Polygenic Risk and Orbitofrontal Cortex Volume Predict Smoking Behavior in Adolescents

Post by Elisa Guma

What's the science?

Tobacco smoking is the leading cause of preventable illness and death in the world. Smoking behavior is thought to be heritable, with contribution from a large number of genetic variants. These genetic variants can be used to calculate a polygenic risk score, which has also been shown to predict brain structure, particularly cortical volume, another heritable feature that is altered in chronic smokers. This week in Psychological Medicine, Li and colleagues investigated the association between the polygenic risk score for smoking, and brain structure in a large population-based European dataset of 14-year old adolescents (IMAGEN cohort).

How did they do it?

The authors analyzed data from the IMAGEN project, a longitudinal imaging genetics study conducted across eight European cities. Blood samples from 1774 secondary school-aged adolescents (50.68% female, mean age 14.43±0.4 years) were collected for DNA extraction from which 477245 single nucleotide polymorphisms (markers of genetic variation) were extracted following quality control. To calculate the polygenic risk score for smoking, the authors used publicly available data from a genome-wide meta-analysis of smoking initiation from the Tobacco and Genetics Consortium (n=74053). Smoking behavior was captured using a self-report questionnaire from which the authors classified participants into 3 categories based on the frequency of smoking: ‘never users’, ‘users on a few occasions’ (1-9 occasions), and ‘users on more occasions’ (10 or more).

Cortical volume was calculated using FreeSurfer from T1-weighted structural magnetic resonance imaging scans acquired across the 8 different sites (on 3T MRI scanners from various manufacturers, but the same scanning protocol. Linear regression models were used to investigate the association between smoking polygenic risk score and smoking behavior (controlling for gender, age, site). Next, a generalized linear model was used to analyze the association between vertex-wise cortical volume and smoking polygenic risk score (controlling for gender, site, and total brain volume). Finally, a mediation model was used to test the effect of brain anatomy on the relationship between smoking polygenic risk score and smoking behavior.

What did they find?

The authors identified a nominally significant association between smoking polygenic risk scores and smoking behaviors. Next, they observed an inverse correlation between smoking polygenic risk score and cortical volume of the right orbitofrontal cortex. The volume of this region was significantly correlated with the number of occasions of tobacco use. This indicated that adolescents with a higher smoking polygenic risk score had a smaller volume of the right orbitofrontal cortex and that adolescents with smaller volume in the right orbitofrontal cortex were more likely to consume cigarettes. Finally, the mediation analysis suggested that variations in the volume of the right orbitofrontal cortex partially accounted for the relationship between the smoking polygenic risk score on smoking behavior in this adolescent sample.

What's the impact?

This study found that higher smoking polygenic risk scores were associated with an increase in the number of cigarette smokes, and smaller cortical volume of the right orbitofrontal cortex in a large adolescent population-based sample. Interestingly, this brain structure has previously been associated with risk-taking behavior and impulsivity. This strategy of combining polygenic risk scores with brain imaging to investigate the etiology of tobacco smoking can be extended to other disorders or behaviors. Finally, further investigating the role of the orbitofrontal cortex in smoking behavior may be an interesting avenue of future research.

Li et al. Orbitofrontal cortex volume links polygenic risk for smoking with tobacco use in healthy adolescents. Psychological Medicine (2020). Access the original scientific publication here.