Children at risk for type 1 diabetes showed fewer signs of beta-cell autoimmunity for up to 10 years if they were fed a highly hydrolyzed casein formula rather than conventional cow’s-milk–based formula during infancy, according to a report in the Nov. 11 issue of the New England Journal of Medicine.

This indicates that “a preventive dietary intervention aimed at decreasing the risk of type 1 diabetes may be feasible,” said Dr. Mikael Knip of the University of Helsinki, Finland, and his associates.

Their pilot study – the Trial to Reduce IDDM in the Genetically at Risk (TRIGR) – was not sufficiently powered to render a definitive conclusion about preventing progression to overt type 1 diabetes. However, a larger ongoing TRIGR study is now underway in 15 countries and is designed to address that issue, the investigators noted.

The pilot study involved 230 neonates born at 15 Finnish hospitals between 1995 and 1997 whose HLA genotypes showed susceptibility to type 1 diabetes and who had at least one first-degree relative with the disorder. The newborns were randomly assigned in a double-blind fashion to receive for at least 6 months either an extensively hydrolyzed casein-based formula (Nutramigen) or a control cow’s-milk–based formula made to smell and taste like the intervention formula.

Both formulas were offered only when breast milk was unavailable. Breast-feeding was encouraged, and mothers breast-fed at their own discretion and without modifying their diets.

Blood samples were obtained periodically to test for five disease-related autoantibodies: islet-cell antibodies; insulin antibodies; and antibodies to glutamic acid decarboxylase (GAD), the tyrosine phosphatase-related insulinoma-associated 2 molecule (IA-2), and zinc transporter 8 (ZnT8).

The intervention formula was associated with a significant decrease in risk of seropositivity for islet-cell antibodies, IA-2 autoantibodies, or to at least one of the five autoantibodies assessed, Dr. Knip and his colleagues said (N. Engl. J. Med. 2010;363:1900-8).

By 10 years of age, 6% of children in the intervention group and 8% of those in the control group developed type 1 diabetes. This difference was nonsignificant, but the study was not designed to demonstrate significance of this measure, they added.

Although preliminary data from small clinical studies and animal studies have suggested that “complex foreign-protein diets during weaning may play a deleterious role in the process leading to autoimmune diabetes,” the mechanism is not yet known. “We speculate that potential mechanisms may involve reduced gut permeability, induction of the maturation of regulatory T cells in the gut-associated lymphoid tissue, modification of the gut microflora, or some combination of these,” the investigators said.

If further study confirms that highly hydrolyzed formula safely protects high-risk children from developing type 1 diabetes, “a next step might be to expand the intervention to a wider infant population, since 83%-98% of children with newly diagnosed type 1 diabetes are from the general population,” Dr. Knip and his associates added.

This strategy could be a relatively easily implemented public health measure, they said.

This study was funded by the Academy of Finland, the European Commission, the Juvenile Diabetes Foundation International, the Novo Nordisk Foundation, and several other nonindustry sources. Infant formulas were provided by Mead Johnson Nutrition, which had no role in the study design, collection or analysis of data, or preparation of the manuscript. Dr. Knip’s associate, Dr. Anu-Maaria Hämäläainen, reported receiving a grant from the Foundation for Paediatric Research. Dr. Knip and the other associates reported no conflicts of interest.

Dietary Triggers and Type 1

This study’s principal finding was that at least one autoantibody developed in 17% of children who were fed a casein hydrolysate formula, compared with 30% of those fed a control formula, Dr. David M. Harlan and Dr. Mary M. Lee wrote in an editorial. Nevertheless, diabetes ultimately developed in a similar percentage of children in the two study groups.

However, the results may have been influenced by “the vagaries of the design of randomized trials,” wrote Dr. Harlan and Dr. Lee, both of the University of Massachusetts, Worcester.

First, infants in the control group were introduced to formula at a median age of 1.1 month, significantly earlier than infants in the intervention group (2.6 months). “Although statistical tools used to account for such differences suggest that the control formula resulted in more autoantibody seroconversions, the possibility that there were unmeasured confounders, such as the quantity of formula ingested, cannot be excluded,” they noted (N. Engl. J. Med. 2010;363:1961-3).

Second, three of the seven study subjects assigned to the intervention formula who later developed type 1 diabetes had actually dropped out of the study before receiving any formula. They were never tested for serum autoantibodies, but if they had been tested they likely would have been positive, effectively erasing the difference in the seroconversion rate between the two study groups.

Dr. Harlan reported ties to the American Diabetes Association, and Dr. Lee reported ties to Tolerx, Diamyd, and Baxter.

Copyright (c) 2010 Elsevier Global Medical News. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.