Context: Children with initial autoantibodies to either insulin (IAA) or glutamic acid decarboxylase (GADA) differ in peak age of seroconversion and have different type 1 diabetes (T1D) risk gene associations, suggesting heterogeneity in the disease process. Objective: To compare the associations of age at seroconversion, HLA risk, and specificity of secondary autoantibodies with the progression of islet autoimmunity between children with either IAA or GADA as their first autoantibody. Design and methods: A cohort of 15,253 children with HLA-associated increased risk of T1D participated in a follow-up program in which islet autoantibodies were regularly measured. The median follow-up time was 6.7 years. Spearman correlation, Kaplan-Meier survival plots, and Cox proportional-hazard models were used for statistical analyses. Results: Persistent positivity for at least one of the tested autoantibodies was detected in 998 children; 388 of children progressed to clinical T1D. Young age at initial seroconversion was associated with a high probability of expansion of IAA-initiated autoimmunity and progression to clinical diabetes, whereas expansion of GADA-initiated autoimmunity and progression to diabetes were not dependent on initial seroconversion age. The strength of HLA risk affected the progression of both IAA- and GADA-initiated autoimmunity. The simultaneous appearance of two other autoantibodies increased the rate of progression to diabetes compared with that of a single secondary autoantibody among subjects with GADA-initiated autoimmunity but not among those with IAA as the first autoantibody. Conclusions: Findings emphasize the differences in the course of islet autoimmunity initiated by either IAA or GADA supporting heterogeneity in the pathogenic process.

Aims/hypothesis In previous studies, the risk of developing familial type 1 diabetes has been reported to be more than two times higher in the offspring of affected fathers than in those of affected mothers. We tested the hypothesis that index children with an affected father may have a more aggressive disease process at diagnosis than those with other affected first-degree relatives. Methods A cross-sectional, observational study was performed using the Finnish Pediatric Diabetes Register. Clinical and metabolic characteristics, beta cell autoantibodies and HLA class II genetics were analysed from index children in Finland diagnosed before the age of 15 years between January 2003 and December 2016. Information on the presence of type 1 diabetes in first-degree relatives was collected at diagnosis using a structured questionnaire. Results Out of 4993 newly diagnosed index children, 519 (10.4%) had familial type 1 diabetes. More than 5% (n = 253, 5.1%) had an affected father, 2.8% (n = 141) had an affected mother, 1.9% (n = 95) had an affected sibling and 0.6% (n = 30) had two or more affected family members. All clinical and metabolic variables were markedly poorer in children with sporadic vs familial diabetes. The index children with an affected father or mother were younger than those with an affected sibling (median age 7.59 vs 6.74 vs 10.73 years, respectively; p <0.001). After age- and sex-adjusted analyses, index children with an affected father presented more often with ketoacidosis (9.7% vs 3.6%; p = 0.033) and had greater weight loss before diagnosis (3.2% vs 0%; p = 0.006) than those with an affected mother. Children with familial disease tested negative for all autoantibodies more often (3.5% vs 2.1%; p = 0.041) and had insulin autoantibodies more frequently (49.8% vs 42.2%; p = 0.004) than those with sporadic disease. Both major HLA risk haplotypes (DR3-DQ2 and DR4-DQ8) were more often lacking among children with sporadic vs familial disease (15.9% vs 11.2%; p = 0.006). The DR4-DQ8 haplotype was more frequent in the familial vs the sporadic group (75.7% vs 68.5%; p = 0.001) and especially among children with an affected father when compared with children with sporadic disease (77.5% vs 68.5%; p <0.05). When comparing index children with affected parents diagnosed before or after the birth of the index child, a clear male preponderance was seen among the affected parents diagnosed before the birth of the index child (fathers 66.2% vs mothers 33.8%; p = 0.006), whereas the proportion of fathers and mothers was similar if type 1 diabetes was diagnosed after the birth of the index child. Conclusions/interpretation The more severe metabolic derangement at diagnosis in children with sporadic type 1 diabetes compared with those with familial type 1 diabetes was confirmed. The higher frequency of diabetic ketoacidosis and increased weight loss at diagnosis in index children with an affected father compared with an affected mother support the hypothesis that paternal type 1 diabetes is associated with more severe disease in the offspring than maternal diabetes. The sex difference seen between affected parents diagnosed before and after the birth of the index child supports the hypothesis that maternal insulin treatment protects against type 1 diabetes.

Objective The aim of this study was to explore the extended family history of type 1 diabetes in children at genetic risk and define the impact of a positive family history on the development of islet autoimmunity and type 1 diabetes. Methods The subjects were participants in The Finnish Type 1 Diabetes Prediction and Prevention (DIPP) study and carried increased HLA-conferred risk for type 1 diabetes. The case children (N = 343) were positive for at least one islet autoantibody, and the control children (N = 343) matched by age, gender and class II HLA genotype were negative for islet autoantibodies at the time of data collection. Extended family history of type 1 diabetes was obtained by using a structured questionnaire. Results Among children who were autoantibody positive and progressed to type 1 diabetes 62.2% (28/45) had at least one relative with type 1 diabetes. Interestingly, 57.8% of these children (26/45) had such a relative outside the nuclear family compared to 30.7% of children with no autoantibodies (P= .001), 35.2% of those with only classical islet cell antibodies (P= .006), and 35.2% of non-progressors with biochemical autoantibodies (P= 0.011). A positive history of type 1 diabetes in the paternal extended family was more common in children with multiple biochemical autoantibodies compared to those with only one biochemical autoantibody (P= .010). No association between the specificity of the first appearing autoantibody and family history of the disease was found. Conclusions Type 1 diabetes in relatives outside the nuclear family is a significant risk factor for islet autoimmunity and progression to clinical disease in HLA susceptible children.