A study of type-1 diabetes associated autoantibodies in schizophrenia.

Epidemiological studies revealed an association between type-1 diabetes (T1D) and schizophrenia but the findings reported to date have been controversial. To clarify the inconsistency across studies, T1D-associated autoantibodies were examined in plasma samples collected from 272 patients with schizophrenia and 276 control subjects. An in-house enzyme-linked immunosorbent assay (ELISA) was developed using three linear peptide antigens, one of which was derived from glutamic acid decarboxylase (GAD) and two were derived from insulinoma-associated antigen 2 (IA2). Mann-Whitney U test showed a significant decrease in the levels of plasma IgG against the IA2b antigen in schizophrenia patients as compared to control subjects (Z=-3.54, p=0.0007), while no significant difference was found between these two groups either in anti-IA2a IgG levels (Z=-1.62, p=0.105) or in anti-GAD IgG levels (Z=-1.63, p=0.104). Linear regression analysis indicated no association of antipsychotic medication with the levels of plasma IgG against IA2a, IA2b or GAD, while the levels of plasma IgG for these 3 peptide antigens were significantly correlated with each other. Binary logistic regression showed that neither the DQ2.5 variant nor the DQ8 variant was associated with circulating levels of 3 T1D-associated autoantibodies in both the patient group and the control group. The coefficient of variation was 10.7% for anti-IA2a IgG assay, 10.1% for anti-IA2b IgG assay and 10.7% for anti-GAD IgG assay. The present work suggests that T1D-associated antibodies are unlikely to confer risk of schizophrenia and that the in-house ELISA developed with linear peptide antigens is highly reproducible.