Fatty acid status in infancy is associated with the risk of type 1 diabetes-associated autoimmunity

Aims/hypothesis We investigated the association of early serum fatty acid composition with the risk of type 1 diabetes-associated autoimmunity. Our hypothesis was that fatty acid status during infancy is related to type 1 diabetes-associated autoimmunity and that long-chain n-3 fatty acids, in particular, are associated with decreased risk. Methods We performed a nested case-control analysis within the Finnish Type 1 Diabetes Prediction and Prevention Study birth cohort, carrying HLA-conferred susceptibility to type 1 diabetes (n = 7782). Serum total fatty acid composition was analysed by gas chromatography in 240 infants with islet autoimmunity and 480 control infants at the age of 3 and 6 months. Islet autoimmunity was defined as repeated positivity for islet cell autoantibodies in combination with at least one of three selected autoantibodies. In addition, a subset of 43 infants with primary insulin autoimmunity (i.e. those with insulin autoantibodies as the first autoantibody with no concomitant other autoantibodies) and a control group (n = 86) were analysed. A third endpoint was primary GAD autoimmunity defined as GAD autoantibody appearing as the first antibody without other concomitant autoantibodies (22 infants with GAD autoimmunity; 42 infants in control group). Conditional logistic regression was applied, considering multiple comparisons by false discovery rate < 0.05. Results Serum fatty acid composition differed between breastfed and non-breastfed infants, reflecting differences in the fatty acid composition of the milk. Fatty acids were associated with islet autoimmunity (higher serum pentadecanoic, palmitic, palmitoleic and docosahexaenoic acids decreased risk; higher arachidonic: docosahexaenoic and n-6: n-3 acid ratios increased risk). Furthermore, fatty acids were associated with primary insulin autoimmunity, these associations being stronger (higher palmitoleic acid, cis-vaccenic, arachidonic, docosapentaenoic and docosahexaenoic acids decreased risk; higher a-linoleic acid and arachidonic: docosahexaenoic and n-6: n-3 acid ratios increased risk). Moreover, the quantity of breast milk consumed per day was inversely associated with primary insulin autoimmunity, while the quantity of cow's milk consumed per day was directly associated. Conclusions/interpretation Fatty acid status may play a role in the development of type 1 diabetes-associated autoimmunity. Fish-derived fatty acids may be protective, particularly during infancy. Furthermore, fatty acids consumed during breastfeeding may provide protection against type 1 diabetes-associated autoimmunity. Further studies are warranted to clarify the independent role of fatty acids in the development of type 1 diabetes.