Genetics and neonatal diabetes: towards precision medicine

934 www.thelancet.com Vol 386 September 5, 2015 Diabetes is a lifelong chronic disease. During the past 100 years, its diagnosis has been based on measurements of raised blood glucose concentrations. In the 1960s, diabetes was subclassifi ed based on age at onset and need for insulin treatment (ie, juvenile or maturity onset; insulin or non-insulinrequiring diabetes). Because diabetes was believed to be an inherited disease, much hope was placed on the identifi cation of genetic markers that would help to diagnose diabetic subgroups. Although investigators in the 1970s noted that type 1 diabetes was strongly associated with the HLA locus on chromosome 6, determination of HLA genotypes did not add substantial diagnostic value because of their high prevalence. The discovery of autoantibodies to diff erent islet antigens in the 1980s added strong discriminatory power to the diagnosis of autoimmune type 1 diabetes, and this knowledge was later applied to a late-onset autoimmune form of diabetes in adults. The fi rst real genetic breakthroughs in diabetes classifi cation came with the discovery that mutations in the genes encoding glucokinase, HNF1A, and HNF4A were associated with diff erent forms of maturity-onset diabetes of the young. Whereas maturity-onset diabetes of the young can show varying penetrance and severity, neonatal diabetes, a rare (1:100 000 births) severe form of diabetes, is diagnosed in infants younger than 6 months. The group in Exeter, UK, pioneered the genetic dissection of neonatal diabetes, and noted that one form could be linked to mutations in the KCNJ11 gene encoding the Kir6.2 subunit of the ATP-dependent potassium channel in pancreatic islets, and could be treated with sulfonylureas. During the past 20 years, more than 20 genes have been identifi ed as causing neonatal diabetes, as discussed by Elisa De Franco and colleagues in their accompanying study in The Lancet. In many of these monogenic diseases, a causal diagnosis has had an important eff ect on choice of treatment and disease outcome. In one striking case, after identifi cation of a mutation in the KCNJ11 gene in a poorly developing child with neonatal diabetes and switching from insulin to la rge doses of sulfonylurea, the child’s diabetes could not only be well controlled, but development, walking, and talking became possible. Kir6.2 is also expressed in the brain, and this combination of diabetes, developmental brain defects, and sometimes epilepsy has been called developmental delay-epilepsy-neonatal diabetes. Clear evidence exists of a genetic diagnosis improving treatment. In patients diagnosed with maturity-onset diabetes of the young, those with mutations in the glucokinase gene do not need any treatment because the mutation only modestly raises the threshold for the phosphorylating capacity of the enzyme, but the slight increase in glucose can fully overcome this defect. Therefore, maturity-onset diabetes of the young 2 caused by glucokinase mutations Genetics and neonatal diabetes: towards precision medicine We declare we have no competing interests. The Commissioners for the Lancet Commission on Dementia Care are David Ames, Clive Ballard, Sube Banerjee, Alistair Burns, Jiska Cohen-Mansfi eld, Claudia Cooper, Nick Fox, Laura Gitlin, Rob Howard, Helen Kales, Gill Livingston, Karen Ritchie, Ken Rockwood, Liz Sampson, Quincy Samus, Lon Schneider, Geir Selback, and Linda Teri.