Poster Presentation The 13th International Congress of the Immunology of Diabetes Society 2013

Homozygosity of the HLA-DRB1*03,DQB1*0201 or DRB1*04,DQB1*0302 haplotypes increases the metabolic severity of T1D onset. (#112)

Arleta Rewers 1 , Sunanda Babu 1 , Liping Yu 1 , Fran Dong 1 , Marian Rewers 1
  1. University of Colorado School of Medicine, Aurora, CO, United States

This study explored a potential effect of the major T1D susceptibility HLA haplotypes: DRB1*03,DQB1*0201 and DRB1*04,DQB1*0302 on the metabolic severity of T1D onset.

Between 1998-2012, 3544 Colorado residents were diagnosed with T1D before the age of 18 years and seen at the Barbara Davis Center for Childhood Diabetes in Denver. Medical records at diagnosis were informative in 3439 (97%) of these cases. Metabolic severity of diabetes at diagnosis was defined by presence/absence of diabetic ketoacidosis (DKA): venous pH< 7.3 or bicarbonate < 15 mmol/L, in presence of hyperglycemia and ketosis. Islet autoantibodies (mIAA, GADA, IA-2, ZnT8) were tested, generally within 2 weeks after diagnosis, to confirm T1a (autoimmune) diabetes. HLA-DR,DQ typing was performed using PCR SSOP in 2287 (67%) of the patients.

Children expressing neither the DRB1*03,DQB1*0201 nor the DRB1*04,DQB1*0302 haplotype were least likely to present in DKA (33.1%); at the highest risk were children homozygous for the DRB1*03,DQB1*0201 (46.9%) or DRB1*04,DQB1*0302 haplotype (40.4%) (Table).  Among children who did not present with DKA, HbA1c followed a similar trend, with the highest mean HbA1c levels (11.3%) in the DRB1*03,DQB1*0201 homozygotes.

Exclusion of children other than non-Hispanic white (n=568) or secondary cases within a family (n=7) did not affect the results. Presence of the DQB1*0602 allele (n=52) offered no protection from DKA if the other haplotype was DRB1*03,DQB1*0201 (33.3% vs 36.1%), but appeared to partially protect if the other haplotype was DRB1*04,DQB1*0302 (26.3% vs 40.2%, p=0.03).


Multivariate logistic regression, controlling for age, sex, family history of T1D, race/ethnicity and insurance status confirmed these findings. DKA was 56% and 47% more likely in, respectively, DRB1*03,DQB1*0201 and DRB1*04,DQB1*0302 homozygotes, than in children expressing neither of these haplotypes. 

T1D onset appears to be metabolically most severe in the DRB1*03,DQB1*0201 and DRB1*04,DQB1*0302 homozygotes, rather than in the DRB1*03,DQB1*0201/DRB1*04,DQB1*0302 heterozygotes thought conventionally to experience the fastest destruction of beta-cells.