Precision medicine in KCNJ11 permanent neonatal diabetes

One of the most important advances towards personalised genetic medicine for diabetes was the discovery that nearly half of neonatal diabetes is caused by de-novo (usually) or inherited mutations in the subunits of the pancreatic β-cell ATP-sensitive potassium (K_ATP) channel.¹ These activating mutations in the KCNJ11 and ABCC8 genes shift the probability of the K_ATP channels to have an open configuration, causing plasma membrane hyperpolarisation, thereby inhibiting insulin secretion, and causing permanent or transient neonatal diabetes. With most of these mutations, channel function can be restored by blockade with sulfonylurea tablets. This simple and low-cost treatment allows most patients to achieve substantially improved glycaemic control— with minimal hypoglycaemia—compared with insulin injections.^2,3 Identification and appropriate treatment of these individuals with sulfonylureas is also remarkably cost-effective over their lifetime.² In many cases, the K_ATP subunit mutations also cause a spectrum of sometimes severe neurological defects that often (but not always) correlate with the open probability of the mutant channels and the ability of sulfonylurea drugs to block the channel.³ Despite this profoundly insightful set of discoveries, some concern has remained about the safety and durability of sulfonylurea treatment in these patients, which often requires higher than approved doses and is given off-label in most countries when used in children.

In The Lancet Diabetes & Endocrinology, Pamela Bowman and colleagues⁴ report findings from long-term follow-up of an international cohort of patients with neonatal diabetes due to KCNJ11 mutations who were switched from insulin to sulfonylurea treatment. After about 10 years of follow-up, 75 (93%) of 81 participants remained on sulfonylurea therapy alone. Very good glycaemic control was maintained for patients for whom paired data were available on HbA_1c and sulfonylurea at all time points (ie, pre-transfer [for HbA_1c], year 1, and most recent follow-up; n=64), with median HbA_1c 6.4% at most recent follow-up, compared with 5.9% at 1 year and 8.1% before transfer to sulfonylurea therapy. There were no reports of severe hypoglycaemia in 809 patient-years of follow-up for the whole cohort. We agree that these positive long-term findings confirm this approach to be one of the best examples of precision genetic medicine in all medical practice.

In type 2 diabetes, sulfonylurea drugs are associated with hypoglycaemia and failure to durably maintain improvements in glucose control, and have also been associated with increased all-cause mortality and cardiovascular events in adults with type 2 diabetes.^5,6 Additional concerns exist about the use of glibenclamide specifically in type 2 diabetes.⁷ These concerns might contribute to reluctance among some clinicians to pursue sulfonylurea therapy in patients known to have KCNJ11 or ABCC8 mutations causing neonatal diabetes. However, compared with adults with type 2 diabetes, most individuals with K_ATP mutation-related diabetes are children or young adults who are non-obese and insulin sensitive, without confounding or contributory histories of coronary artery disease, hyperlipidaemia, hypertension, and renal insufficiency, among other issues more common in type 2 diabetes. The evidence that HbA_1c is improved by sulfonylurea treatment more than with insulin is robust,^8,9 and the results of Bowman and colleagues’ study,⁴ together with previous research,¹⁰ show that hypoglycaemia is mild and other side-effects of high-dose sulfonylureas are not clinically significant. Some evidence exists that sulfonylurea therapy might have a beneficial effect on the neurological features seen in patients with more severe mutations, but treatment might need to start at an early age and the benefit might be small.¹¹ Bowman and colleagues report that,⁴ despite some initial improvement in some patients, neurological features persisted in their international cohort of long-term sulfonylurea-treated patients.

Bowman and colleagues’ findings⁴ show that during more than 10 years of treatment there was no loss of the effect of sulfonylurea therapy on insulin secretion; rather, there was a small but detectable continued lowering of the dose (expressed as dose per kg bodyweight) necessary to maintain normal glycaemia. Why this might be the case is unclear, but several possibilities exist. It could be that there is continued improvement of the capacity of β cells for glucose-stimulated insulin secretion during sulfonylurea treatment, or that sulfonylureas, often hydrophobic in nature, can accumulate in tissues over time. This theory might also explain why we have heard anecdotally from patients that occasional doses can be missed without a prompt rise in blood sugar.

In individuals showing signs of developmental delay and neurological dysfunction due to K_ATP mutations, there might only be a modest effect on improving the neurological features with sulfonylurea drugs.^11–14 CNS effects are clearly variable in their response and might depend on getting the drug to affected individuals as early as possible,⁹ and in a way that overcomes the low concentration of sulfonylurea in the CNS itself.¹⁵

Overall, Bowman and colleagues’ findings⁴ illustrating the durability of sulfonylurea therapy without substantial off-target effects will be of general interest, not only to those following advances in this field, but as an exemplar of precision medicine. These results also suggest that there are important differences between sulfonylurea therapy in type 2 diabetes and neonatal K_ATP-related diabetes, the mechanisms of which remain to be fully elucidated.