The study aimed to compare the HLA specificities of AAb-positive healthy schoolchildren with those of patients with type 1 diabetes (T1D). HLA-DRB1 and DQB1 alleles were determined in 178 AAb-positive and 339 AAb-negative schoolchildren aged 6-17 years without first-degree relatives with T1D and in 274 patients with T1D. AAbs against glutamic acid decarboxylase (GADA), protein tyrosine phosphatase (IA-2A), and insulin (IAA) were determined by (125)I-antigen binding, and islet cell cytoplasmic antibodies (ICAs) immunohistochemically. Here, 82.6% (147/178) of AAb-positive schoolchildren had single AAbs and 17.4% (31/178) had multiple AAbs. In both groups, GADA occurred with highest and IAA with lowest frequency. In children with single AAbs at levels between the 99th and 99.9th percentile, frequencies of the diabetes-associated DRB1 (*03, *04) and DQB1 (*02, *0302) alleles and the protective DRB1 (*15) and DQB1 (*0602) alleles did not differ from those of controls. In patients, the positive associations were confirmed for DRB1*04 (OR = 5.39) and DQB1*0302 (OR = 9.05), whereas DRB1*15 (OR = 0.05) and DQB1*0602 (OR = 0.06) were negative-associated (p < 0.001). The same association was found in schoolchildren with multiple AAbs for DRB1*04 (OR = 3.84), DQB1*0302 (OR = 4.95), and DRB1*15 (OR = 0.1; p < 0.001-0.014), and with high-titer single AAbs (>/=99.9th percentile), but none of them had DQB1*0602. The highest risk genotype DQB1*02/*0302 occurred in 36.5% of patients (OR = 21.07) and in 19.3% of children with multiple AAbs (OR = 8.8; p<0.001). It is concluded that probands with multiple and high-titer single AAbs in the general population have the same genetic predisposition for T1D as patients and are therefore at highest risk for the disease.