MILLENNIUM AWARD RECIPIENT CONTRIBUTION Identification of Children with Early Onset and High Incidence of Anti-islet Autoantibodies David T. Robles,* , † ,1 George S. Eisenbarth,* , † , ‡ , § Tianbao Wang,* Henry A. Erlich, ¶ Teodorica L. Bugawan, ¶ Sunanda R. Babu,* Kathy Barriga, Jill M. Norris, Michelle Hoffman, Georgeanna Klingensmith,* Liping Yu,* and Marian Rewers,* ,  ,2 for the Diabetes Autoimmunity Study in the Young (DAISY) *Barbara Davis Center, †Department of Immunology, ‡Department of Medicine and §Department of Pediatrics, Department of Preventive Medicine and Biometrics, University of Colorado, Denver, Colorado 80262; and ¶ Roche Molecular Systems, Alameda, California 94501 A total of 21,000 general population newborns (NECs) and 693 young siblings– offspring of patients with type 1A diabetes (SOCs) were class II genotyped and 293 NECs and 72 SOCs with the high-risk geno- type, DR3/4, DQB1*0302 have been prospectively eval- uated. Seventeen individuals who converted to persis- tent autoantibody positivity and two autoantibody- negative control groups (35 SOCs and 24 NECs) were typed for HLA-A class I alleles. The A1, A2 genotype was significantly increased among the autoantibody- positive subjects (47%) compared to autoantibody-neg- ative SOCs (14%, P  0.01) and NECs (13%, P  0.02). Life-table analysis of DR3/4, DQB1*0302 siblings re- vealed a risk of 75% for development of islet autoanti- bodies by the age of 2 years for those with A1, A2. The HLA-A2 phenotype frequency was increased among an independent DR3/4, DQB1*0302 young diabetes cohort (64% versus 33% for autoantibody-negative NECs). These results suggest that a high incidence and early appearance of islet autoantibodies for siblings of pa- tients with type 1A diabetes are associated with DR3/4, DQB1*0302 and potentially increased with HLA-A ge- notype A1, A2. © 2002 Elsevier Science (USA) Key Words: diabetes mellitus; autoimmunity; immu- nogenetics; HLA; MHC autoantibodies. INTRODUCTION Type 1A diabetes, formerly termed insulin-depen- dent diabetes mellitus (IDDM) is a genetically complex disorder resulting from immune-mediated destruction of the insulin-secreting beta cells. The disease is char- acterized by the presence of islet autoantibodies (e.g., anti-insulin, glutamic acid decarboxylase, and ICA512 (IA-2) autoantibodies) that typically precede the devel- opment of diabetes. Several prospective studies have reported that these autoantibodies can appear early in childhood and the presence of two or more of these autoantibodies is highly predictive for the development of diabetes (1, 2). The Diabetes Autoimmunity Study in the Young (DAISY) 3 investigates early genetic and en- vironmental factors contributing to risk of beta cell autoimmunity and progression to diabetes in relatives (siblings and offspring cohort (SOCs)) of patients with type 1A diabetes and a general population newborn cohort (NEC). Children with HLA-DR and DQ geno- types associated with diabetes risk are followed from birth for the development of islet autoantibodies. Genetic susceptibility for development of type 1A diabetes is associated with over 15 genetic loci, with the best characterized loci within the human leukocyte antigen (HLA) region on chromosome 6p21. Common alleles of class II HLA genes, in particular DQ and DR genes, are strongly associated with diabetes risk. To date, the highest risk genotype for development of type 1A diabetes is DQA1*0501, DQB1*0201 (DR3) with DQA1*0301, DQB1*0302 (DR4), which occurs in ap- proximately 35–50% of new onset type 1A diabetics compared to a population frequency of 2.4%. A number of studies suggest that other genes within the HLA region might influence diabetes susceptibility but the extensive linkage disequilibrium within the region makes the identification of these genes problematic. Recently, several reports in both humans and animal models of type 1A diabetes have suggested that alleles of class I genes (e.g., HLA-A) may contribute to diabe- tes risk and the age of diabetes development (3– 6). 1 Recipient of a 2001 Millennium Trainee Award. 2 To whom correspondence and reprint requests should be addressed at Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences Center, Campus Box B140, 4200 E. 9th Avenue, Den- ver, CO 80262. Fax: (303) 315-4892. E-mail: marian.rewers@uchsc.edu. 3 Abbreviations used: HLA, human leukocyte antigen; IAA, insulin autoantibodies; GADA, glutamic acid decarboxylase autoantibodies; 512A, ICA512 (IA-2) autoantibodies; SOC, sibling– offspring cohort; NEC, newborn nonrelative cohort; DAISY, Diabetes Autoimmunity Study in the Young. Clinical Immunology Vol. 102, No. 3, March, pp. 217–224, 2002 doi:10.1006/clim.2001.5171, available online at http://www.idealibrary.com on 1521-6616/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved. 217