A Randomized Phase 3 Study Evaluating the Efficacy and Safety of Alogliptin in Pediatric Participants with Type 2 Diabetes Mellitus

Abstract

Introduction

There is an unmet need for pharmacological therapies for children with type 2 diabetes mellitus (T2DM). We assessed the efficacy and safety of an oral dipeptidyl peptidase-4 inhibitor, alogliptin, 25 mg once daily (QD), as a potential treatment for pediatric patients with T2DM.

Methods

This phase 3, 52-week, multicenter, randomized, double-blind, placebo-controlled trial was conducted in children and adolescents (10–17 years old) with T2DM. Participants had glycosylated hemoglobin (HbA1c) ≥ 6.5% at baseline (≥ 6.5% to < 11% without treatment or on metformin alone; ≥ 7.0% to < 11% on insulin alone or in combination with metformin). Where required, participants underwent prerandomization stabilization of their background metformin and/or insulin therapy. All received diabetes education and home glucose-monitoring training (during screening, prerandomization stabilization, and specified visits through week 26). Participants were then stratified based on previous antihyperglycemic therapy for 12 weeks before screening into schedule A (antihyperglycemic treatment-naïve) or B (metformin and/or insulin). The primary efficacy endpoint was change in HbA1c levels from baseline at week 26. Safety was assessed as a secondary endpoint at week 52.

Results

Overall, 152 participants (median age, 14 years; 68.9% female) were randomized (1:1) to receive either alogliptin (n = 75) or placebo (n = 77). The majority were white (58.3%), had a body mass index of ≥ 30 kg/m² (60.3%), and had received previous antihyperglycemic therapy (82.1%). The difference in HbA1c levels from baseline to week 26 between the alogliptin and placebo groups (least squares mean change [95% confidence interval]) was 0.10 (− 0.63, 0.83; p = 0.78). There was no difference in efficacy endpoints between alogliptin and placebo across both subgroups. No new safety concerns were observed with alogliptin treatment.

Conclusion

Alogliptin 25 mg QD did not significantly improve glycemic control versus placebo in pediatric patients with T2DM. Alogliptin treatment was safe and well tolerated, and no new safety concerns were observed in this study.

Trial Registration

ClinicalTrials.gov: NCT02856113; EudraCT: 2015-000208-25.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13300-025-01700-3.

Key Summary Points

Why carry out this study?

Multiple treatment options were available for managing adults with type 2 diabetes mellitus (T2DM); however, the range of options, especially oral agents, is relatively limited for children and youth with T2DM

This study evaluated the efficacy and safety of alogliptin 25 mg once daily (QD) in pediatric patients with T2DM

What was learned from the study?

Alogliptin 25 mg QD did not significantly improve glycemic control compared with placebo in pediatric patients with T2DM. This is consistent with other dipeptidyl peptidase-4 (DPP-4) inhibitors, such as sitagliptin, linagliptin, and saxagliptin, suggesting that the DPP4 inhibitor class is unlikely to be effective in treating pediatric T2DM

Alogliptin treatment was safe and well tolerated, and no new safety concerns were observed

This study provides valuable insights into the use of DPP-4 inhibitors in the pediatric population and underscores the need for continued research and exploration of various treatment approaches to better manage pediatric patients with T2DM

Introduction

The incidence of type 2 diabetes mellitus (T2DM) has been increasing among children and adolescents, mainly because of the global increase in childhood obesity [1–3]; this concerning trend is projected to continue [4]. Pediatric T2DM is also associated with the rapid emergence of complications and comorbidities, including diabetic neuropathy, diabetic kidney disease, diabetic retinopathy, dyslipidemia, depression, and increased cardiovascular-related events, indicating a major economic burden in later life [4, 5].

There are multiple approved pharmacological treatment options available for managing adults with T2DM [6–8]. However, the range of options is relatively limited for children and youth with T2DM compared with that for adults [9]. Until recently, the only approved treatments in the United States (US) and Europe for pediatric patients with T2DM were metformin and insulin [8]. Although metformin is effective in many patients, it has certain limitations. Metformin treatment is often associated with gastrointestinal (GI) side effects, affecting approximately 20–30% of patients and negatively affecting quality of life and adherence [10]. Furthermore, using metformin alone may not maintain durable glycemic control in children; consequently, many youth with T2DM may require combination treatment or insulin therapy relatively quickly after diagnosis [11]. Insulin therapy is a mainstay treatment option with a long history of safe and effective use in the clinical setting for T2DM [12]. However, insulin treatment often involves multiple daily injections and adverse events (AEs), such as weight gain and hypoglycemia, which may limit its acceptability and compliance in younger patients [12]. It is also important to highlight that, to date, no pivotal clinical trial has been conducted with insulin exclusively in pediatric patients with T2DM [8].

To address the unmet need for more effective treatments in children, researchers have explored alternative options for managing pediatric T2DM. Injectable glucagon-like peptide-1 receptor agonists (GLP1-RAs) liraglutide, exenatide, and dulaglutide have been approved and licensed for the treatment of pediatric T2DM since 2019, 2021, and 2022, respectively [13–15]. However, identifying effective oral agents would provide children with additional therapeutic options and potentially improve their adherence to treatment plans. At the time of the initiation of this study, there was an unmet need to explore alternative oral therapies as an add-on or alternative to metformin or insulin for pediatric T2DM.

Alogliptin is an oral, highly selective dipeptidyl peptidase-4 (DPP-4) inhibitor with an established antihyperglycemic profile indicated for the treatment of T2DM in adults [16]. A pharmacokinetic (PK)/pharmacodynamic (PD) study conducted with alogliptin (25 mg) in pediatric patients aged 10–17 years reported that this dose achieved alogliptin exposure and DPP-4 inhibition levels similar to those observed in adult patients with T2DM. Importantly, no safety concerns were identified [17]. Consequently, a 25-mg dose of alogliptin was recommended for study in pediatric phase 3 trials [17].

This study evaluated the use of alogliptin 25 mg once daily (QD) in pediatric patients with T2DM, both as a standalone treatment and in combination with metformin, insulin, or both. The objective of this phase 3 trial was to assess the efficacy of alogliptin on glycemic control and demonstrate the safety of alogliptin compared with placebo in youth with T2DM.

Methods

Participants

This was a phase 3, multicenter, randomized, double-blind, placebo-controlled, 52-week study in children and adolescents aged 10–17 years with a confirmed diagnosis of T2DM. Diagnosis was based on the American Diabetes Association and World Health Organization criteria: laboratory determinations of fasting plasma glucose (FPG) ≥ 126 mg/dl, random glucose ≥ 200 mg/dl (≥ 11.10 mmol/l), glycosylated hemoglobin (HbA1c) ≥ 6.5%, or 2-h oral glucose tolerance test glucose ≥ 200 mg/dl. Additional criteria included an HbA1c level of ≥ 6.5% to < 11.0% if the participant was treatment-naïve or on metformin alone or ≥ 7.0% to ≤ 11.0% if the participant was on insulin therapy alone or in combination with metformin. Additionally, in participants with a diagnosis of T2DM for < 1 year and/or who were taking insulin before randomization, islet antigen-2 antibody levels were required to be below the upper limit of normal reference range at randomization. Key exclusion criteria were a confirmed diagnosis of type 1 diabetes mellitus or maturity-onset diabetes in youth, history of hemoglobinopathy, bariatric surgery, more than one episode of pancreatitis, or an episode of diabetic ketoacidosis or pancreatitis at any time after T2DM diagnosis. A serum creatinine level of ≥ 1.5 mg/dl for male patients or ≥ 1.4 mg/dl for female patients or creatinine clearance of < 60 ml/min using the Schwartz formula [18] for estimated glomerular filtration rate at the screening visit were also key exclusion criteria for the study. Other key exclusion criteria are listed in Supplementary Table S1.

Study Design

The study included a screening period of up to 2 weeks. Participants who met all entry criteria proceeded to randomization on day 1. The study also included a prerandomization stabilization period, which allowed participants a period of time to wash out prior medications and establish a stable regimen of ongoing metformin and/or insulin treatment. For participants on metformin, the stabilization period enabled them to reach their maximum tolerated dose (MTD) with a treatment goal of at least 1000 mg twice daily (BID). For participants already receiving insulin alone or in combination with their MTD of metformin, the stabilization period necessitated a stable insulin regimen, defined as a period of at least 1 month with no insulin dose changes, to prevent hypoglycemia or symptomatic hyperglycemia. Insulin doses may have been reduced or discontinued in those who maintained HbA1c levels of < 8.0%. Conversely, participants with HbA1c levels ≥ 11.0% on the MTD of metformin alone were initiated on insulin therapy at the discretion of the investigator.

Once the participants met the objectives of the prerandomization stabilization period to achieve the desired HbA1c levels for study entry, they were reassessed to confirm their qualification for all other study inclusion and exclusion criteria. Additional randomization criteria for participants included in the prerandomization stabilization period are shown in Supplementary Table S2.

Participants were assigned to one of two groups based on their previous antihyperglycemic therapy usage for the 12 weeks before the screening period. The schedule A group included participants who were naïve to antihyperglycemic therapy, and the schedule B group included participants who were previously treated with metformin and/or insulin. On study day 1, eligible participants were randomly assigned in a 1:1 ratio to one of two treatment groups: alogliptin 25 mg QD or matching placebo (Fig. 1). Participants were required to maintain a continuous stable dose of their background antihyperglycemic therapy (metformin and/or insulin, if applicable) throughout the first 26 weeks of the double-blind treatment period. In addition, participants received diabetes education and fingerstick home glucose-monitoring training during the prerandomization stabilization period and during the first 26 weeks of treatment, which included instructions on proper nutrition and exercise, recognition of signs and symptoms of hypoglycemia, and the use of the glucometer to enable them to maintain blood glucose diaries for each study visit. A detailed description of the diabetes education and home glucose-monitoring training is available in the Supplementary Appendix in the electronic supplementary material.

Fig. 1.

Study design. DB double-blind, HbA1c glycosylated hemoglobin, Med medication, QD once daily, Wk week

The study (ClinicalTrials.gov: NCT02856113; EudraCT: 2015-000208-25) was conducted in accordance with the Declaration of Helsinki, International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use, Guidelines for Good Clinical Practice, and all applicable laws and regulations. The study was approved by the appropriate regulatory agencies and institutional review boards (IRBs)/ethics committees (IRB Protocol ID: 2,000,022,953). Written informed consent was provided by the parents or legal guardians, and the children/adolescents provided assent to participate in the study.

Study Objectives and Efficacy Endpoints

The primary study objective was to evaluate the efficacy of alogliptin (25 mg QD) compared with placebo when administered as monotherapy or when added to a background of metformin alone, insulin alone, or a combination of metformin and insulin, as measured by the change from baseline in HbA1c levels at week 26. The secondary objective was to evaluate the change from baseline in HbA1c levels after treatment with alogliptin compared with placebo at weeks 12, 18, 39, and 52. Additionally, safety was evaluated for an additional 26 weeks through week 52 by the collection of treatment-emergent AEs (TEAEs) and AEs of special interest (AESIs; serious hepatic abnormalities, pancreatitis, infections [including urinary tract infections], severe hypersensitivity reactions [including angioedema, anaphylaxis, and Stevens-Johnson syndrome], and hypoglycemic events). Finally, clinical laboratory parameters, electrocardiogram readings, physical examinations, vital signs, and compliance rates were recorded. Exploratory analyses were also conducted to evaluate the incidence and time of hyperglycemic rescue and plasma concentrations of alogliptin.

PK Analysis

To assess the time course of alogliptin plasma concentrations, five PK samples were collected from each participant. The plasma concentrations of alogliptin were analyzed using population PK (PopPK) modeling. A two-compartment PopPK model with first-order absorption and elimination with fixed body weight-based allometric scaling was developed from the PK data from the phase 1 single-dose study to characterize the time course of alogliptin plasma concentrations [17]. This model was then used to compute individual estimates of the steady-state area under the curve to the end of the dosing period (AUC_0–t) and maximum serum concentration (C_max) of participants receiving alogliptin 25 mg QD dosing.

Statistical Analysis

A total of 150 children/adolescents were planned to be randomized in a 1:1 ratio into the two treatment groups independent of schedule A or B. This would ensure at least 90% power to detect a difference in mean change from baseline in HbA1c levels at week 26 between alogliptin and placebo, assuming a treatment effect of 0.5%, standard deviation (SD) of 0.9%, and two-sided false rejection rate of 5% [19, 20]. The drop-out rate was taken into consideration in the sample size calculation to ensure sufficient power to detect the intended treatment effects. The primary analysis was conducted using the full analysis set, which included all randomized participants who received at least one dose of the study treatment. A mixed model for repeated measures (MMRM) analysis was used for the assessment of the primary efficacy and secondary endpoints used for statistical inference, where appropriate. An unstructured covariance matrix was used to model the correlation among repeated measurements; parameter estimates were calculated using restricted maximum likelihood estimation. In a scenario where the unstructured covariance matrix sometimes caused a nonconvergence issue in computational practice, it was replaced with a compound symmetry covariance matrix. Degrees of freedom were estimated using the Kenward-Roger approximation, and the Shapiro-Wilk test was used to test the normality of the residuals from each continuous endpoint that was examined using an MMRM model. If the hypothesis of normality was rejected, the natural logarithm was applied to the measurements at each time point, the model was refit, and model-based quantities were back-transformed and expressed on the ratio scale. The number of participants was summarized by prior antihyperglycemic therapy, either schedule A or B. Data collected following discontinuation of double-blind study medication or hyperglycemic rescue were not included in the efficacy analysis. The population for the safety analysis included all participants who took at least one dose of the study medication. The incidence of hyperglycemic rescue events was analyzed using logistic regression modeling. All statistical analyses were conducted using SAS^® (NC, USA), version 9.4 or higher.

Results

Participant Disposition

A total of 285 participants were screened globally across six countries (Brazil, Israel, Italy, Mexico, the Russian Federation, and the US) at 37 sites between October 2016 and August 2021, and 152 children/adolescents were randomized to receive treatment (alogliptin, n = 75 and placebo, n = 77). Most participants had a pre-randomization stabilization period of < 100 days. The prerandomization stabilization period distribution was similar between the placebo and alogliptin groups. Participants not treated with metformin or insulin were not included before randomization. One participant in the placebo group was randomized in error because the inclusion criterion for HbA1c was not met; however, the participant was withdrawn after being randomized and thus did not receive the double-blind study medication. Participant disposition is shown in Fig. 2. A total of 126 (82.9%) participants completed all study visits, 25 (16.4%) discontinued the study treatment, and 22 (14.5%) discontinued the study.

Fig. 2.

Participant disposition. AE adverse event

Demographics and Baseline Characteristics

The median age of the randomized participants was 14.0 years overall (Table 1). Of the 151 participants treated, 51 (33.8%) were aged 10–13 years, and 100 (66.2%) were aged 14–17 years. In the overall population, most participants were female (n = 104, 68.9%) and white (n = 88, 58.3%). Ethnicity was not recorded in 79 participants (52.3%), but in those in whom ethnicity was recorded, most were not Hispanic or Latino (n = 57, 37.7%). The mean body mass index (BMI) was 33.7 kg/m² overall and similar across both treatment groups, with most participants having BMI ≥ 30 kg/m² (overall: n = 91, 60.3%; alogliptin: n = 44, 58.7%; placebo: n = 47, 61.8%). The mean (SD) HbA1c level was 8.1% (1.42) among all participants, ranging from 5.7% to 11.5%. The mean (SD) baseline blood glucose level was 8.4 (3.31) mmol/l among all patients, and it was slightly higher in the alogliptin group at 8.8 (3.49) mmol/l vs. the placebo group at 8.1 (3.11) mmol/l. In terms of previous antihyperglycemic therapy, most participants entered the study on antihyperglycemic therapy (n = 124, 82.1%) and were assigned to schedule B. Only 27 patients (17.9%) were naïve to antihyperglycemic therapy when they entered the study and were assigned to schedule A. The mean (SD) overall compliance rates, measured by the number of unused medications at each study visit, for the alogliptin and placebo treatment groups were similar at 78.8% (10.58) and 72.7% (14.28), respectively.

Table 1.

Demographic and baseline characteristics (SAS)

Parameter	Alogliptin 25 mg QD (n = 75)	Placebo (n = 76)	Total (N = 151)
Age (years), median	14.0	14.0	14.0
Age (years), n (%)
10–13	23 (30.7)	28 (36.8)	51 (33.8)
14–17	52 (69.3)	48 (63.2)	100 (66.2)
Sex, n (%)
Male	22 (29.3)	25 (32.9)	47 (31.1)
Female	53 (70.7)	51 (67.1)	104 (68.9)
Race, n (%)
American Indian or Alaska Native	11 (14.7)	14 (18.4)	25 (16.6)
Asian	0	1 (1.3)	1 (0.7)
Black or African American	16 (21.3)	16 (21.1)	32 (21.2)
White	44 (58.7)	44 (57.9)	88 (58.3)
Multiraciala	4 (5.3)	1 (1.3)	5 (3.3)
Ethnicity, n (%)
Hispanic or Latino	9 (12.0)	6 (7.9)	15 (9.9)
Not Hispanic or Latino	26 (34.7)	31 (40.8)	57 (37.7)
Missing	40 (53.3)	39 (51.3)	79 (52.3)
BMI (kg/m2), mean (SD)	33.7 (8.7)	33.6 (7.9)	33.7 (8.2)
BMI categories (kg/m2), n (%)
< 25	6 (8.0)	7 (9.2)	13 (8.6)
25– < 30	25 (33.3)	22 (28.9)	47 (31.1)
≥ 30	44 (58.7)	47 (61.8)	91 (60.3)
Baseline HbA1c (%), mean (SD)	8.16 (1.513)	8.11 (1.330)	8.13 (1.420)
Baseline blood glucose (mmol/l), mean (SD)	8.8 (3.49)	8.1 (3.11)	8.4 (3.31)
Glomerular filtration rate (ml/min/1.73 m2), mean (SD)	147.3 (36.9)	139.9 (40.5)	143.6 (38.8)
Albumin/creatinine (mg/mmol creatinine), mean (SD)	4.4 (7.5)	5.3 (6.5)	4.8 (7.1)
Cholesterol (mmol/l), mean (SD)	4.1 (0.9)	4.2 (0.9)	4.1 (0.9)
HDL cholesterol (mmol/l), mean (SD)	1.0 (0.3)	0.9 (0.2)	1.0 (0.2)
LDL cholesterol (mmol/l), mean (SD)	2.6 (0.7)	2.6 (0.7)	2.6 (0.7)
Triglycerides (mmol/l), mean (SD)	1.8 (1.0)	1.9 (1.1)	1.8 (1.1)
Diastolic blood pressure (mmHg), mean (SD)	74.6 (10.14)	72.5 (9.89)	73.5 (10.03)
Systolic blood pressure (mmHg), mean (SD)	118.6 (11.83)	117.3 (11.94)	117.9 (11.86)
Prior antihypertensive medications, n (%)	6 (8.0)	9 (11.8)	15 (9.9)
Prior antihyperglycemic therapy, n (%)
Schedule A (naïve to antihyperglycemic therapy)	13 (17.3)	14 (18.4)	27 (17.9)
Schedule B (entered study on antihyperglycemic therapy)	62 (82.7)	62 (81.6)	124 (82.1)

BMI body mass index, HbA1c glycosylated hemoglobin, HDL high-density lipoprotein, LDL low-density lipoprotein, QD once daily, SAS safety analysis set, SD standard deviation

^aEach participant marked with more than one race was coded as multiracial

Efficacy Results

Primary and Secondary Endpoints

No statistically significant difference was observed between participants following treatment with alogliptin compared with placebo for the primary efficacy endpoint of HbA1c change from baseline to week 26. The least squares (LS) mean (95% confidence interval [CI]) of the difference between alogliptin and placebo in HbA1c change from baseline was 0.10 (− 0.63, 0.83; p = 0.78; Table 2 and Fig. 3a). A similar result was observed across all subgroups, regardless of previous antihyperglycemic therapy, sex, age, race, or BMI category (p = not significant for all subgroups; Fig. 3b). Moreover, no statistically significant differences were observed between treatment with alogliptin and placebo in the secondary endpoints of HbA1c change from baseline at weeks 12, 18, 39, and 52 (Table 2 and Fig. 3a).

Table 2.

Efficacy endpoints—HbA1c percent change from baseline (FAS)

	Alogliptin 25 mg QD (n = 75)	Placebo (n = 76)	p value
Baseline
Mean (SD)	8.2 (1.5)	8.1 (1.3)
Week 12
n	62	62
Mean (SD)	7.9 (2.13)	8.1 (2.0)
Change from baselinea	− 0.36 (0.22)	− 0.32 (0.22)
Between-group differenceb	− 0.05 (0.26; − 0.57, 0.48)		0.86
Week 18
n	54	61
Mean (SD)	8.1 (2.4)	8.1 (2.2)
Change from baselinea	− 0.20 (0.25)	− 0.21 (0.25)
Between-group differenceb	0.004 (0.31; − 0.61, 0.62)		0.99
Week 26
n	54	56
Mean (SD)	7.9 (2.1)	8.2 (2.4)
Change from baselinea	0.09 (0.28)	− 0.01 (0.28)
Between-group differenceb	0.10 (0.37; − 0.63, 0.83)		0.78
Week 39
n	49	47
Mean (SD)	7.7 (2.1)	8.4 (1.8)
Change from baselinea	0.09 (0.29)	0.50 (0.28)
Between-group differenceb	− 0.41 (0.37; − 1.14, 0.32)		0.26
Week 52
n	39	39
Mean (SD)	7.6 (2.0)	8.3 (2.2)
Change from baselinea	0.28 (0.32)	0.76 (0.31)
Between-group differenceb	− 0.48 (0.42; − 1.31, 0.34)		0.25

All values are presented as mean (SD)

CI confidence interval, FAS full analysis set, HbA1c glycosylated hemoglobin, LS least squares, QD once daily, SD standard deviation, SE standard error

^aLS mean SE

^bDifference between alogliptin (25 mg QD) vs placebo in LS indicates SE (95% CI)

Fig. 3.

a HbA1c percent change from baseline over time by treatment group; b forest plot of the estimator for difference in mean HbA1c change from baseline to week 26 for the primary analysis and subgroups. Patients in schedule A were naïve to antihyperglycemic therapy, and patients in schedule B were enrolled in the study on antihyperglycemic therapy (prior treatment with metformin and/or insulin). Negative LS mean difference values indicate alogliptin advantage. BMI body mass index, CI confidence interval, F female, HbA1c glycosylated hemoglobin, LS least squares, M male

Exploratory Analyses

Incidence and Time to Hyperglycemic Rescue

Overall, 39 (25.7%) patients required hyperglycemic rescue. Among them, 17 of 75 (22.7%) and 22 of 77 (28.6%) patients in the alogliptin and placebo groups, respectively, required hyperglycemic rescue during the study treatment. No statistically significant difference was observed between the treatment groups in the incidence of hyperglycemic rescue events over any duration assessed. Furthermore, no statistically significant difference was reported between the treatment groups for the time to first hyperglycemic rescue event (alogliptin vs placebo hazard ratio [95% CI]: 0.75 [0.40, 1.41]; p = 0.37).

PopPK Analysis

The PopPK model demonstrated a linear PK profile for alogliptin 25 mg QD, which closely aligns with previously reported results in both adult and pediatric populations [17]. The use of background antihyperglycemic therapies of metformin, insulin, or their combination had no discernible impact on the PK concentration of alogliptin 25 mg QD. Similarly, sex, race, dose, and estimated glomerular filtration rate did not exhibit any significant influence on the PK profile. However, body weight was found to be a significant predictor of the alogliptin PK concentration. The geometric means of individual estimates for steady-state AUC_0–t and C_max from the final PopPK model were calculated to be 1158 ng·h/ml/h and 110 ng/ml, respectively (Table 3).

Table 3.

Estimates of the steady-state exposure of alogliptin 25 mg QD

Statistic	AUC0–t (ng·h/ml)	Cmax (ng/ml)
N	74	74
Mean (SD)	1380 (1240)	118 (54.4)
Geometric mean (%CV)	1158 (54.6)	110 (37.6)
95% CI of geometric mean	1030, 1301	101, 119
5th percentile	577	65.4
10th percentile	663	73.8
25th percentile	880	92.1
50th percentile	1110	109
75th percentile	1340	131
90th percentile	1940	148
95th percentile	2360	176

AUC_0–t area under the curve to the end of the dosing period, CI confidence interval, C_max maximum serum concentration, CV coefficient of variation, QD once daily, SD standard deviation

Safety and Tolerability

Safety events were generally similar in the alogliptin and placebo groups at week 52 (Table 4). Overall, 431 AEs were reported in 118 patients (78.1%) during the study, and most AEs (390 events; 90.5%) were not related to the study treatment. Only one hyperglycemia event was considered related to the study procedure in the alogliptin group. Most events had mild (311; 72.7%) or moderate (109; 25.3%) intensities. The most common (≥ 1.0%) TEAEs related to study treatment were diarrhea (3.3% overall; 1.3% alogliptin; 5.3% placebo), C-telopeptide increase (2.6% overall; 2.7% alogliptin; 2.6% placebo), alanine aminotransferase increase (1.3% overall; 1.3% alogliptin; 1.3% placebo), and aspartate aminotransferase increase (1.3% overall; 0.0% alogliptin; 2.6% placebo). No deaths occurred during the study period. Five participants (3.3%) reported 12 serious AEs (SAEs) overall: five events in two participants in the alogliptin group and seven events in three participants in the placebo group. No SAEs were reported to be related to the study treatment or study procedure. One SAE leading to study treatment dose interruption was reported for one (1.3%) participant in each treatment group, and three SAEs leading to treatment withdrawal were reported for one (1.3%) participant in the alogliptin group. Overall, no meaningful difference in the frequency of TEAEs was observed between the treatment groups. The overall incidence of AESIs (hypersensitivity, pancreatitis, hepatic, and hypoglycemic events) was low, with 30 events reported for 20 of 151 treated patients (13.2%), which were comparable between the alogliptin and placebo treatment groups (9 participants, 12%; 11 participants, 14.5%, respectively). No new safety concerns for treatment with alogliptin were observed, and the overall benefit-risk profile remained unchanged from that observed in adults.

Table 4.

Overview of TEAEs

Parameter	Alogliptin 25 mg QD (n = 75)	Placebo (n = 76)	Total (N = 151)
Total number of AEs	205	226	431
AEs related to study treatment	20	21	41
Total number of SAEs	5	7	12
SAEs related to study treatment	0	0	0
Patients with any TEAE	60 (80.0)	58 (76.3)	118 (78.1)
Infections and infestations	28 (37.3)	26 (34.2)	54 (35.8)
Metabolism and nutrition disorders	17 (22.7)	23 (30.3)	40 (26.5)
Gastrointestinal disorders	15 (20.0)	17 (22.4)	32 (21.2)
Nervous system disorders	17 (22.7)	11 (14.5)	28 (18.5)
Investigations	11 (14.7)	11 (14.5)	22 (14.6)
Injury, poisoning and procedural complications	6 (8.0)	8 (10.5)	14 (9.3)
Skin and subcutaneous tissue disorders	5 (6.7)	7 (9.2)	12 (7.9)
General disorders and administration site conditions	9 (12.0)	2 (2.6)	11 (7.3)
Musculoskeletal and connective tissue disorders	5 (6.7)	6 (7.9)	11 (7.3)
Reproductive system and breast disorders	7 (9.3)	3 (3.9)	10 (6.6)
Respiratory, thoracic, and mediastinal disorders	5 (6.7)	4 (5.3)	9 (6.0)
Renal and urinary disorders	5 (6.7)	3 (3.9)	8 (5.3)
Psychiatric disorders	1 (1.3)	5 (6.6)	6 (4.0)
Immune system disorders	2 (2.7)	3 (3.9)	5 (3.3)
Blood and lymphatic system disorders	3 (4.0)	1 (1.3)	4 (2.6)
Vascular disorders	2 (2.7)	2 (2.6)	4 (2.6)
Cardiac disorders	1 (1.3)	2 (2.6)	3 (2.0)
Eye disorders	2 (2.7)	1 (1.3)	3 (2.0)
Ear and labyrinth disorders	1 (1.3)	1 (1.3)	2 (1.3)
Hepatobiliary disorders	1 (1.3)	1 (1.3)	2 (1.3)
Congenital, familial, and genetic disorders	1 (1.3)	0	1 (0.7)
Endocrine disorders	1 (1.3)	0	1 (0.7)
Neoplasms benign, malignant, and unspecified (including cysts and polyps)	0	1 (1.3)	1 (0.7)

All data values are presented as n (%) or n

AE adverse event, QD once daily, SAE serious adverse event TEAE treatment-emergent adverse event

Discussion

The efficacy of alogliptin in adults has been assessed through a comprehensive clinical development program involving over 50 clinical studies with > 20,000 adult participants. Adults treated with alogliptin 12.5 or 25 mg QD for 26 weeks demonstrated significant reductions in HbA1c and FPG levels compared with those receiving placebo. Notably, the treatment effect was evident as early as the 4th week of treatment, with continued effects at weeks 26 and 52 [19–23].

However, in the present study, no significant difference in HbA1c change from baseline at weeks 12, 18, 26, 39, or 52 was observed among pediatric participants aged 10–17 years. Subgroup analyses based on prior antihyperglycemic treatment, age, sex, race, and BMI produced similar results (Fig. 3b).

Generally, clinical studies involving children/adolescents tend to be more challenging and have a higher rate of failure in establishing both efficacy and safety than studies involving adults [24]. Several common factors contribute to the challenges faced in pediatric trials, including dosing issues, placebo response, study design, and potential differences between the disease processes in pediatric and adult populations [24]. However, the lack of efficacy in this study did not appear to be influenced by dosing, drug exposure, placebo response, or the study design. The dosage of alogliptin used in this study was also used in a PK/PD study in pediatric patients aged 10–17 years that reported similar alogliptin exposure and DPP-4 inhibition levels between pediatric and adult patients with T2DM, and no notable placebo response was observed [17]. Moreover, the study design, which was double-blind and placebo-controlled, was the same as that used in successful adult phase 3 studies showing the positive efficacy of alogliptin [19–23]. After excluding the aforementioned potential factors, it became apparent that the failure of this study in the pediatric population may be attributed to rapid disease progression and the moderate potency of alogliptin as a DPP-4 inhibitor.

Although the pathophysiology of T2DM in pediatric patients was previously reported as comparable to that in adults [25], despite a lack of pediatric vs. adult studies, increasing evidence indicates that the decline in pancreatic beta-cell function may be faster in the pediatric population than in the adult population [9, 26–28]. For example, Weiss et al. observed impaired insulin release following glucose stimulation in youth with obesity with T2DM compared with peers with obesity without diabetes as an indicator of early stages of beta-cell function decline [28]. Moreover, several distinct features of T2DM have been observed in pediatric patients compared with adults.

Children with T2DM exhibit higher insulin resistance than adults, even when accounting for the same relative amount of adipose tissue [29–31]. The Restoring Insulin Secretion (RISE) studies were designed to directly compare the effects of medication on beta-cell function in pediatric and adult patients with T2DM, using identical hyperglycemic clamp protocols [29]. Youths were reported to have lower insulin sensitivity, hyperresponsive beta cells, reduced insulin clearance [30], and increased beta-cell function deterioration [31] compared with adults. Substantial evidence exists for a more aggressive disease process in youth-onset T2DM compared with adult-onset T2DM [9].

This rapid disease progression also suggests the need for more potent or high–glycemic-efficacy pharmacological therapies. DPP-4 inhibitors, while valuable additions to the treatment armamentarium for T2DM, are generally regarded as less potent than most GLP1-RAs and sodium-glucose cotransporter-2 (SGLT2) inhibitors in terms of their glucose-lowering effects [32]. GLP1-RAs and SGLT2 inhibitors also have a known benefit in significantly lowering body weight [33]. This becomes even more important because most patients in this trial were overweight.

Unlike evidence from trials in adults with T2DM where HbA1c-lowering benefits have been observed [34–39], the lack of treatment effect in pediatric patients with T2DM has also been observed in trials with three other DPP4 inhibitors: sitagliptin, linagliptin, and saxagliptin.

There have been three phase 3 studies involving sitagliptin in pediatric patients with T2DM. One study by Shanker et al. investigated the use of sitagliptin (100 mg QD) as an initial therapy for youth with T2DM in a 54-week, double-blind, randomized controlled clinical trial [40]. The study included 190 participants aged 10–17 years with HbA1c levels of 6.5–10% (7.0–10% if on insulin). All participants were overweight or obese at screening and tested negative for pancreatic autoantibody. The study used a placebo control for the first 20 weeks, after which metformin replaced the placebo. The primary efficacy endpoint was the change in HbA1c levels from baseline to week 20. However, the results showed that DPP-4 inhibition with sitagliptin did not significantly improve glycemic control [40].

Similarly, data were pooled from two other 54-week, double-blind, randomized, placebo-controlled studies involving sitagliptin (100 mg QD) or placebo, in addition to the existing treatment for young patients with T2DM aged 10–17 years who had inadequate glycemic control on metformin with or without insulin (ClinicalTrials.gov: NCT01472367 and NCT01760447). The 220 randomized and treated participants had HbA1c levels of 6.5–10.0% (7.0–10% if on insulin) at baseline and were overweight or obese at screening or diagnosis while testing negative for pancreatic autoantibodies. The primary endpoint was the change in HbA1c levels from baseline to week 20 [41]. Consistent with the findings of study by Shanker et al., in participants naïve to treatment, this pooled analysis also demonstrated that the addition of sitagliptin to metformin did not provide a lasting improvement in glycemic control in youth with T2DM [40].

A study comparing linagliptin, another DPP4 inhibitor, to empagliflozin, an SGLT-2 inhibitor, in youth aged 10–17 years with T2DM was recently published [42]. The DINAMO study compared the efficacy and safety of linagliptin (5 mg) with those of empagliflozin (10 mg) and placebo. This trial involved 158 pediatric patients with T2DM previously treated with metformin or insulin. The primary outcome was the change in HbA1c levels from baseline at 26 weeks. The study showed that the empagliflozin group had a significant reduction of 0.84% in HbA1c at week 26 compared with the placebo group (95% CI − 1.5, − 0.2; p = 0.01); however, linagliptin did not show the same HbA1c benefit [42]. Empagliflozin has subsequently been approved as an adjunct treatment to diet and exercise to improve glycemic control in adults and pediatric patients aged ≥ 10 years with T2DM, whereas linagliptin was another DPP-4 inhibitor that failed to demonstrate efficacy in pediatric trials [43].

Saxagliptin, another DPP-4 inhibitor, also failed to demonstrate efficacy in children and adolescents [44]. In this 26-week, randomized, phase 3 trial in patients aged 10–17 years, with uncontrolled T2DM (HbA1c levels 6.5–10.5%) treated with metformin, insulin, or both, patients received saxagliptin, dapagliflozin, or placebo. At week 26, the difference in the adjusted mean change in HbA1c levels versus placebo was significant for dapagliflozin (− 1.03 percentage points [95% CI − 1.57, − 0.49, p < 0.001]) and non-significant for saxagliptin (− 0.44 percentage points [95% CI − 0.93, 0.05, p = 0.078]) [44].

In contrast to the DPP4 inhibitors, injectable incretin-based treatments with the GLP1-RAs liraglutide, exenatide, and dulaglutide have been reported to improve glycemic control after 24, 26, and 52 weeks of treatment in youth with T2DM [45–47], with all three currently approved for use in this population [13–15]. It has been well established that GLP1-RAs are more efficacious than DPP-4 inhibitors in adults with T2DM [48]; therefore, it could be postulated that the stronger potency of GLP1-RAs was sufficient to overcome the severe insulin resistance in pediatric T2DM, whereas the potency of DPP-4 inhibitors was not.

Another issue worth mentioning is related to managing the influence of background therapy with metformin and/or insulin from confounding the outcomes of placebo-controlled studies. We introduced a prerandomization stabilization period for this trial to overcome this. This step was important to ensure patient eligibility and stabilize the background therapy before randomization. Despite the infrequent use of these run-in periods, we recommend their inclusion in additional clinical trials in the future to minimize the risk of confounding.

This study has several limitations that also need consideration. Although the study was statistically powered, the subgroups based on previous antihyperglycemic therapy were underpowered, making it challenging to draw meaningful comparisons between the different groups. Another important factor to consider is that the participants were recruited from six different countries, representing a global population. This variability in geographic location may have introduced unaccounted differences in access to healthcare and other resources relevant to diabetes management, which could have influenced the study results; conversely, it also provides a good representation from a global perspective. Furthermore, the frequency, consistency, and content of the diabetes education and home glucose-monitoring training sessions may have varied according to local guidelines, resources, and training available at each study site. Additionally, the exact duration or continuity of training sessions for each participant was not tracked, which may have impacted the depth of education and support they received over the course of the study. Lastly, it was not feasible to determine whether other medication use, specifically insulin use, increased or decreased during the trial or if there were any differences in insulin dosage between the alogliptin and placebo groups. Given the complexities of insulin dosing in this cohort, some of whom were treatment-naïve and some on basal plus/minus multiple daily dosing, and variable duration of insulin use and timing of hyperglycemic rescue, it was challenging to conduct a meaningful analysis on the change in insulin dosing throughout the trial. However, the study design included two features that we believe balanced background therapies between the two treatment groups. First, randomization was performed in a blinded manner; second, the study protocol required participants to maintain their background antihyperglycemic therapy (if applicable) at the same dose (e.g., baseline insulin dose) as at the time of randomization throughout the first 26 weeks of the double-blind treatment period. Thus, no dosage difference before and after the start of the trial would be expected for individual participants.

Overall, this study contributes to the body of scientific literature on the efficacy and safety of DPP-4 inhibitors in pediatric patients with T2DM, and the findings are consistent with those of other studies exploring the use of DPP-4 inhibitors (i.e., sitagliptin, linagliptin, and saxagliptin) in the pediatric population [40–42, 44].

Conclusion

The findings of this study indicate that alogliptin treatment (25 mg QD) did not show significant efficacy compared with placebo in pediatric participants aged 10–17 years with T2DM who had inadequate glycemic control, either with or without prior antihyperglycemic therapy. However, it is important to note that alogliptin treatment was generally safe and well tolerated, with no new safety concerns observed in this participant group. Overall, together with the pediatric studies on sitagliptin, linagliptin, and saxagliptin, this study provides valuable insights into the use of DPP-4 inhibitors in the pediatric population and underscores the need for continued research and exploration of various treatment approaches to better manage pediatric patients with T2DM, particularly with oral treatments.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Xuejun Victor Peng is responsible for the work described in this study. Xuejun Victor Peng conceived, designed, planned the study, and acquired and analyzed the data. Richard Czerniak was responsible for the population PK analysis and interpretation. All authors (Xuejun Victor Peng, Amy S. Shah, Georgeanna Klingensmith, Daniel S. Hsia, Yunlong Xie, and William V. Tamborlane) interpreted the results, critically reviewed and/or revised the manuscript for important intellectual content and provided final approval of the version to be published, and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

Funding for this study and the journal Rapid Service fee was provided by Takeda Pharmaceuticals Company Limited, Cambridge, MA, USA.

Data Availability

The datasets, including the redacted study protocol, redacted statistical analysis plan, and individual participant data supporting the results reported in this article, will be made available within 3 months from the initial request to researchers who provide a methodologically sound proposal. The data will be provided after its de-identification, in compliance with applicable privacy laws, data protection, and requirements for consent and anonymization.

Declarations

Conflict of Interest

Xuejun Victor Peng, Richard Czerniak, and Yunlong Xie are employees of Takeda Pharmaceuticals Company Limited, Cambridge, MA, USA. Georgeanna Klingensmith, Daniel S. Hsia, William V. Tamborlane, and Amy S. Shah have no conflicts of interests to declare. The new affiliation of Daniel S. Hsia is Emory University, Department of Pediatrics, Division of Endocrinology and Children's Healthcare of Atlanta, Atlanta, GA, USA. Georgeanna Klingensmith is now retired from Barbara Davis Center and Department of Pediatrics, University of Colorado, Aurora, CO, USA.

Ethical Approval

The study (ClinicalTrials.gov: NCT02856113; EudraCT: 2015-000208-25) was conducted in accordance with the Declaration of Helsinki, International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use, Guidelines for Good Clinical Practice, and all applicable laws and regulations. The study was approved by the appropriate regulatory agencies and institutional review boards (IRBs)/ethics committees (IRB Protocol ID: 2000022953). Written informed consent was provided by the parents or legal guardians, and the children/adolescents provided assent to participate in the study.