substantially greater statistical power in a more heterogeneous study population compared with previous studies and, therefore, provides a more definitive answer to whether weaning to an extensively hydrolyzed formula might be protective of diabetes. This trial showed that among infants at risk for T1D, weaning to a hydrolyzed formula compared with a conventional formula did not reduce the cumulative incidence of T1D. Thus, the absolute risk of T1D was 8.4% among participants randomized to the casein hydrolysate vs 7.6% among those randomized to the conventional formula; the hazard ratio for T1D adjusted for HLA risk group, duration of breastfeeding, duration of study formula consumption, sex, and region while treating study center as a random effect was 1.1 (95% CI, 0.8 to 1.5; P = 0.46). The median age at diagnosis of type 1 diabetes was similar in the 2 groups (6.0 years [Q1–Q3, 3.1–8.9] vs 5.8 years [Q1–Q3, 2.6–9.1]) [19].
In conclusion, studies investigating cow's milk as an environmental factor show inconsistent results and do not currently support the need to revise the dietary recommendations for infants at risk for T1D.
Another outstanding question concerns the association between timing of gluten introduction and IA development. The BABYDIET study was an intervention study aimed to determine whether primary intervention through delayed introduction of dietary gluten was feasible and could reduce the incidence of IA in high‐risk first degree relatives of patients with T1D [20]. The study was based on the premise that introduction of foods containing gluten or cereal before the age of 3 months was associated with an increased risk of IA in childhood. New‐born children were eligible if they were younger than 3 months, were offspring or siblings of patients with T1D and had HLA genotypes that confer a high T1D risk. In this study, authors did not find a benefit in delaying gluten exposure with respect to autoimmunity associated with diabetes and celiac disease at 3 years of age.
The follow‐up findings of the BABYDIET study confirm previous results, failing to demonstrate that an intervention based on delayed gluten introduction will reduce the risk of developing autoimmunity related to T1D [21].
Conversely, in The Environmental Determinants of Diabetes in the Young (TEDDY) study, which prospectively followed 8676 children with increased genetic risk of T1D in the U.S., Finland, Germany, and Sweden, authors observed that later introduction of gluten‐containing cereals was associated with increased risk of any IA using Cox regression models [22]. In this study, the HRs for every 1‐month delay in gluten introduction were 1.05 (95% CI 1.01, 1.10; P = 0.02) and 1.08 (95% CI 1.00, 1.16; P = 0.04), respectively. The risk of IA associated with introducing gluten before 4 months of age was lower (HR 0.68; 95% CI 0.47, 0.99), whereas the risk of IA associated with introducing it after 9 months of age was higher (HR 1.57; 95% CI 1.07, 2.31) than introduction between 4 and 9 months of age. Interestingly, another sub‐analysis performed on TEDDY participants shows that administration of probiotics during the first 27 days of life reduced the risk for a first‐appearing β‐cell autoantibody in children with the HLA‐DQ2/8 genotype [23]. However, once again, authors acknowledged the need to confirm these results in further studies before any recommendation of dietary intake or probiotics use is made.
Also noteworthy are some observational studies suggesting a protective role of vitamin D and long‐chain n‐3 fatty acid against T1D by modulating the immune system. Nonetheless, clinical trials have been inconclusive so far [14, 24].
Finally, the Diabetes Prediction and Prevention Project (DIPP Study) (Clinical trial NCT03269084; www.clinicaltrials.gov) was a longitudinal study on T1D prediction and prevention carried out in the university hospitals of Turku, Tampere, and Oulu (Finland). The aim of the study was to investigate longitudinally the dietary factors in relation to the development of diabetic autoantibodies and clinical T1D. The diet of children was followed up by a structured questionnaire and by 3‐day dietary records at various ages. A food frequency questionnaire was applied for studying the dietary intake of pregnant and lactating mothers.
The aims of this project were: (1) to identify infants at increased genetic risk for T1D from the general population at birth; (2) to monitor such children for the appearance of diabetes‐associated autoantibodies, to identify those at high risk to develop clinical disease and to characterize the natural course of T1D; (3) to identify the environmental factors inducing the seroconversion to autoantibody‐positivity in children at increased genetic risk; and (4) to evaluate whether it is possible to delay or prevent progression to clinical T1D by daily administration of intranasal insulin. Whereas points 1–4 have been fulfilled and useful information has been obtained, the trial with intranasal insulin started soon after detection of autoantibodies, and did not show any beneficial effect of this treatment in preventing or delaying the disease [25].
In conclusion, since the failure of ENDIT and DPT1 trials (see secondary prevention) in preventing the onset of T1D in subjects who are β‐cell autoantibody positive, interest has switched to prevention trials starting before islet‐cell autoimmunity has developed. These primary prevention trials of T1D offer an exciting view of how our knowledge of the pathogenesis of this disease can lead to the possibility of intervening at birth. There is still a long way to go; however, the rationale is sound and the prospects seem good.
Secondary Prevention
Secondary prevention of T1D aims to reduce the incidence of the disease by stopping progression of β‐cell destruction in individuals with signs of such a process. This approach may represent a viable alternative to an actual cure by blocking the autoimmune response while β‐cell mass is still functional. For this reason, in the last few decades, prevention studies have started to focus on subjects at risk of developing the disease, in which the autoimmune process has begun but the β‐cell mass is still preserved. The most‐used approach at this stage of T1D is as an antigen‐specific therapy, based on insulin administration to establish tolerance. Generally, mechanistic long‐term effects of these treatments rely on effector T‐cell depletion with expansion of T regulatory cells (T‐reg) (Figure 2.2).
Three large multicenter trials of diabetes prevention in autoantibody‐positive relatives have been completed. The European Nicotinamide Diabetes Intervention Trial (ENDIT), the Diabetes Prevention Trial‐Type 1 (DPT‐1) and the Type 1 Diabetes Prediction and Prevention Study (DIPP). However, no significant benefits of these treatments on the prevention of clinical onset of T1D were found (Table 2.1).
European Nicotinamide Diabetes Intervention Trial (ENDIT)
The ENDIT study, conducted predominantly in Europe, examined whether nicotinamide could lead to a reduction in the rate of progression to T1D in at risk relatives of T1D probands. Over 40,000 first‐degree relatives aged 5–40 years were screened in centers in Europe and North America. The study was designed to recruit at least 422 subjects with ICA titers ≥ 20 JDF units to be randomized to either a nicotinamide‐ or a placebo‐treated group. With an expected rate of progression to diabetes of 40% in the placebo arm, the proposed 5‐year observation period should have allowed a 90% power to observe a 35% reduction in the incidence of disease [26, 27].
Nicotinamide treatment at the doses used did not show any significant effect on the primary outcome – progression to T1D. A total of 159 participants developed the disease within 5 years of randomization to treatment, 82 (30%) in the active treatment group and 77 (28%) in the placebo group. The unadjusted Cox proportional hazard estimate showed no difference between the placebo and nicotinamide groups on an intention to treat basis. Nor was any difference was found between groups after adjustment for age at baseline, glucose concentrations at 2‐h glucose in the OGTT, and number of islet autoantibodies. The proportion of relatives who developed diabetes within 5 years was almost identical in those treated with nicotinamide and those treated with placebo, and there was no suggestion of a treatment effect in any of the subgroups defined by well established markers of additional risk.
A useful message of this trial has been that large‐scale collaborations were essential