Outcomes in youth-onset type 2 diabetes during the transition period: glycemic control, microvascular complications, and effects of second-line agents

Article information

Ann Pediatr Endocrinol Metab. 2026;31(1):45-54

Publication date (electronic) : 2026 February 28

doi : https://doi.org/10.6065/apem.2550068.034

Jeesun Yoo¹, Yun Jeong Lee¹, Choong Ho Shin^1,2, Young Ah Lee^,1,2

¹Department of Pediatrics, Seoul National University Children's Hospital, Seoul, Korea

²Department of Pediatrics, Seoul National University College of Medicine, Seoul, Korea

Address for correspondence: Young Ah Lee Department of Pediatrics, Seoul National University Children’s Hospital, 101 Daehak-ro, Jongno-gu, Seoul 03080, Korea Email: nina337@snu.ac.kr

Received 2025 February 23; Revised 2025 April 2; Accepted 2025 April 25.

Abstract

Highlights

· Over one third of young adults with youth-onset type 2 diabetes mellitus (T2DM) developed microvascular complications by early adulthood, and the majority required insulin or second-line agents during follow-up.

· Glycemic control improved during the transition from late adolescence to young adulthood, with greater benefits observed in females, those diagnosed at a younger age, and those with suboptimal glycemic control at start of the transition.

· The addition of second-line antidiabetic agents was associated with improvements in glycated hemoglobin, supporting the need to broaden approved therapeutic options and insurance coverage for adolescents with youth-onset T2DM in Korea.

Introduction

Youth-onset type 2 diabetes mellitus (T2DM) has a greater burden of complications and comorbidities than adult-onset T2DM, with more rapid progression of complications [1]. Moreover, the annual decline in pancreatic beta-cell function in youth-onset T2DM has been reported to occur approximately twice as rapidly as in adult-onset T2DM [2,3], highlighting the need for earlier and more aggressive therapeutic intervention in early phase of T2DM.

The prevalence of youth-onset T2DM has exhibited a marked increase in recent years, possibly due to lifestyle changes and the increasing rate of obesity [4-6]. Data from the SEARCH for Diabetes in Youth Study revealed an annual percentage change in T2DM incidence of 5.31% from 2002 to 2018 [4]. In South Korea, the prevalence of T2DM in pediatric populations has also increased more than four-fold since the early 2000s [7]. Considering the high severity and rapid progression of youth-onset T2DM, the treatment paradigm has been reevaluated and updated through pediatric clinical trials of drugs approved for adults. The International Society for Pediatric and Adolescents Diabetes [8] and the American Diabetes Association [9] have recently approved the use of glucagon-like peptide-1 (GLP-1) receptor agonists (e.g., liraglutide, dulaglutide) or sodium-glucose cotransporter 2 (SGLT2) inhibitors (e.g., empagliflozin, dapagliflozin) as a secondary treatment option for pediatric patients with T2DM aged 10 years and older, based on pediatric randomized controlled trials (RCTs) demonstrating the efficacy and safety of GLP-1 receptor agonists including liraglutide, dulaglutide, and exenatide [10-12], and SGLT2 inhibitors including dapagliflozin and empagliflozin [13,14]. Despite this change in the global treatment paradigm, metformin and insulin are still the only approved drugs in Korean adolescents with T2DM to date.

In this study, we investigated the prevalence of diabetic nephropathy and retinopathy during the transition period between age of 19 and 22 years in patients with youth-onset T2DM. Next, we evaluated trends in glycemic control during this transition period. Lastly, we analyzed the impact of adjunctive second-line antidiabetic agents on glycemic outcomes.

Materials and methods

1. Patient identification and data collection

Among 319 patients who were diagnosed with T2DM at <18 years of age and followed for >1 year at Seoul National University Children’s Hospital between January 2001 and July 2023, 282 patients were retrospectively reviewed. Among 174 patients who reached the age of 20, 84 patients who had available glycated hemoglobin (HbA1c) data from at least 2 distinct ages between 19 to 22 years old were finally enrolled (Fig. 1).

Fig. 1.

Flowchart showing inclusion of the subjects. SNUCH, Seoul National University Children’s Hospital; HbA1c, glycated hemoglobin.

Age, height, weight, body mass index (BMI) and HbA1c data at each time point (at diagnosis, at the last visit during the transition period between the ages 19 to 22, and at the last follow-up visit) were collected. Duration of diabetes, history of antidiabetic medication, and the presence of microvascular complications including nephropathy or eye disease were evaluated. Obesity was defined as a BMI z-score ≥95th percentile for age and sex (<18 years of age) and a BMI absolute value ≥25 kg/m² (≥18 years of age). Medication use was defined as cases prescribed within 6 months at each time point, and documented records of continuous use for at least 3 months. Treatment was classified into metformin monotherapy, insulin monotherapy, metformin and insulin therapy, and addition of second-line agents (sulfonylurea, thiazolidinedione, GLP-1 receptor agonist, SGLT2 inhibitors, etc.).

Microalbuminuria (stage 3) was defined as a urine albumin-to-creatinine ratio (ACR) ≥30 mg/g in at least 2 out of 3 determinations within a 6-month period, or 3 or more consecutive ACR values ≥30 mg/g irrespective of the time interval between measurements. Macroalbuminuria (stage 4) was similarly defined, using an ACR threshold of ≥300 mg/g. Cases in which proteinuria or albuminuria was presumed to be caused by other underlying renal disorders (such as IgA nephropathy, acute tubulointerstitial nephritis, etc.) were excluded from the prevalence count of nephropathy. Nonproliferative diabetic retinopathy, proliferative diabetic retinopathy, macular edema, vitreous hemorrhage, and cataract were included as diabetic eye diseases. The determination of diabetic eye diseases was based on medical record documentation by an ophthalmologist.

Results

1. Patient characteristics and treatment regimen

Table 1 described the clinical characteristics of 84 patients (47 males and 37 females) at each time point. At diagnosis, mean age was 14.0±2.1 years. The median diabetes duration at the last follow-up visit was 9.6 (IQR, 6.5–14.6) years. Mean HbA1c level was 10.5% at initial diagnosis, 7.9% at the last visit during the transition period (mean, 21.5 years), and 8.1% at last follow-up (mean, 24.9 years).

Table 1.

Clinical characteristics and treatment regimens of subjects (N=84)

Treatment regimen at each time point is presented in Table 1. At diagnosis, 3 patients (4%) managed diabetes by lifestyle modification, and 32 patients (38%) were treated with metformin monotherapy, 15 (18%) with insulin monotherapy, and 27 (32%) with metformin and insulin combination therapy. The proportion of patients treated with metformin monotherapy decreased to 15% at the last visit during the transition period, and to 12% at the last follow-up visit. Moreover, 46 patients (57%) were dependent on insulin therapy at the last follow-up. The proportion of patient utilizing second-line agents increased markedly from 8% at diagnosis to 59% at the last follow-up (Table 1).

3. Change in glycemic control during the transition period

Fig. 2A shows HbA1c trajectory during the transition period between age 19 and 22. In overall study subjects, mean HbA1c levels showed a decreasing trend with age from 8.2% at age 19 to 7.7% at age 22 (P=0.007). Next, change in HbA1c category (<7%, ≥7%–<9%, and ≥9%) was evaluated (Fig. 2B). A decreased proportion in the ≥9% category (from 37% at age 19 to 25.6% at age 22) and increased proportion in the <7% category (from 37% at age 19 to 51.2% age 22) was found, although these changes were not significant.

Fig. 2.

Glycemic control during the transition period from age 19 to 22. (A) Box-whisker plot showing mean glycated hemoglobin (HbA1c) trajectory during the transition period. The box represents the interquartile range containing the middle 50% of the data, with the median indicated by the horizontal line within the box. The lower and upper whiskers extend to the minimum and maximum values. Blue dots represent the mean values of HbA1c of each age group. (B) 100% stacked bar plot of HbA1c categories during the transition period.

Fig. 3 exhibits the results of LMM analysis. Females (P=0.049 vs. males), patients diagnosed before age 15 (P=0.001 vs. after age 15), and those with HbA1c ≥7% at age 19 (P<0.001 vs. <7%) had significantly higher HbA1c levels during the transition period. Significant interactions were observed for age with sex (P=0.04), age at diagnosis category (P=0.09), and HbA1c category at age 19 (P<0.001). Females (P=0.002), patients diagnosed before age 15 (P=0.002), and those with HbA1c ≥7% at age 19 (P<0.001) showed a significant decrease in HbA1c levels during the transition period, while their respective counterparts—males, those diagnosed at or after age 15, and individuals with HbA1c <7% at age 19—did not exhibit significant decreases.

Fig. 3.

Change in glycemic control during the transition period and associated factors. (A) Sex. (B) Age at diagnosis. (C) Glycated hemoglobin (HbA1c) category at age 19. F, female; M, male.

4. Glycemic control after adding second-line antidiabetic medication

Fig. 4 shows mean HbA1c levels (Fig. 4A) and HbA1c category (Fig. 4B) at each time points. Mean HbA1c levels significantly decreased from 9.3% at baseline to 8.5% at 1 year after addition of second-line medication (P=0.03 vs. baseline), and to 8.1% at the last follow-up (P=0.03 vs. baseline, Fig. 4A). The proportion of patients in the <7% category significantly increased from 13.6% at baseline to 28.8% at 1 year after the addition of second-line drugs (P=0.04 vs. baseline), and to 35.6% at the last follow-up (P=0.006 vs. baseline) (Fig. 4B). The proportion of patients in the ≥9% category decreased from 59.3% at baseline to 45.8% at 1 year after addition of second-line medication without significance (P=0.12 vs. baseline), but significantly decreased to 33.9% at the last follow-up (P=0.005 vs. baseline).

Fig. 4.

Glycemic control after adding second-line antidiabetic agents. (A) Box-whisker plot showing changes in glycated hemoglobin (HbA1c) following the initiation of second-line agents. Each individual patient is represented by a dot at each time point, and changes in their values are shown as a gray line connecting the dots. The asterisk (*) indicates a P-value of less than 0.05. (B) Sankey diagram showing changes in HbA1c categories following the initiation of second-line agents.

Fig. 5A presents a pie chart of the proportion of patients who initiated any second-line agent during the follow-up period, as well as the distribution of initially prescribed second-line agents (by class). Of the 84 patients included in the study, second-line antidiabetic medications were prescribed to 59 (70.2%). The most frequently prescribed second-line agent was sulfonylureas (29%), followed by dipeptidyl peptidase-4 (DPP-4) inhibitors (26%), and GLP-1 receptor agonists and SGLT2 inhibitors at 5% each. Fig. 5B and C demonstrate the distribution of initially prescribed second-line agents before and after May 2016, respectively. Prior to this time point, sulfonylureas predominated, accounting for 75% of second-line agent selections, with DPP-4 inhibitors comprising the majority of the remainder. Post-May 2016, greater diversity in agent selection was observed. DPP-4 inhibitors appears to be the most frequently selected second-line agent, followed by SGLT2 inhibitors, GLP-1 receptor agonists, and sulfonylureas, each accounting for approximately 10% or more of all initial prescriptions.

Fig. 5.

Treatment regimen and the class of second-line antidiabetic agents. (A) Distribution of overall medication usage and first prescribed second-line agent during the total follow-up period. *‘Other’ category included drugs from the glucagon-like peptide-1 (GLP-1) receptor agonist and thiazolidinedione classes. (B) First prescribed second-line agent before May 2016 (N=28). (C) First prescribed second-line agents from May 2016 (N=31). *‘Other’ category includes drugs from the GLP-1 receptor agonist and thiazolidinedione classes.

Discussion

In patients with youth-onset T2DM, diabetic microvascular complications were identified in 20% during the transition period, and this increased to more than 33% at the last follow-up visit. Mean HbA1c levels significantly improved during the transition period. In particular, females, patients diagnosed before age 15, and those with HbA1c levels of 7% or higher at age 19 exhibited significant downward trajectories in HbA1c during the transition period. Addition of second-line agents has been shown to improve glycemic control. Second-line options have also expanded according to changes in domestic insurance coverage.

In our study of subjects with a median diabetes duration of 9.6 years, one-third of patients developed one or more microvascular complications, including nephropathy in 25% and diabetic eye diseases in 20%. This high occurrence of complications within a few years of T2DM diagnosis is similar to another report of the high incidence of complications within a similar period in the Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) study [15]. A higher odds ratio for predicting microalbuminuria, retinopathy, and neuropathy has been reported in patients with youth-onset T2DM compared to those with T1D [16-19]. Furthermore, a higher incidence of diabetic complications and more rapid disease progression have been reported in youth-onset T2DM compared to adult-onset T2DM [20]. Research from the TODAY study showed that the prevalence of elevated urine albumin excretion increased from 6.3% to 16.6% over 3.9 years of follow-up and to 18% at 5-year follow-up in youth with T2DM [21,22]. Rates of diabetic kidney disease progression are more rapid in youth-onset T2DM than adult-onset T2DM, likely driven by early metabolic programming and pubertal physiological changes that exacerbate renal energy imbalance and hypoxia [23]. Considering the higher incidence of diabetic complications in youth-onset T2DM and their aggressive clinical course [24], early intensive intervention is required.

Glycemic control in patients with adolescent-onset T2DM tended to improve between 19 and 22 years of age. There has been a paucity of research investigating the trajectory of glycemic control in patients with youth-onset T2DM, in their transition to adulthood. Improvement of insulin resistance at the end of puberty may be contributing factors to glycemic improvement [25]. The nadir of insulin sensitivity occurs around Tanner stage III, with a subsequent recovery to prepubertal levels by Tanner stage V [26,27]. In addition, the increase in self-efficacy, relief of emotional distress, and independence in adulthood could positively affect a patient’s ability to manage diabetes [28]. Meanwhile, according to the SEARCH for Diabetes in Youth Study [29], young adults with youth-onset T2DM often exhibit worsening glycemic control and loss to follow-up during the transfer from pediatric to adult care, highlighting need for a tailored clinical program to support them. Physiological and behavioral shifts may contribute to the glycemic improvement in transition-aged patients with T2DM. Further prospective studies are needed to identify predictors affecting glycemic control during the transition period and incorporate them in clinical programs.

In particular, a significant decrease in HbA1c level was found in females, patients diagnosed before age 15, and those with poor glycemic control (≥7% of HbA1c) at age 19. These 3 groups all exhibited worse glycemic control during the transition period compared to their counterparts (males, patients diagnosed after age 15, and those with good glycemic control at age 19). The association of female sex and longer disease duration with poor glycemic control has been reported in pediatric and adult patients with T2DM [30,31]. In addition, those with poor glycemic control at the age of 19 may be prescribed a second-line antidiabetic medication. In this study, the addition of second-line agents during the transition period improved glycemic control at the 1-year follow-up and at the last follow-up. Policy improvements for adolescents with T2DM are needed to allow the initiation of prescriptions for second-line antidiabetic medications during adolescence.

In our study subjects, sulfonylureas and DPP-4 inhibitors were the most frequently prescribed second-line medications. With the Korean reimbursement approval of 2 agents for adult patients in May 2016—empagliflozin (Jardiance), an SGLT2 inhibitor, and dulaglutide (Trulicity), a GLP-1 receptor agonist—there was a decrease in the frequency of sulfonylurea prescriptions, and an increase in the utilization of DPP-4 inhibitors. Also, the use of SGLT2 inhibitors, and GLP-1 receptor agonists has increased since May 2016. This reflects how reimbursement policies for specific medications influence clinical practice in South Korea, potentially impacting glycemic outcomes.

Sulfonylurea stimulates insulin secretion by binding to the SUR subunit of the ATP-sensitive potassium channel and inducing channel closure in pancreatic beta cells [32]. While data on the use of sulfonylureas in youth-onset T2DM are limited, a randomized, single-blind comparative study of 285 pediatric patients demonstrated significant HbA1c reductions in both the glimepiride and metformin groups [33]. Linagliptin, a DPP-4 inhibitor that enhances the effects of incretin by increasing active GLP-1 levels [32], is well-tolerated and effective in pediatric patients with T2DM [34], but systematic review of 5 RCTs indicated that the efficacy of DPP-4 inhibitors in youth-onset T2DM remains unclear [35]. Recent advances in pharmacotherapy research have expanded treatment options for youth-onset T2DM, shifting focus towards GLP-1 receptor agonists and SGLT2 inhibitors, rather than sulfonylureas or DPP-4 inhibitors. GLP-1 receptor agonists not only stimulate glucose-dependent insulin secretion and suppress glucagon release, but also delay gastric emptying, and SGLT2 inhibitors reduce hyperglycemia through increased urinary glucose excretion by inhibiting renal glucose reabsorption [32]. These options are approved for pediatric use by the Food and Drug Administration (FDA) and the European Medicines Agency (EMA), and have demonstrated superior glycemic control and safety profiles in RCTs [10-14]. According to a pilot study on the efficacy of dulaglutide, glycemic control was significantly improved during 3 months and 1-year posttreatment in 5 Korean pediatric patients with T2DM, with minimal side effects and good adherence [36]. Based on international guidelines for youth-onset T2DM [8,9], insurance coverage needs to be updated so that FDA- or EMA-approved drugs (GLP-1 receptor agonists and SGLT2 inhibitors) can be prescribed to Korean adolescents.

This study has several limitations. First, it is limited by its retrospective design in a single tertiary center and the lack of information on medication adherence. Second, there is a selection bias, as patients with poor treatment compliance or loss to follow-up due to adverse events were excluded, since our study included subjects who had at least 2 HbA1c measurements across different age groups between 19 and 22 years. Third, risk factors affecting glycemic trends during the transition period could not be determined in detail by a retrospective design. Fourth, glycemic trajectories between patients who received second-line agents and those who did not could not be compared due to variation in drug type and the timing of initiation.

In conclusion, one-third youth-onset T2DM patients in this study experienced microvascular complications in young adulthood. Transition-aged patients showed improved glycemic control, with significant downward HbA1c trajectories in females, patients diagnosed at a younger age, and those with poor glycemic control at age 19. Considering the beneficial effects of second-line antidiabetic agents on glycemic control during the transition period, early introduction of second-line agents in adolescence is expected to improve glycemic control and prevent the development or progression of microvascular complications.

Supplementary materials

Article information Continued