Abstract
Anti-glutamic acid decarboxylase antibody (GADA) is an important islet cell-associated autoantibody for the diagnosis of autoimmune type 1 diabetes mellitus. In Japan, the GADA assay kit was recently changed from radioimmunoassay (RIA) to enzyme-linked immunosorbent assay (ELISA). Thereafter, a mismatched measurement between the two tests became apparent in clinical situations. The present study aimed to clarify the actual extent of mismatch between the two measurements on a larger-scale real-world clinical practice. In this cross-sectional non-local/non-hospital-based study, we collected anonymized data on GADA levels of 598 participants, who were simultaneously measured with GADA-RIA and GADA enzyme-linked immunosorbent assay tests. We found that 34% of the GADA-RIA-positive participants showed negative results in the GADA enzyme-linked immunosorbent assay test; the mismatch was predominantly observed in participants with relatively low GADA-RIA levels (<32 U/mL). This considerable mismatch might lead to physicians' confusion in diagnosing type 1 diabetes mellitus.