Youth -Onset Type 2 Diabetes: Lessons Learned from the TODAY Study

Abstract

Type 2 diabetes is increasingly diagnosed in obese children and adolescents. Evidence suggests that this disease commonly progresses more rapidly in youth compared to adults, and is associated with high rates of early microalbuminuria, hypertension and dyslipidemia. The Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY) study was the first multiethnic, multicenter randomized trial in the United States to compare three treatment approaches in new onset type 2 diabetes in obese youth (n=699; ages 10–17): monotherapy with metformin, metformin with roziglitazone, and metformin with an intensive lifestyle intervention. The primary outcome was glycemic control. Diabetes-related complications and cardiovascular risk factors were also examined. Approximately half of the participants were unable to maintain glycemic control using metformin alone. Combination therapy of metformin with rosiglitazone resulted in better durability of glycemic control, and the metformin plus intensive lifestyle was intermediate but not superior to metformin alone. Deterioration in glycemic control was associated with rapid loss of beta cell function, not worsened insulin sensitivity, and could not be explained by differences in adherence or BMI. After 3.9 years, 33.8% of participants had hypertension and 16.6% had microalbuminuria. Only 55.9% of participants had an LDL-cholesterol <100 mg/dl after 3 years and 13.7% had retinopathy. The significance of the findings from this important trial for the management of youth and young adults with youth-onset type 2 diabetes and its complications is discussed. An aggressive multifaceted approach is needed to prevent or forestall premature microvascular and macrovascular complications in youth-onset type 2 diabetes.

Introduction

The incidence of type 2 diabetes in children and adolescents is increasing as the rates of childhood obesity have increased. The natural history of youth-onset type 2 diabetes and its complications, and ways in which it differs from that of adult onset diabetes, will be described in this review. Early development of hypertension, microalbuminuria and dyslipidemia has important implications for monitoring and treatment. The Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY) study is the first multiethnic, multicenter randomized trial of youth-onset type 2 diabetes to compare three therapeutic approaches to maintain glycemic control, and to examine diabetes-related complications and cardiovascular risk factors over time. This review summarizes the problem of type 2 diabetes in youth and results from the TODAY study. The significance of the findings from this important trial for the management of type 2 diabetes and its complications in the pediatric population as well as the implications for providers for the care of these individuals as they enter adulthood are discussed (Table 1).

Table 1.

Characteristics of Type 2 Diabetes in Youth and Adults

Characteristic	Youth in TODAY	Adults
Obesity	Common Poor response to intensive lifestyle change program	Common Lifestyle change programs more effective in older adults
Pancreatic beta cell function	20–35% decline per year	7–11% decline per year
Initial therapeutic response to metformin monotherapy	Poor durability of glycemic control in approximately half of youth	Durability of glycemic control better in older adults
Microalbuminuria	6.3% at baseline 16.6% by 4 years Long-term data unavailable	38% of UKPDS participants after 15 yrs34; Diabetic nephropathy occurs in 20- 40% of adults with type 2 diabetes and is the leading cause of kidney failure in the US53
Hypertension	11.6% at baseline 33.8 % over 3.9 years Long-term data unavailable	Common (67% of adults with diabetes53)
Dyslipidemia	Triglyceride ≥150 mg/dL or LLM: 23.3% at 3 yrs LDL-chol <100 mg/dL: 55.9% at 3 yrs; LDL-chol ≥130 mg/dL or LLM: 10.7% at 3 yrs Long-term data unavailable	Triglyceride ≥150 mg/dL: >50% of adults54 LDL-chol <100 mg/dL: less than 1/3 of men and 20% of women54
Retinopathy	13.7% at 2.8 yrs diabetes duration	12.6% in adults 3 yrs after diagnosis of diabetes
Depressive symptoms	15%	17.6%55

chol=cholesterol; LDL=low density lipoprotein; LLM= lipid lowering medication; UKPDS=United Kingdom Prospective Diabetes Study; US=United States

Scope of the Problem

Obesity is the most important risk factor for the development of type 2 diabetes in adolescents. Body mass index (BMI) between the 85^th and 95^th percentile classifies children as overweight and BMI at or above the 95^th percentile for age classifies children as obese(). The prevalence of BMI above the 95^th percentile among children has significantly increased from 1999 to 2004: from 14% to 16% in female adolescents and from 14% to 18% in males(Ogden et al., 2006). There is a disproportionate increase in prevalence of overweight and obesity among children from minority groups, approaching 23% in non- Hispanic Black and Mexican American adolescents compared to 13% in non-Hispanic White adolescents(Ogden, et al., 2006). Unfortunately, overweight and obesity in childhood predispose to obesity in adulthood(Steinberger, Moran, Hong, Jacobs, & Sinaiko, 2001). There is a strong relationship between early onset obesity and increased risk of diabetes in adults(Chan, Rimm, Colditz, Stampfer, & Willett, 1994).

Type 2 diabetes, previously diagnosed only in adults, is now increasingly diagnosed in children(Fagot-Campagna, 2000; Hale & Rupert, 2006; Pihoker, Scott, Lensing, Cradock, & Smith, 1998; Pinhas-Hamiel et al., 1996; Scott, Smith, Cradock, & Pihoker, 1997). Recent United States (US) data indicate the rate of new cases of type 2 diabetes among those < 20 years of age is 8.5 per 100,000(Dabelea et al., 2007): approximately 20,000 children in the US are affected by type 2 diabetes(Pettitt et al., 2013). For Asian/Pacific Islander and American Indian youth aged 10–19 years, the rate of new cases of type 2 diabetes is now higher than for type 1 diabetes(Dabelea, et al., 2007; Prevention, 2011)². Risk factors in children include being overweight, having a family history of diabetes, race/ethnicity (African American, Hispanic, Native American, Asian/Pacific Islander), signs or conditions associated with insulin resistance (including acanthosis nigricans, hypertension, dyslipidemia, polycystic ovarian disease, having been small for gestational age), and maternal history of diabetes or gestational diabetes during the child’s gestation^13,14.

The TODAY Study: Rationale and Design

As type 2 diabetes emerged in the pediatric population, it was recognized that an understanding of biological differences between youth and adults with type 2 diabetes and their development of diabetes complications is essential for optimal treatment. There was a lack of knowledge concerning the safety and efficacy of oral glycemic control medications used in adults for the management of type 2 diabetes in youth. Only metformin and insulin were approved for use in the pediatric population. Hormonal changes in adolescence coupled with varying physical and emotional factors including family interactions, peer influences and poor socioeconomic conditions typically seen in youth onset type 2 diabetes and obesity¹⁶ pose unique challenges, especially when promoting lifestyle change. The hypothesis tested in the multisite TODAY trial was that the use of combination treatment in new-onset type 2 diabetes in youth would be more effective in maintaining glycemic control than metformin monotherapy.

The TODAY trial evaluated the effectiveness of three treatment approaches on glycemic control and studied diabetes-related complications, cardiovascular risk factors and psychosocial factors in new-onset obese type 2 diabetes in youth¹⁵ (Figure 1). Eligibility criteria included age of 10–17 years, recent onset of type 2 diabetes (< 2 years), BMI ≥85^th percentile for age and sex, availability of a ‘family support person’- an adult caregiver willing to participate and support the youth participant, negative antibodies to glutamic acid decarboxylase (GAD-65) and insulinoma-associated protein 2 (IA-2), and fasting c-peptide > 0.6 ng/mL. After a run-in period, in which patients were treated with metformin alone, received diabetes education, demonstrated adherence to medications and attendance to visits and maintained HbA1C < 8%, participants were randomized to receive metformin alone (1000 mg twice daily; 500 mg twice daily if higher doses were not tolerated), metformin with rosiglitazone (4 mg twice daily), or metformin with an intensive lifestyle intervention program¹⁵. The intensive lifestyle program promoted weight loss through a family-based behavior-change program that encouraged healthy eating and activity behaviors and moderate weight loss (a reduction of 7–10% of percent overweight was considered meaningful). This program was delivered with in-person visits by trained Personal Activity/nutrition Leaders. The rationale, design, methods and the details of the lifestyle intervention program have been reported^{15, 17}. A licensed psychologist at each site supervised the intensive lifestyle-intervention program. Enrollment was from July 2004 to February 2009, and follow-up continued through February 2011, ensuring a minimum follow up of 2 years and a maximum of 6.5 years.

Figure 1. Design of the TODAY Study.

ACE-I=angiotensin converting enzyme inhibitor; BMI=body mass index; GAD=glutamic acid decarboxylase; HbA1C: hemoglobin A1C; IA-2= insulinoma-associated protein 2

Adapted from Pediatr Diabetes¹⁵, with permission. Data also from N Engl J Med¹⁸

TODAY Study Participant Characteristics and Glycemic Outcomes

Of the 1211 children and adolescents who were screened for entry into the TODAY study, 927 (76.5%) entered the run-in phase and 699 (75.4%) were randomized (Figure 1)¹⁸. The mean duration of type 2 diabetes at the time of randomization was 7.8 months and participants were followed for an average of 3.9 years. Baseline demographic, clinical and biochemical characteristics of the cohort have been previously reported and are summarized below^{15, 18}.

The cohort was 64.7% female, with a mean age of 14.0 years. The mean BMI z-score was 2.23, indicating this cohort was significantly more obese than the threshold BMI 85^th percentile for age and sex. 39.7% were Hispanic, 32.5% non-Hispanic Black, 20.3% non-Hispanic White, 5.9% American Indian and 1.6% Asian (Figure 1). A family history of diabetes was reported for 89.4% and almost 60% had a first degree relative with type 2 diabetes. A third of participants were born after a pregnancy complicated by diabetes: 17.2% of those born full term were large for gestational age (>4000g) and 9% were small for gestational age (<2500g). More than 60% participants lived in homes without both biological parents and 41.5% had a reported annual household income <$25,000¹⁸. Acanthosis nigricans was present in 85.6% of participants, and 88% were in Tanner stage 4 or 5 of puberty. Significant depressive symptoms were reported in 15% of participants, and clinical and subclinical binge eating in 6% and 20% respectively^{19, 20}.

The primary outcome of the study was time to treatment failure defined as persistently elevated HbA1C ≥ 8% over 6 months or inability to wean from insulin within three months of its initiation for acute metabolic decompensation or the need to begin insulin therapy for the second time within a three month period for metabolic decompensation. Of the 699 randomized participants, 46% reached primary outcome, with a median time to treatment failure of 11.5 months (Figure 2)¹⁸. Although rosiglitazone is not currently recommended to treat youth with type 2 diabetes (Table 2), metformin plus rosiglitazone was associated with a significantly lower occurrence of primary outcome (failure rate 39%) compared to metformin alone (failure rate 52%). Metformin therefore was less effective in providing durable glycemic control in youth with type 2 diabetes than has been shown in adult studies (Table 1). The outcome with metformin plus intensive lifestyle intervention was intermediate (failure rate 47%) and not significantly different from metformin alone (Figure 2). Primary outcome results did not change after adjusting for sex, race, ethnic group, baseline HbA1C and BMI at baseline. Medication adherence was 84% and 57% at months 8 and 60 respectively, and did not differ significantly between the 3 treatment groups. Attendance in the Intensive Lifestyle Program was 75% over the first 24 months with 54% attending ≥75% of sessions.

Figure 2. The TODAY Study: Overall Primary Outcome Results.

Survival curves for freedom from glycemic failure are shown. Data are for up to 60 months of follow-up (accounting for 98.4% of cases of glycemic failure), although the rates and analysis are based on the complete data set.

From N Engl J Med¹⁸, with permission.

Table 2.

Treatment Guidelines for Type 2 Diabetes

	Youth	Adults
Lifestyle change
Diet	6–12 years of age: 900–1200 kcal/day 13–18 years of age: ≥1200 kcal/day Increase fruits and vegetables Limit sugary drinks Limit high fat foods, snacks and fast food Consider short-term (10 wks) supervised protein-sparing modified fast if body weight > 120% ideal body weight and there is need for rapid weight loss for a serious medical condition	Individualize Saturated fat <7%; low trans fat; For weight loss: low carbohydrate, low-fat low-calorie or Mediterranean diet Reduce protein if renal disease present (0.8–1.0 gm/kg/day)
Physical     activity	Moderate-rigorous ≥ 60 min/day Nonacademic screen time <2 hrs/day	Moderate intensity aerobic ≥ 150 min/wk over ≥3 days/wk Resistance training ≥ 2days/wk
Smoking     cessation	Recommended	Recommended
Education (including     preconception     counseling)	Recommended	Recommended
Glycemic Control Medications
Metformin	Initial therapy. Inexpensive, effective, weight neutral; can cause diarrhea, abdominal discomfort, anorexia, nausea; effective in half of youth in the TODAY trial	Initial therapy. Inexpensive, effective, weight neutral; can cause diarrhea, abdominal discomfort, anorexia, nausea; rare lactic acidosis
Sulfonylureas     (glyburide, glipizide,     glimepiride)	Not approved	Inexpensive, can cause hypoglycemia and weight gain
Meglitinides     (repaglinide,     nateglinide)	Not approved	Medium cost, short-acting, can cause hypoglycemia and weight gain
DPP-4 inhibitors     (sitagliptin, saxagliptin,     linagliptin, alogliptin)	Not approved	Expensive, weight neutral, risk of pancreatitis
GLP-1 receptor agonists     (exenatide, liraglutide)	Not approved	Injectable, expensive, associated with weight loss, gastrointestinal side effects, risk of pancreatitis
Thiazolidinediones     (pioglitazone,     rosiglitazone)	Not approved	Rarely used because of adverse effects: weight gain, heart failure, edema, fracture, bladder cancer risk (pioglitazone) and possible increased CVD risk (rosiglitazone)
Alpha glucosidase     inhibitors (acarbose,     miglitol)	Not approved	Medium cost, moderate efficacy, gastrointestinal side effects
SGLT2 inhibitors     (canagliflozin)	Not approved	Expensive, no hypoglycemia; weight loss possible. Risks: genital mycotic and urinary tract infections, hypotension, impaired renal function, hyperkalemia, increased LDL-chol.
Insulin	Initial therapy with marked hyperglycemia; main adverse effect is hypoglycemia	Initial therapy with marked hyperglycemia; main adverse effect is hypoglycemia
Dilated eye examinations	Annual; can consider every 2–3 yrs if previous normal eye exams	Annual; can consider every 2–3 yrs if previous normal eye exams
Nephropathy screening and treatment	Annual urine microalbumin:creatinine ratio; consider ACE-I or angiotensin receptor blocker (ARB) if persistent elevation (>30 mg/G Cr) on 2 different days. Annual serum Cr and GFR calculation	Annual urine microalbumin:Cr and consider ACE-I or ARB if persistent elevation (>30 mg/G Cr) on 2 different days Annual serum Cr and GFR estimation
CVD Risk Reduction (in addition to lifestyle change)
Blood pressure
Goal	< 90th percentile or <130/80 mm Hg	<140/80 mm Hg; <130/80 mm Hg in younger adults
Treatment	Initial Therapy: Low salt diet, moderation of alcohol, ACE-I or ARB Multiple drugs commonly needed	Initial Therapy: Low salt diet, moderation of alcohol, ACE-I or ARB Multiple drugs commonly needed
LDL-chol
Goal	<100 mg/dL	<100 mg/dL without overt CVD, <70 mg/dL with CVD
LDL -chol treatment	Step 2 American Heart Association diet (saturated fat 7% and total cholesterol 200 mg/day) Statin drug- add if > 10 years of age and LDL-chol > 160 mg/dL after lifestyle changes or LDL-chol > 130 mg/dL and one or more risk factor*	Lifestyle change (diet, physical activity as noted above) Initial therapy with a statin drug. Statin recommended without CVD if >40 yrs of age with one additional risk factora
Aspirin	Not recommended	Use with known CVD and with high CVD risk (most men >50 yrs or women >60 yrs with ≥1 additional risk factora)

^aFamily history of cardiovascular disease (CVD), hypertension, smoking, dyslipidemia, albuminuria

ACE-I=angiotensin converting enzyme inhibitor; ARB=angiotension receptor blocker; chol=cholesterol; Cr=creatinine; CVD=cardiovascular disease; DPP-4= dipeptidyl peptidase; GLP-1= glucagon-like peptide 1; GFR=glomerular filtration rate; LDL=low density lipoprotein; SGLT2=sodium-glucose cotransporter 2; TODAY: Treatment Options for type 2 Diabetes in Adolescents and Youth

Data from Diabetes Care¹³ and Pediatrics⁴⁵

The group that received the intensive lifestyle program showed the smallest increase in BMI and percent overweight in the first 6 months of the study as well as long term (up to 60 months), whereas the metformin plus rosiglitazone group had the greatest increase in BMI in the first 6 months and long term. However, the effect of the intensive lifestyle program on BMI was small and did not result in greater durability of glycemic control. The change in percent overweight at 24 months was not significantly different in those treated with metformin alone compared to metformin plus intensive lifestyle. On the other hand, metformin plus rosiglitazone treatment resulted in a modest increase in adiposity, but better durability of glycemic control²¹. All treatments were well tolerated. Gastrointestinal side effects were the most common adverse events, and were lowest in the metformin plus rosiglitazone group.

Gender and racial differences were observed. Metformin with rosiglitazone was more effective in girls than in boys, and was more effective than metformin alone in girls (but not in boys). The highest failure rate was observed in non-Hispanic Blacks (53%) followed by Hispanics (45%) and non-Hispanic Whites (37%). Monotherapy with metformin was less effective in non-Hispanic Blacks than in Hispanics or non-Hispanic Whites.

Insulin sensitivity and beta cell function were studied using oral glucose tolerance tests that were performed at baseline, 6 months, 24 months and yearly thereafter²². At baseline, there was an inverse correlation between HbA1C and beta cell function relative to insulin sensitivity. Over the course of 4 years, the TODAY trial demonstrated that the combination of metformin plus rosiglitazone resulted in greater improvement in insulin sensitivity and beta cell function relative to insulin sensitivity compared to the metformin alone group and the metformin plus intensive lifestyle arm²². However, in contrast to adults, youth with type 2 diabetes treated with metformin plus rosiglitazone showed continued decline in insulin sensitivity after short term improvement in the first six months. Beta cell function deteriorated at a rate of 20–35% per year, in contrast to adult studies that showed a slower (7–11%) decline in beta cell function per year^23–25. Youth with type 2 diabetes who had higher HbA1C, higher fasting glucose levels and lower beta cell function were less likely to be able to maintain glycemic control, as the insulin secretion indices declined with increasing HbA1C. Youth with type 2 diabetes who had more advanced disease at baseline (higher HbA1C, severe impairment of beta-cell function) experienced progressive and more rapid loss of beta cell function and were unable to maintain satisfactory glycemic control²².

After the trial was in progress there were reports of increased fractures in adults taking rosiglitazone and pioglitazone, resulting in the American Diabetes Association recommending avoidance of these drugs in people with type 2 diabetes with risk factors for fracture¹³. Because of this, bone mineral density was measured. Accrual of bone mineral content (BMC) and bone mineral density (BMD) was lower in the metformin plus rosiglitazone group, but the differences between groups were small and not considered clinically significant. It is not yet known if the thiazolidinedione effects on bone are reversible in adolescents or if this treatment affects attainment of peak bone mass²¹. Additional factors could also affect bone health in this population, including vitamin D deficiency and the use of progestational agents (eg. Depot-MPA) for contraception, but these were not analyzed in the TODAY trial.

Hypertension, Microalbuminuria, and Retinopathy in Youth-Onset Type 2 Diabetes

Microalbuminuria and hypertension have been reported in cross-sectional studies on youth with type 2 diabetes, with a higher prevalence than observed in youth with type 1 diabetes^26–29. The high incidence of microalbuminuria is of great concern, since youth-onset type 2 diabetes is associated with more end stage renal disease and greater mortality in middle-aged adults³⁰. Whether retinopathy is more common in adolescents with type 2 than type 1 diabetes is unclear^{28, 29,31}.

The TODAY cohort provided the opportunity to study the incidence and progression of hypertension, nephropathy and retinopathy and their association with glycemic control and other risk factors over a 4 year period³². Blood pressures were measured at every visit and urine samples collected at baseline and annually, unless there was an abnormal result necessitating more frequent collection. Treatment of hypertension and microalbuminuria was initiated per algorithms beginning with angiotensin converting enzyme inhibitor (ACE-I) therapy. Retinopathy was studied in the last year of the TODAY trial using fundus photography (n=524)³³.

In the TODAY study, hypertension was seen in 11.6% of youth at baseline and increased to 33.8% over an average study duration of 3.9 years³². Males had 81% greater risk than females of developing hypertension and every 1 kg/m² rise in BMI increased the risk of hypertension by 6%. The risk of hypertension was not associated with glycemic control or treatment arm. Hispanics had the lowest prevalence at baseline (8%) compared to non-Hispanic Blacks (14%) and non-Hispanic Whites (18%). Over one third of patients initially treated with an ACE-I for hypertension required an additional medication over the first 3.9 years of study.

Microalbuminuria was observed in 6.3% of TODAY youth at baseline and almost tripled to 16.6% after an average follow-up of 3.9 years³². HbA1C elevation was strongly correlated with microalbuminuria, with every 1% increase in HbA1C increasing the risk of microalbuminuria by 17%. There was no racial/ethnic or gender difference in the development of microalbuminuria in youth with type 2 diabetes. In comparison, in adults with recently diagnosed type 2 diabetes [the United Kingdom Prospective Diabetes Study UKPDS)], 25% developed microalbuminuria within 10 years, 38% within 15 years and 29% had renal impairment over a median of 15 years (over half of this group did not have albuminuria)³⁴. The high rates of hypertension and microalbuminuria observed over a short period of time in the TODAY youth with recent onset of type 2 diabetes may indicate greater long-term risk of premature end stage renal disease and cardiovascular disease.

The prevalence of retinopathy was 13.7% in the TODAY study³³, similar to adult data from the Diabetes Prevention Program (12.6% after 3 years of type 2 diabetes)³⁵. Higher prevalence of retinopathy was seen in TODAY with older age, longer duration of diabetes (average duration was 4.9 +/− 1.5 years) and poorer glycemic control. Surprisingly, higher BMI was associated with a lower prevalence of retinopathy. This apparent “obesity paradox” has been reported in Asian Malay adults, and raises the question whether insulin resistance is protective for development of retinopathy ^{33, 36}.

Dyslipidemia, Inflammatory Markers and Atherosclerotic Disease in Youth-Onset Type 2 Diabetes

Macrovascular diseases (cardiovascular and cerebrovascular disease) are the major cause of morbidity and mortality in adults with type 2 diabetes. Of great concern is the finding that earlier onset type 2 diabetes is associated with a higher risk of having a myocardial infarction³⁷. Youth with type 2 diabetes have abnormal vascular stiffness as measured by aortic pulse wave velocity³⁸. Early changes reflective of atherosclerosis can be detected by assessment of carotid intima-media thickness (IMT)³⁹. In adolescents with type 2 diabetes, every 1% increase in HbA1C or each additional year of diabetes increased the odds of finding greater carotid IMT by 30%⁴⁰. In addition, male sex, and higher LDL-cholesterol (LDL-chol) and tumor necrosis factor–alpha levels increased the likelihood for abnormal carotid IMT⁴⁰. In the SEARCH for Diabetes in Youth Study, 43% of youth with poorly controlled type 2 diabetes had an LDL-chol ≥ 130 mg/dL, and 40% had triglyceride levels ≥ 200 mg/dL⁴¹.

Atherogenic profiles of youth in the TODAY study were assessed annually by measuring lipid levels and inflammatory markers including high-sensitivity c-reactive protein (hsCRP)⁴². Lipid goals defined in the study protocol included LDL-chol < 100 mg/dL and triglycerides < 150 mg/dL. Statin therapy was started for LDL-chol ≥130 mg/dL or triglyceride ≥300 mg/dL. Only 55.9% of the youth remained at the LDL-chol goal over the first 3 years. The number of participants with high risk LDL-chol (≥ 130 mg/dL or using lipid lowering therapy) more than doubled from baseline (4.5% to 10.7%) over 36 months of the study. Levels of triglycerides as well as apolipoprotein B and non-HDL-chol rose significantly from baseline to the end of the first year and remained at a higher level for the next two years of observation. Inflammatory factors [hsCRP, homocysteine and plasminogen activator inhibitor-1 (PAI-1)] also increased during the first year, and low HDL-chol levels were common. All three treatment groups showed worsening of cardiovascular risk factors. LDL-chol levels rose with increasing HbA1C levels in all 3 treatment groups. Triglyceride levels worsened with increasing HbA1C, except in the intensive lifestyle plus metformin arm. Ethnic-specific differences seen in the general population were also reflected in the TODAY cohort, with lower triglyceride levels observed in non-Hispanic Blacks compared to both Hispanic and non-Hispanic Whites. Adherence to diet, exercise and lipid lowering medication is important to lower cardiovascular risk, but this can be particularly difficult in adolescents and young adults.

The high incidence of atherogenic lipid profiles and worsening of cardiovascular risk factors observed during the first 36 months of the TODAY study suggest the possibility of premature onset of cardiovascular disease in the future for youth-onset type 2 diabetes. It has been estimated that children with type 2 diabetes with onset at age 10 years will lose 19 life years⁴³. This underscores the importance of developing effective treatment strategies and interventions to lower this risk.

Treatment Recommendations for Youth with Type 2 Diabetes

The American Academy of Pediatrics recently published recommendations for treatment of type 2 diabetes in children and adolescents (Table 2)^{44, 45}. Lifestyle modification is recommended for all individuals with diabetes. The “Pediatric Weight Management Evidence-based Nutrition Practice Guidelines” from the Academy of Nutrition and Dietetics is helpful for initial and ongoing nutrition counseling and management (http://www.andevidencelibrary.com/topic.cfm?cat=4102&auth=1). Moderate-rigorous exercise is recommended, but adolescents with type 2 diabetes may have impaired exercise capacity⁴⁶. In the TODAY study, all participants received standard diabetes education prior to randomization, which included discussions about healthy eating and physical activity goals (youth who proceeded to randomization decreased weight by 0.68 kg during this run-in period⁴⁷). Further research is needed to identify more effective diet and physical activity plans for these youth and to learn how to best promote and sustain lifestyle change over time.

Metformin is recommended as first line therapy at diagnosis. Approximately half of the youth in the TODAY study were successfully managed on metformin monotherapy. Metformin was safe and is inexpensive. Newer glycemic control medications used in adults with type 2 diabetes, such as dipeptidyl peptidase 4 (DPP-4) inhibitors, glucagon-like peptide 1 (GLP-1) receptor agonists and sodium-glucose cotransporter 2 (SGLT2) inhibitors, were not available when the TODAY study was designed, are not approved for use in the pediatric population and have not been well-studied in youth. Monotherapy with glimepiride has been compared to metformin monotherapy over 24 weeks in pediatric patients with type 2 diabetes⁴⁸. Similar reductions in A1c were observed, but with greater weight gain with glimepiride.

Insulin therapy is recommended when children and adolescents present with ketosis or ketoacidosis, or when random blood glucose levels are ≥250 mg/dL or HbA1C > 9% at diagnosis. Ketoacidosis can be the presenting feature in 5%–25% of patients, requiring careful fluid and electrolyte management as well as insulin therapy. Elevated blood glucose levels can cause reversible beta cell toxicity, further reducing insulin secretion. Insulin therapy can help restore glycemic control quickly and theoretically allow some beta cell recovery. In youth, insulin is also added to metformin therapy when glycemic goals are not maintained on metformin alone.

Bariatric surgery has been shown to be an effective treatment of type 2 diabetes of short duration in adults^{49, 50}, reducing weight and increasing insulin sensitivity. Outcome data for bariatric surgery are lacking in adolescents, except from few specialized centers⁵¹. In a small number of adolescents with type 2 diabetes, Inge et al.⁵² demonstrated significant improvements in BMI, HbA1C and fasting insulin levels after Roux-en-Y gastric bypass surgery compared to a medically managed group. There were also improvements in measures of fatty liver disease, blood pressure and serum triglycerides. Although these outcomes are encouraging, data from larger studies and with longer follow-up are needed before bariatric surgery can be recommended as a safe and effective treatment for type 2 diabetes in adolescents and young adults.

Summary

Results from the TODAY study of youth-onset type 2 diabetes were significant for the following: (1) Metformin monotherapy provided durable glycemic control in only half the participants, indicating that this disease can be more aggressive and rapidly progressive when onset is early in life. It is therefore important to carefully and frequently monitor these patients and intensify therapy, including initiating insulin therapy early, as needed. (2) Although the combination of metformin plus rosiglitazone was better than metformin alone in maintaining glycemic control and preserving beta cell function, given the potential long-term adverse effects with thiazolidinediones, the use of these drugs cannot be recommended. (3) Metformin plus intensive lifestyle did not result in meaningful sustained weight loss and did not significantly improve glycemic control compared to metformin alone. Further evaluation of data from the intensive lifestyle arm is ongoing. (4) The high incidence of microalbuminuria as well as hypertension, dyslipidemia and elevated markers of inflammation early in the course of type 2 diabetes suggest that the development of serious diabetes complications may occur prematurely in youth-onset type 2 diabetes. Careful monitoring for and treatment of hypertension, albuminuria, impaired renal function and retinopathy are essential (Table 2).

Conclusions

The TODAY trial demonstrated the difficulty in treating youth-onset type 2 diabetes, with only half of participants with recent-onset diabetes able to maintain glycemic control on metformin monotherapy. Disappointingly, there was little weight loss even with an intensive lifestyle modification program. The difficulty in treating type 2 diabetes in youth highlights the critical need to promote healthy lifestyle to prevent or postpone the development of type 2 diabetes in those at risk. For individuals with early onset type 2 diabetes, glycemic control must be carefully monitored and treated, adding insulin when necessary. Physical and psychosocial changes of puberty and adolescence in addition to socioeconomic challenges make adherence to treatment regimens particularly difficult in this population. It is important to engage the family, understand the patient and family’s perception of diabetes, and use a culturally appropriate, family centered approach to education and ongoing care. Ongoing surveillance and treatment for cardiovascular risk factors, diabetes-related comorbidities and complications is essential to prevent early morbidity and mortality. Future research will hopefully suggest better treatment options and prevention strategies.

Highlights.

1. Risk factors for development of type 2 diabetes in youth are being overweight, having a family history of diabetes, race/ethnicity (African-American, Hispanic, Native American, Asian/Pacific Islander), signs of insulin resistance (acanthosis nigricans, hypertension, dyslipidemia, polycystic ovarian disease, having been small for gestational age), maternal history of diabetes or gestational diabetes during the child's gestation

2. Metformin is less effective in youth-onset diabetes compared to adults, with poor durability of glycemic control observed in half of youth in the TODAY study

3. Deterioration in glycemic control in the TODAY study was associated with rapid loss of pancreatic beta cell function

4. The addition of an intensive lifestyle change program to standard diabetes education and metformin therapy did not result in meaningful sustained weight loss and did not significantly improve glycemic control in youth in the TODAY study

5. Hypertension, microalbuminuria and dyslipidemia are common in youth-onset type 2 diabetes.

Abbreviations

ACE-I

angiotensin converting enzyme inhibitor

ARB

angiotension receptor blocker

BMI

body mass index

chol

cholesterol

Cr

creatinine

CVD

cardiovascular disease

DPP-4 inhibitors

dipeptidyl peptidase 4 inhibitors

GAD

glutamic acid decarboxylase

GLP-1

glucagon-like peptide 1

GFR

glomerular filtration rate

HbA1C

hemoglobin A1c

IA-2

insulinoma-associated protein 2

LDL

low density lipoprotein

TODAY

Treatment Options for type 2 Diabetes in Adolescents and Youth

SGLT2

sodium-glucose cotransporter 2

UKPDS

United Kingdom Prospective Diabetes Study

US

United States