Genetic correlation between smoking behaviors and schizophrenia

Nicotine dependence is highly comorbid with schizophrenia, and the etiology of the comorbidity is unknown. To determine whether there is a genetic correlation of smoking behavior with schizophrenia, genome-wide association study (GWAS) meta-analysis results from five smoking phenotypes (ever/never smoker (N=74,035), age of onset of smoking (N=28,647), cigarettes smoked per day (CPD, N=38,860), nicotine dependence (N=10,666), and current/former smoker (N=40,562)) were compared to GWAS meta-analysis results from schizophrenia (N=79,845) using linkage disequilibrium (LD) score regression. First, the SNP heritability (h²_g) of each of the smoking phenotypes was computed using LD score regression (ever/never smoker h²_g=0.08, age of onset of smoking h²_g=0.06, CPD h²_g=0.06, nicotine dependence h²_g=0.15, current/former smoker h²_g=0.07, p<0.001 for all phenotypes). The SNP heritability for nicotine dependence was statistically higher than the SNP heritability for the other smoking phenotypes (p<0.0005 for all two-way comparisons). Next, a statistically significant (p<0.05) genetic correlation was observed between schizophrenia and three of the five smoking phenotypes (nicotine dependence r_g=0.14, CPD r_g=0.12, and ever/never smoking r_g=0.10). These results suggest that there is a component of common genetic variation that is shared between smoking behaviors and schizophrenia.