GAD Antibody Positivity Predicts Type 2 Diabetes in an Adult Population

OBJECTIVE

To evaluate the significance of GAD antibodies (GADAs) and family history for type 1 diabetes (FH_T1) or type 2 diabetes (FH_T2) in nondiabetic subjects.

RESEARCH DESIGN AND METHODS

GADAs were analyzed in 4,976 nondiabetic relatives of type 2 diabetic patients or control subjects from Finland. Altogether, 289 (5.9%) were GADA⁺—a total of 253 GADA⁺ and 2,511 GADA⁻ subjects participated in repeated oral glucose tolerance tests during a median time of 8.1 years. The risk of progression to diabetes was assessed using Cox regression analysis.

RESULTS

Subjects within the highest quartile of GADA⁺ (GADA⁺_high) had more often first-degree FH_T1 (29.2 vs. 7.9%, P < 0.00001) and GADA⁺ type 2 diabetic (21.3 vs. 13.7%, P = 0.002) or nondiabetic (26.4 vs. 13.3%, P = 0.010) relatives than GADA⁻ subjects. During the follow-up, the GADA⁺ subjects developed diabetes significantly more often than the GADA⁻ subjects (36/253 [14.2%] vs. 134/2,511 [5.3%], P < 0.00001). GADA⁺_high conferred a 4.9-fold increased risk of diabetes (95% CI 2.8–8.5) compared with GADA⁻—seroconversion to positive during the follow-up was associated with 6.5-fold (2.8–15.2) and first-degree FH_T1 with 2.2-fold (1.2–4.1) risk of diabetes. Only three subjects developed type 1 diabetes, and others had a non–insulin-dependent phenotype 1 year after diagnosis. GADA⁺ and GADA⁻ subjects did not clinically differ at baseline, but they were leaner and less insulin resistant after the diagnosis of diabetes.

CONCLUSIONS

GADA positivity clusters in families with type 1 diabetes or latent autoimmune diabetes in adults. GADA positivity predicts diabetes independently of family history of diabetes, and this risk was further increased with high GADA concentrations.Latent autoimmune diabetes in adults (LADA) was introduced nearly 2 decades ago to separate a GAD antibody (GADA)-positive subgroup of adult patients initially diagnosed with type 2 diabetes (1,2). Using this definition with the add-on criteria of no exogenous insulin during the first 6–12 months, the prevalence of LADA among unselected “type 2 diabetic patients” is ∼25% in subjects younger than 35 years and between 4 and 13% in subjects older than 35 years at diagnosis in populations of European origin (3–9). In follow-up studies, a progressive defect in insulin secretion was observed in ∼50–60% of LADA patients within 6–10 years (3,10), which led to the inclusion of these patients as a slowly progressing form of type 1 diabetes in the last World Health Organization (WHO) classification of diabetes (11). However, both the existence of LADA as a distinct subgroup of diabetes and the criteria that should be used to diagnose it have been challenged (e.g., (12,13). The LADA group is heterogeneous, and most studies have been cross-sectional, whereas prospective studies including patients at or before diagnosis and population-based studies are few (3,4,14–16). Genetic background, especially for type 1 diabetes, may be a confounding factor, and we have shown that LADA was more frequent in families with both type 1 and type 2 diabetes than in families with type 2 diabetes only (17). Moreover, some data support that type 1 and type 2 diabetes cluster in same families (17–20), although this has been contradicted in a large U.K. study on parents of type 1 diabetic patients (21).In children, progression to diabetes has been associated with high antibody levels and early development of multiple autoantibodies, whereas subjects with a later appearance of antibodies had a slower progression (22–25). We have previously hypothesized that GADAs would be a marker of a subclinical autoimmune process and showed that GADA positivity was associated with a decrease in maximal insulin secretory capacity in nondiabetic subjects (26). If that is the case, GADAs should also be a predictor of future diabetes in adults. This was not supported by two studies on the general population (16,27), but a Swedish study reported a sixfold increased risk for diabetes in GADA⁺ subjects (15).In a prospective follow-up study of a large cohort of relatives of type 2 diabetic patients and population control subjects from Finland, we have now evaluated the predictive value of GADAs and family history for type 1 or type 2 diabetes in conjunction with the traditional risk factors for diabetes.