Five‐year follow‐up of incretin‐based therapy in a patient with KCNJ11‐related permanent neonatal diabetes: Achieving insulin discontinuation and sulfonylurea reduction

ABSTRACT

Permanent neonatal diabetes mellitus (PNDM) is commonly caused by pathogenic variants in KCNJ11 that impair insulin secretion through ATP‐sensitive potassium (K_ATP) channel dysfunction. Sulfonylureas (SUs) can stimulate insulin secretion and enable insulin discontinuation, but insulin withdrawal may remain difficult even under high‐dose SU, especially when therapy is initiated later in life. We describe a woman with KCNJ11‐related PNDM who achieved insulin discontinuation after adding sitagliptin, followed by further improvement with semaglutide during a five‐year follow‐up. HbA_1c remained between 6.1% and 6.5% on semaglutide 0.25 mg/week, with a 60% reduction in glibenclamide and preserved endogenous insulin secretion. A transient HbA_1c rise after semaglutide interruption, followed by improvement upon resumption, confirmed its reversible effect. No episodes of severe hypoglycemia or gastrointestinal adverse events occurred. This case suggests that combining insulin secretagogues targeting both K_ATP‐dependent (triggering: SU) and K_ATP‐independent (amplifying: incretin‐based therapy) pathways may provide additive or synergistic benefits in PNDM.

In a woman with KCNJ11‐related permanent neonatal diabetes mellitus, incretin‐based therapy enabled insulin discontinuation and maintained glycemic control for more than 5 years, with HbA_1c between 6.1% and 6.5% on the lowest semaglutide dose, dose‐dependent improvement in insulin secretion, and an approximately 60% reduction in glibenclamide dose.

INTRODUCTION

Permanent neonatal diabetes mellitus (PNDM), typically diagnosed within the first 6 months of life, is commonly caused by a pathogenic variant in the KCNJ11 gene, which encodes Kir6.2, a subunit of the ATP‐sensitive potassium (K_ATP) channel. This defect impairs insulin secretion and usually necessitates lifelong insulin therapy. Although sulfonylureas (SUs), which stimulate insulin secretion by closing K_ATP channels, can improve glycemic control and enable insulin discontinuation in many cases, some patients, especially those who initiate SU at an older age, still require insulin despite high‐dose SU treatment ¹ , possibly due to progressive loss of β‐cell function.

CASE REPORT

A woman with KCNJ11‐related PNDM was initially diagnosed with type 1 diabetes in infancy and treated with insulin ² . At age 13, genetic testing confirmed a heterozygous KCNJ11 mutation (c.754G>T, p.V252L). Glibenclamide (15 mg/day, twice daily) was initiated, reducing HbA_1c from 10.0% to 7.0%. However, HbA_1c later increased to 10.0–11.0% despite an insulin dose of 47 U/day (14 U degludec, 33 U aspart) and high‐dose SU (30 mg/day, twice daily) ² . At age 24, the addition of sitagliptin (50 mg/day), a dipeptidyl peptidase‐4 (DPP‐4) inhibitor, improved endogenous insulin secretion as assessed by the C‐peptide index, defined as (serum C‐peptide [ng/dL]/serum glucose [mg/dL]) × 100, measured under non‐fasting conditions without standardized meal timing or content and enabled insulin discontinuation. HbA_1c decreased from 11.2% to 7.4% within 3 months ² . Sitagliptin dose escalation to 100 mg/day enabled a further >1% reduction in HbA_1c, and switching to the once‐weekly glucagon‐like peptide‐1 receptor agonist semaglutide (from 0.25 to 0.5 mg) further reduced HbA_1c to 6.1% after 6 months ³ . The C‐peptide index improved dose‐dependently with sitagliptin and semaglutide, as previously reported ³ . These findings suggest that combining insulin secretagogues targeting both K_ATP‐dependent (triggering: SU) and K_ATP‐independent (amplifying: incretin‐based therapy) pathways may provide additive or synergistic benefits in PNDM. Here, we report the 5‐year follow‐up of this patient, comprising the previously reported 3 years and 2 additional years.

Over 2.5 years after switching to semaglutide, glycemic control remained stable, and HbA_1c consistently remained between 6.1% and 6.5% on the lowest dose of semaglutide (0.25 mg/week) (Figure 1a). HbA_1c transiently increased to 7.3% at month 59 following a patient‐initiated interruption of semaglutide during extended travel due to concerns about hypoglycemia but returned to 6.3% by month 62 after the treatment was resumed (Figure 1a–c). Glibenclamide was reduced to approximately 40% of the baseline dose used at incretin initiation (from 30 mg/day, twice daily, to 12.5 mg/day, once daily). The C‐peptide index remained stable throughout the clinical course, indicating preserved endogenous insulin secretion. Continuous glucose monitoring profiles, including intra‐ and inter‐day variability, also remained stable, with a mean coefficient of variation of 35% and mean time‐in‐range (70–180 mg/dL) of 70% during the most recent 12 months. Body weight remained at approximately 54 kg during sitagliptin treatment (height 158.6 cm, BMI 21.5 kg/m²). After switching to semaglutide (0.25 mg/week), body weight was 54.3 kg and decreased to 51.6 kg within 1 month after up‐titration to 0.5 mg/week, then remained around 50 kg during follow‐up (BMI 19.9 kg/m²). The patient did not report intentional weight loss or major changes in diet or physical activity during this period. She did not follow any specific diet and had been dancing regularly as exercise. No episodes of severe hypoglycemia were observed; CGM at month 61 showed 2% of time in the 54–69 mg/dL range and 0% of time in the <54 mg/dL range. No adverse gastrointestinal effects were observed. Treatment adherence remained high throughout the follow‐up period. No diabetic microvascular or macrovascular complications were observed during follow‐up up to month 62, and the complication status remained unchanged after initiation of semaglutide.

Figure 1.

Long‐term trajectory and continuous glucose monitoring profiles following incretin‐based therapy in KCNJ11‐related permanent neonatal diabetes. (a) Five‐year trajectories of HbA_1c and C‐peptide index. Data from months 0–35 were previously reported ¹ , ² and are shown for continuity, whereas months 36–62 represent new follow‐up data. The C‐peptide index was calculated as (serum C‐peptide concentration [ng/dL]/serum glucose concentration [mg/dL]) × 100, under non‐fasting conditions without standardized meal timing or content. (b) Ambulatory glucose profile at month 59, during a period that included a patient‐initiated interruption of semaglutide while traveling. (c) Ambulatory glucose profile at month 61, after resumption of semaglutide.

DISCUSSION

This report presents the longest documented follow‐up of incretin‐based therapy in KCNJ11‐related PNDM, extending beyond 5 years. HbA_1c remained between 6.1% and 6.5% after switching to the lowest semaglutide dose, with dose‐dependent improvement in insulin secretion and an approximately 60% reduction in glibenclamide compared with the initial dose. These findings suggest that high‐dose SU is not always necessary when combined with incretin‐based agents. Previous reports have demonstrated the effectiveness of incretin‐based therapies over 1–3 years ⁴ , ⁵ , ⁶ (Table 1); this case extends those observations to 5 years, demonstrating sustained efficacy and safety. The greatest HbA_1c reduction (−4.9%) among reported cases (typically 1.6–2.1%) was achieved despite the highest pre‐treatment insulin dose (47 U/day vs. 10–20 U/day), with successful discontinuation of insulin. The reason for this pronounced effect remains unclear. However, as previous cases were reported in European patients and this case involved a Japanese patient, ethnic differences in insulin secretion or incretin responsiveness may have contributed. No episodes of severe hypoglycemia or gastrointestinal side effects occurred during treatment. Furthermore, the transient increase in HbA_1c following patient‐initiated discontinuation of semaglutide, followed by improvement upon resumption, provides strong evidence of its reversible and causal role in glycemic control. By demonstrating the long‐term effectiveness and safety of combining agents that target both K_ATP channel‐dependent and K_ATP channel‐independent insulin secretory pathways, these findings offer new insights into conventional treatment strategies for PNDM, which have traditionally relied on insulin and high‐dose SU.

Table 1.

Previously reported cases of incretin‐based therapy in KCNJ11‐related permanent neonatal diabetes

Author (year)	Country	Age at initiation of IBT	IBT used	Duration of IBT use (months)	Glibenclamide dose (mg/day)	Insulin dose	HbA1c (%)	BW (kg)
Sastre J, et al. (2014) 4	Spain	19	Sita 100 mg/day	39	48* → 15.3	20* U/day → discontinued	8.1%* → 6.3%	60* → 51
Hindsø M, et al. (2021) 5	Denmark	40	Sita 100 mg/day → Lira (0.3 → 0.6 mg/day) → Sema (0.5 mg/week)	26 †	59.5 → 52.5	10–12 U/day → discontinued	8.1% → 6.5%	NA
Crowley M, et al. (2024) 6	Ireland	31	Lira (0.3 → 0.6 mg/day) → Sema (0.25 → 0.5 mg/week)	13	80 → 35	12 U/day → discontinued	8.6% → 6.5%	64.2 → 58
Present case	Japan	24	Sita (50 → 100 mg/day) → Sema (0.25 → 0.5 → 0.25 mg/w)	62	30 → 12.5	47 U/day → discontinued	11.2% → 6.3%	54 → 50

BW, body weight; IBT, incretin‐based therapy; Lira, Liraglutide; mg/day, milligrams per day; mg/week, milligrams per week; Sema, semaglutide; Sita, sitagliptin; U/day, units per day.

^*Baseline HbA_1c and insulin doses were obtained six months prior to IBT initiation.

^† Duration of sitagliptin use was not reported.

Incretin‐based therapy, especially once‐weekly semaglutide, improves glycemic control and offers additional benefits by enhancing treatment adherence and eliminating the need for multiple daily insulin injections. This approach may be particularly beneficial for patients who exhibit a partial response to SU after long‐term insulin therapy.

DISCLOSURE

The authors declare no conflicts of interest.

Approval of the research protocol: N/A.

Informed consent: Written informed consent was obtained from the patient.

Registry and the registration no. of the study/trial: N/A.

Animal studies: N/A.

DATA AVAILABILITY STATEMENT

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.