A Cut-Off Value for Anti-GAD Ab ELISA Test for Predicting Future Insulin Dependency in SPIDDM May Not Exist

We previously demonstrated that patients initially diagnosed with slowly progressive insulin-dependent (type 1) diabetes mellitus (SPIDDM) using a radioimmunoassay (RIA) test for anti-glutamic acid decarboxylase antibody (GADA) (GADA-RIA), who were also positive on a GADA enzyme-linked immunosorbent assay (ELISA) test (GADA-ELISA), showed a reduced insulin secretory capacity early in the SPIDDM disease process [Endocr J. 2017;64:163]. When we used GADA-RIA test, in SPIDDM with non-insulin dependent state, a GADA-RIA value ≥10 U/mL(GADA-RIA-High) was believed to indicate a higher risk for future insulin dependency than a value <10 U/mL(GADA-RIA-Low). In our previous study also, the insulin secretory capacity may be lower in “GADA-RIA-High and GADA-ELISA-positive SPIDDM patients” (A group) than those with “GADA-RIA-Low and GADA-ELISA-positive SPIDDM patients” (B group) early in the disease process, but this point was not yet confirmed. Therefore, we have now performed additional analysis to compare serum C-peptide levels in these two patient groups (group A vs. B). When restricted to a shorter disease duration, <11 years based on our previous study, serum C-peptide levels were 0.73 ± 0.61 ng/mL in group B (n = 9) and 0.68 ± 0.65 ng/mL in group A(n = 15). There was no significant difference between the two measures, but each of these two measurements was significantly lower than those for patients with type 2 diabetes (2.20 ± 0.85 ng/mL; n = 15) and those for “GADA-RIA-Low and GADA-ELISA-NEGATIVE SPIDDM patients” (2.60 ± 1.66 ng/mL; n = 6). These findings suggest that, regardless of their GADA-RIA titer, GADA-ELISA-positive patients with SPIDDM may show a reduced insulin secretory capacity early in the disease process.

In conclusion, unlike GADA-RIA test, a cut-off value for GADA-ELISA test for predicting future insulin dependency in SPIDDM may not exist.