No Deterioration in Glycemic Control in HNF-1α Maturity-Onset Diabetes of the Young Following Transfer From Long-Term Insulin to Sulphonylureas

Maturity-onset diabetes of the young (MODY) accounts for ∼1% of diabetes in the U.K. It is characterized by autosomal dominant inheritance of young-onset diabetes that is not insulin dependent. MODY is frequently misdiagnosed as type 1 diabetes and treated immediately with insulin due to its presentation with marked hyperglycemia in slim adolescents/young adults (1–4). With diagnostic molecular genetic testing now well established, it is possible to make a definitive diagnosis of specific genetic subtypes of MODY (5).

HNF-1α mutations account for ∼65% of U.K. MODY (MODY 3) cases. Patients with a mutation in HNF-1α are sensitive to the hypoglycemic action of sulfonylureas (6). This suggests that patients with HNF-1α MODY started on insulin from diagnosis could transfer to a sulfonylurea, as described in isolated cases (1, 2). However, there have been no systematic studies. HNF-1α MODY is characterized by β-cell dysfunction that deteriorates over time (7,8), which leads to some concern over taking patients off insulin, particularly after a prolonged period on this treatment.

We aimed to assess the short-term safety and effectiveness of transferring patients with HNF-1α mutations on insulin from diagnosis to sulfonylureas. The characteristics of the eight U.K. Caucasian patients were median age 34 years (range 17–48), median age of diagnosis 14 years (range 8–17), and median time on insulin 20 years (range 4–35); four patients had been on insulin for >27 years. The median dose of insulin was 0.5 U/kg (range 0.1–2.2). Insulin was stopped, and gliclazide 20 mg was started and rapidly increased to a maximum dose of 160 mg b.d. unless good control was achieved or the patient suffered hypoglycemia. Patients monitored blood glucose levels four times daily initially and tested their urine for ketones. Telephone support was provided by our national team based in Exeter, U.K., and local support was provided in four cases by MODY link nurses (9). HbA_1c was measured on insulin and after at least 2 months on sulfonylureas (median 6 months [range 2–11]).

All patients were able to discontinue insulin and were maintained on sulfonylureas without developing ketonuria or marked hyperglycemia. There was heterogeneity in response with the majority of patients (6 of 8) showing an improvement in control. The median reduction in HbA_1c following transfer to sulfonylureas was 0.8% (range −2.5 to 3.2) (P = 0.26). The median dose of gliclazide was 80 mg daily (range 20–320). Only one patient had a marked deterioration in HbA_1c (3.2%) on transfer to sulfonylureas. She had a long duration (35 years) of diabetes and may have coinherited type 2 diabetes genes from her father (who had type 2 diabetes), which can result in a more severe phenotype (10). All patients reported improvements in quality of life as a result of stopping insulin. The longest any of these patients has been off insulin is 18 months (range 7–18).

We conclude that transferring insulin-treated HNF-1α MODY patients to sulfonylureas was safe in the short term, even when patients have been on prolonged insulin treatment, and is not associated with a deterioration of control in most patients. Although the improved glycemic control seen in most patients may be partly attributed to increased blood glucose monitoring and/or attention from health care professionals, this could only be seen if treatment with sulfonylureas was effective. Sulfonylurea therapy should be considered in patients with HNF-1α MODY on insulin from diagnosis, but this should be closely monitored. This emphasizes the clinical utility of performing diagnostic molecular genetic testing. These preliminary short-term data need to be repeated in larger series with long-term follow-up. Many of these patients may require insulin again in the future.