CAUSAL ASSOCIATIONS BETWEEN SCHIZOPHRENIA AND CANCERS: MENDELIAN RANDOMIZATION ANALYSES

Abstract Background Schizophrenia is one of the most prevalent neurological disorders that affect the physical, cognitive, behavioral, and emotional functioning of an individual. Cancer as a chronic disease is closely related to shorten lifespan in patients with schizophrenia. Aims & Objectives At present, the associations between these two diseases still remains unclear. So we preformed Mendelian randomization (MR) methodological approach to address whether this association is causal using genetic variants identified in genome-wide association studies (GWAS). Methods GWAS statistics data, including schizophrenia (n =533,86) from the Psychiatric Genomics Consortium and cancer patients (n = 12,409,533) from the GWAS Catalog of European ancestry, was analyzed. In this study, a two-sample MR method was used to determine the causality between schizophrenia and cancer and two-step MR design to explore whether some (including smoking and sex hormone levels) mediates the causal pathway from schizophrenia to specific cancer outcome. The inverse-variance weighted approach was used as the principal analysis, complemented with weighted median, weighted mode, simple median method, and MR- Egger analyses, while several sensitivity analyses were applied to evaluate heterogeneity and pleiotropy. Results In total, we obtained instrumental variables of schizophrenia and 20 major site-specific cancers using 38 summary association statistics. Two-sample MR analysis revealed a significant positive causal influence for schizophrenia on increased cancer risk, including breast cancer, subtype oestrogen receptor (ER) positive breast cancer, subtype ER negative breast cancer,bowel cancer if ever had screening, lung cancer, lung cancer with illnesses of mother, cancer in upper lobe, bronchus or lung and myeloid leukaemia. Two-step MR analysis showed smoking was a mediator of the positive cause-effect relationship between schizophrenia and cancer of lung. Discussion & Conclusion Our present MR study identified a causal relationship between schizophrenia and cancer. Smoking is a confounder when evaluating the association between schizophrenia and lung cancer. Further research is needed on additional risk factors as potential mediators.