Can Depression Polygenic Risk Score Predict Postinfective Fatigue Syndrome (PIFS) in the Dubbo Infection Outcomes Study (DIOS)?

Prospective cohorts show that a postinfective fatigue syndrome (PIFS) meeting criteria for chronic fatigue syndrome (CFS) develops in roughly 1 in 10 people after diverse acute infections, independent of the triggering pathogen. Whether constitutional liability to affective disorders contributes has been debated but not tested at the level of measured genetic risk. Using the Dubbo Infection Outcomes Study (DIOS), we tested whether polygenic liability to major depressive disorder (MDD), and to CFS, predicts PIFS. Participants with serologically confirmed infection were followed at regular intervals before a physician- and psychiatrist-assessed diagnosis of PIFS at 6 months. Polygenic scores for MDD and CFS were derived from external GWAS using SBayesRC and tested as predictors of PIFS diagnosis using logistic regression, adjusted for sex, age, and ten principal components. Effect modification by sex was examined as an exploratory analysis. Of 464 participants (89 cases, 375 controls), neither MDD-PGS (adjusted OR = 0.99, 95% CI [0.78, 1.26], p = .95) nor CFS-PGS (adjusted OR = 0.97, 95% CI [0.77, 1.23], p = .81) predicted PIFS diagnosis. In exploratory analysis, the MDD-PGS × sex interaction was suggestive (Wald p = .06; Firth penalized p = .04), with higher MDD-PGS associated with increased odds of PIFS in women but decreased odds in men. The CFS-PGS × sex interaction was nonsignificant. Common-variant polygenic liability to major depression or to chronic fatigue did not predict PIFS diagnosis in this cohort, but a suggestive sex-dependent effect of major depression polygenic risk warrants replication in adequately powered cohorts.