Prevalence of Glutamic Acid Decarboxylase (GAD) Antibody Positivity in Young-Onset Diabetes: A Retrospective Study From Northeast India

Background: Latent autoimmune diabetes in adults (LADA) is a distinct form of diabetes characterized by autoimmune β-cell destruction with a slower progression compared to classical type 1 diabetes. It is frequently misclassified as type 2 diabetes, particularly in younger individuals, leading to suboptimal management. The presence of anti-glutamic acid decarboxylase (GAD) antibodies is a key diagnostic marker for LADA. However, data on the prevalence and distribution of GAD antibodies among young-onset diabetes patients in Northeast India are limited. Methods: This retrospective cross-sectional study was conducted at the Multidisciplinary Research Unit of a tertiary care center in Northeast India using archived serum samples from the Indian Council of Medical Research (ICMR)-funded PHENOEINDY-2 (Phenotyping North East Indian Young Type 2 Diabetes) cohort study (2017-2021). A total of 149 individuals diagnosed with diabetes before 40 years of age were included based on the availability of adequate stored serum samples. Serum anti-GAD antibody levels were measured using a standardized enzyme-linked immunosorbent assay (ELISA). Participants were categorized as GAD-negative (<5 U/mL) or GAD-positive (≥5 U/mL), the latter considered suggestive of an autoimmune diabetes phenotype. Descriptive analyses were performed to determine the prevalence and distribution of GAD antibody levels. Results: The overall mean GAD antibody level was 3.17 U/mL, with a median of 1.07 U/mL and a wide range (0.26-109.15 U/mL), demonstrating a markedly right-skewed distribution (skewness 7.52; kurtosis 63.29). Fourteen patients (9.4%) were classified as GAD-positive, indicating a probable diagnosis of LADA. The GAD-negative group (n=135) showed a tightly clustered distribution with low variability (mean 0.95 U/mL; SD 0.54 U/mL), consistent with non-autoimmune diabetes. In contrast, the GAD-positive group exhibited substantial heterogeneity (mean 24.58 U/mL; SD 29.67 U/mL), with several individuals demonstrating markedly elevated antibody levels exceeding 70-100 U/mL. Visual analyses confirmed a clear separation between the two groups, with compact clustering in the GAD-negative cohort and a wide dispersion in the GAD-positive subgroup. Conclusion: Approximately 1 in 10 young-onset diabetes patients in this cohort demonstrated GAD positivity, suggesting that autoimmune diabetes represents an important subset of young-onset diabetes in Northeast India. These findings provide preliminary epidemiological evidence on the prevalence of GAD positivity in this population and support the need for larger prospective studies incorporating detailed clinical phenotyping, metabolic follow-up, and evaluation of the clinical utility of GAD antibody testing to improve diagnostic classification and patient management.