Mendelian randomization study on the effect of tumor necrosis factor on schizophrenia

Previous observational studies have shown that the levels of tumor necrosis factor (TNF) increased in patients with schizophrenia. The present two-sample Mendelian randomization (MR) study aims to identify the causal link between TNF and schizophrenia.

To date, the largest genome-wide association study (GWAS) for TNF (n = 23 141) and for schizophrenia (53 386 cases and 77 258 controls) was used. All participants were of European ancestry. The MR-egger_intercept test and Cochran's Q statistic were used to determine the pleiotropy and heterogeneity, respectively. Weighted median and inverse variance weighted (IVW) were used to evaluate the causal association of TNF with schizophrenia.

We found no significant pleiotropy or heterogeneity of all three selected plasma TNF genetic instrumental variants in breast cancer GWAS. Interestingly, the odds ratio (OR) = 1.517 with 95% confidence interval (CI), 1.006-2.288 and P = 0.047 of schizophrenia correspond to one unit increase in natural log-transformed TNF levels using IVW method. The increased trend was further proven using weighted median (OR = 1.585; 95% CI, 1.017-2.469; P = 0.042). Reverse MR analysis shows no causal effect of schizophrenia on plasma TNF levels.

Our analysis suggested a causal association between genetically increased TNF signaling and increased risk of schizophrenia in the European population. Thus, TNF may be a potential risk for schizophrenia.