of smoking are associated with smoking behaviour.

Abstract

Formal genetics studies show that smoking is influenced by genetic factors; exploring this on the molecular level can offer deeper insight into the etiology of smoking behaviours. Summary statistics from the latest wave of the GWAS and Sequencing Consortium of Alcohol and Nicotine (GSCAN) were used to calculate polygenic risk scores (PRS) in a sample of ~2200 individuals who smoke/individuals who never smoked. The associations of smoking status with PRS for Smoking Initiation (i.e., Lifetime Smoking; SI-PRS), and Fagerström Test for Nicotine Dependence (FTND) score with PRS for Cigarettes per Day (CpD-PRS) were examined, as were distinct/additive effects of parental smoking on smoking status. SI-PRS explained 10.56% of variance (Nagelkerke-R²) in smoking status (p=6.45x10^-30). In individuals who smoke, CpD-PRS was associated with FTND score (R²=5.03%, p=1.88x10^-12). Parental smoking alone explained R²=3.06% (p=2.43×10^-12) of smoking status, and 0.96% when added to the most informative SI-PRS model (total R²=11.52%). These results show the potential utility of molecular genetic data for research investigating smoking prevention. The fact that PRS explains more variance than family history highlights progress from formal to molecular genetics; the partial overlap and increased predictive value when using both suggests the importance of combining these approaches.