Effectiveness and safety of long-term treatment with sulfonylureas in patients with neonatal diabetes due to KCNJ11 mutations: an international cohort study

KCNJ11 mutations cause permanent neonatal diabetes through pancreatic ATP-sensitive potassium channel activation. 90% of patients successfully transfer from insulin to oral sulfonylureas with excellent initial glycaemic control; however, whether this control is maintained in the long term is unclear. Sulfonylurea failure is seen in about 44% of people with type 2 diabetes after 5 years of treatment. Therefore, we did a 10-year multicentre follow-up study of a large international cohort of patients with KCNJ11 permanent neonatal diabetes to address the key questions relating to long-term efficacy and safety of sulfonylureas in these patients. In this multicentre, international cohort study, all patients diagnosed with KCNJ11 permanent neonatal diabetes at five laboratories in Exeter (UK), Rome (Italy), Bergen (Norway), Paris (France), and Krakow (Poland), who transferred from insulin to oral sulfonylureas before Nov 30, 2006, were eligible for inclusion. Clinicians collected clinical characteristics and annual data relating to glycaemic control, sulfonylurea dose, severe hypoglycaemia, side-effects, diabetes complications, and growth. The main outcomes of interest were sulfonylurea failure, defined as permanent reintroduction of daily insulin, and metabolic control, specifically HbA1c and sulfonylurea dose. Neurological features associated with KCNJ11 permanent neonatal diabetes were also assessed. This study is registered with ClinicalTrials.gov, number NCT02624817. 90 patients were identified as being eligible for inclusion and 81 were enrolled in the study and provided long-term (>5·5 years cut-off) outcome data. Median follow-up duration for the whole cohort was 10·2 years (IQR 9·3–10·8). At most recent follow-up (between Dec 1, 2012, and Oct 4, 2016), 75 (93%) of 81 participants remained on sulfonylurea therapy alone. Excellent glycaemic control was maintained for patients for whom we had paired data on HbA1c and sulfonylurea at all time points (ie, pre-transfer [for HbA1c], year 1, and most recent follow-up; n=64)—median HbA1c was 8·1% (IQR 7·2–9·2; 65·0 mmol/mol [55·2–77·1]) before transfer to sulfonylureas, 5·9% (5·4–6·5; 41·0 mmol/mol [35·5–47·5]; p