Racial/ethnic disparities in prevalence and care of patients with type 2 diabetes mellitus

Abstract

Background:

As of 2012, nearly 10% of Americans had diabetes mellitus. People with diabetes are at approximately double the risk of premature death compared with those in the same age groups without the condition. While the prevalence of diabetes has risen across all racial/ethnic groups over the past 30 years, rates are higher in minority populations. The objective of this review article is to evaluate the prevalence of diabetes and disease-related comorbidities as well as the primary endpoints of clinical studies assessing glucose-lowering treatments in African Americans, Hispanics, and Asians.

Methods:

As part of our examination of this topic, we reviewed epidemiologic and outcome publications. Additionally, we performed a comprehensive literature search of clinical trials that evaluated glucose-lowering drugs in racial minority populations. For race/ethnicity, we used the terms African American, African, Hispanic, and Asian. We searched PubMed for clinical trial results from 1996 to 2015 using these terms by drug class and specific drug. Search results were filtered qualitatively.

Results:

Overall, the majority of publications that fit our search criteria pertained to native Asian patient populations (i.e., Asian patients in Asian countries). Sulfonylureas; the α-glucosidase inhibitor, miglitol; the biguanide, metformin; and the thiazolidinedione, rosiglitazone have been evaluated in African American and Hispanic populations, as well as in Asians. The literature on other glucose-lowering drugs in non-white races/ethnicities is more limited.

Conclusions:

Clinical data are needed for guiding diabetes treatment among racial minority populations. A multi-faceted approach, including vigilant screening in at-risk populations, aggressive treatment, and culturally sensitive patient education, could help reduce the burden of diabetes on minority populations. To ensure optimal outcomes, educational programs that integrate culturally relevant approaches should highlight the importance of risk-factor control in minority patients.

Keywords: :

• Ethnicity
• Insulin
• Racial disparities in access to treatment
• Type 2 diabetes mellitus

Introduction

As of 2012, 29.1 million Americans, 9.3% of the population of the United States, had diabetes mellitus, of whom 8.1 million (27.8%) were unaware that they had already sustained end-organ damage from undiagnosed diabetes1. People with diabetes are at approximately double the risk of death than people without the condition in the same age groups1^,2. Disturbingly, disparities still exist in the care and health outcomes of racial/ethnic minorities within the United States. This review was undertaken to examine the trends in type 2 diabetes mellitus (T2DM) prevalence and treatment among racial/ethnic minority populations, and their inclusion in clinical research for new therapies to treat T2DM. We hope that a better understanding of these issues will lead to optimal treatment across all populations affected by T2DM.

While the prevalence of diagnosed diabetes is already significantly elevated at 7.6% among non-Hispanic whites, African Americans are disproportionately affected with a prevalence of 13.2%, a risk for diagnosis that is 74% higher than that for non-Hispanic whites (2014 National Diabetes Statistics Report; 2010–2012 data)1. Among the Latin American/Hispanic population overall, the prevalence of diagnosed diabetes is 12.8%; however, the toll is unevenly distributed among different ethnic groups1. Among Cubans, 9.3% have diabetes, and 8.5% of Central and South Americans have diabetes, whereas the prevalence for Mexican Americans and Puerto Ricans is 13.9% and 14.8%, respectively1. For Asian Americans, the prevalence of diagnosed diabetes is 9.0%1. In the American Indian adult population, the overall prevalence is 15.9%, ranging from 6.0% among Alaskan tribes to 24.1% among those living in southern Arizona1. In a separate evaluation of National Health and Nutrition Examination Survey (NHANES) from 1999 to 2010, the most dramatic increase in the prevalence of diabetes was seen among African Americans. Diabetes prevalence rose from 6.4% in 1999 to 10.3% in 2010 for non-Hispanic whites (p for linear trend = 0.115), whereas African Americans experienced almost a doubling as prevalence grew from 7.9% in 1999 to 14.1% (p for linear trend = 0.008) in 20103. With an increasing incidence of diabetes and decreasing mortality rate, the lifetime risk of developing diabetes has risen to 40% for both men and women (2000–2011 data), compared with 20% and 27%, respectively (1985–1989 data). Hispanics and African American women are at the greatest risk of developing diabetes, and have a lifetime risk exceeding 50%4.

Results of the 2014 National Diabetes Statistics Report confirmed that the prevalence of diabetes increases with age1. The estimated prevalence of 4.1% for both diagnosed and undiagnosed diabetes in the 20–44-year age group may appear trivial; however, rates increase dramatically over time, rising to 16.2% in those aged 45–64 years, and peaking at the substantial prevalence of 25.9% in patients older than 65 years (2009–2012 data)1. The Centers for Disease Control (CDC) reports worrisome trends among American youth, noting surges not only in type 1 diabetes mellitus (T1DM), formerly known as ‘juvenile’ diabetes, but also in T2DM, formerly termed ‘adult-onset’ diabetes. As expected, new cases were recorded more often for T1DM than for T2DM among non-Hispanic white youth aged 10–19 years, but the opposite was true for age-matched Asian/Pacific Islanders and American Indians1. New diagnoses of T1DM and T2DM were reported at similar rates for non-Hispanic black and Hispanic youth aged 10–19 years (2008–2009 data)1.

Disease-treatment outcomes are influenced by a complex network of factors such as genetics, education, adherence to treatment, and socioeconomic status that can ultimately impact access to proper medical care. The progressive nature of T2DM and its link to comorbidities add to the complexity of disease treatment. The objective of this review article is to evaluate the prevalence of diabetes and disease-related comorbidities as well as the primary endpoints of clinical studies assessing glucose-lowering treatments in African Americans, Hispanics, and Asians.

Race/ethnicity and diabetes comorbidities: cardiovascular disease, hypertension, and kidney failure

Diabetes is the leading cause of cardiovascular disease (CVD), hypertension, and kidney failure1^,2. Death certificate data from 2010 indicate that diabetes was the seventh leading cause of death, although the true burden probably was – and remains – under-reported. Between 35% and 40% of death certificates mentioned the diagnosis of diabetes as a co-morbid condition, and 10% to 15% cited diabetes as the cause of death1. Among economically disadvantaged populations that have higher incidences of CVD, early detection and management of risk factors are particularly important5.

Data from the NHANES survey of the predicted 10-year risk of CVD and coronary heart disease (CHD) in adults in the United States between 1999 and 2010 revealed an overall increase in the prevalence of diabetes (p for linear trend = 0.013) while the rate of CVD was maintained (p = 0.152), and CHD significantly improved (p for linear trend = 0.005)3. When the data were stratified by race/ethnicity, no significant improvements in the rates of CVD were identified among non-Hispanic whites, African Americans, or Mexican Americans. No significant improvements in mean systolic blood pressure or total cholesterol were identified among African Americans. A striking feature of the NHANES analysis was that while Americans did better overall in controlling the major modifiable risk factors – blood pressure, cholesterol, and smoking – more of the population developed obesity, and the prevalence of diabetes increased. Excess weight and obesity are well-established risk factors for T2DM as well as dyslipidemia, hypertension, and metabolic syndrome. In addition, the relative risks associated with each of these factors have frequently been reported to be elevated in ethnic populations1.

Hypertension, like diabetes, is a leading risk factor for CVD. In a CDC morbidity and mortality report from 2007 to 2010, African Americans had the highest age-adjusted prevalence (41.3%) of hypertension compared with non-Hispanic whites (28.6%)6. Overall, individuals with diabetes or obesity had a higher rate of hypertension than those without6. A recent and comprehensive cross-sectional study examined the potential relationship between abdominal visceral adipose tissue and cardio-metabolic risk factors in 2035 African Americans from the Jackson Heart Study and 3170 European Americans from the Framingham Heart Study7. Participants from the Jackson cohort were found to have higher rates of obesity, hypertension, metabolic syndrome, and diabetes than Framingham participants (all p = 0.001). The relationship between cardio-metabolic risk factors and visceral adipose tissue was found to be weaker in women from the Jackson cohort, versus those from the Framingham cohort. However, there were stronger associations reported between serum lipids (triglycerides and high-density lipoprotein cholesterol levels) and cardio-metabolic risk factors in the Jackson men than males in the Framingham group. In an investigation of NHANES data from 2005 to 2010, the joint occurrence of abdominal obesity, high blood pressure, and elevated triglycerides correlated with an increased risk of elevated levels of glycated hemoglobin (HbA_1c) (>7%) in non-Hispanic whites (odds ratio [OR]: 2.3), and was even greater in African Americans (OR: 9.1) and Mexican Americans (OR: 4.8)8.

Analysis of the United States Renal Data System from 1995 to 2005 revealed an overall increase in the age-adjusted rate of end-stage renal disease (ESRD) in patients with diabetes in the United States9. In the 2000s the increasing age-adjusted rate of ESRD among African Americans plateaued, and significantly decreased among American Indians, Asians, and Hispanics. In 2005, African Americans with and without diabetes were approximately four times more likely to have ESRD compared with non-Hispanic whites. Although improvements in the rate of ESRD have been noted, ethnic minorities are still disproportionately affected by a high incidence of hypertension and CVD. While excess weight and obesity clearly play a role in the development of CVD, metabolic syndrome, and diabetes, the particular impacts of early childhood factors, sex, and ethnicity will require further study. A strategy proposed by the United States National Prevention Council emphasizes the importance of eliminating health disparities among minority patients and offers recommendations aimed at fostering healthy behavior with the goal of achieving positive change2.

Disparities in outcomes of diabetes care

Studies over decades have shown that racial/ethnic disparities complicate both the delivery and the outcomes of healthcare in the United States. In populations of adults with T2DM, a meta-analysis of 11 studies from 1993 to 2005 revealed that African American patients have mean HbA_1c ∼0.65% higher than non-Hispanic whites10, and a meta-analysis of 11 studies from 1993 to 2007 showed that Hispanic patients had mean HbA_1c ∼0.5% higher than non-Hispanic whites11. Poorer glycemic control and variances in hemoglobin glycation or red cell survival have been suggested as possible reasons for the disparities in HbA_1c levels among racial/ethnic minorities12. On average, African Americans, Hispanics, American Indians, and Asians have elevated HbA_1c levels in the presence of impaired glucose tolerance (IGT)12. More recently, a retrospective analysis of data derived from a large national patient registry indicated that African American patients were significantly more likely than non-Hispanic white patients to receive recommended process of care measures for diabetes and significantly more likely to be receiving insulin therapy13. Despite better adherence to processes of care, however, this population was significantly less likely to have well controlled glycemia than non-Hispanic whites (36.7% vs. 44.0% [p < 0.001]). In addition, African American patients were significantly less likely than non-Hispanic white patients to have well-controlled cholesterol (39.5% vs. 46.8% [p < 0.001]) and blood pressure (29.0% vs. 35.4% [p < 0.001]). Similar observations were made in a United Kingdom cohort study of 24,111 patients of diverse ethnicities14. The researchers found that ethnicity had a stronger effect on diabetic control than socioeconomic status (SES), with South Asian and Black African/Caribbean patients having worse control than non-Hispanic whites.

Racial/ethnic disparities are also evident in the management and outcomes of diabetes-related complications. In 2009, the age-adjusted hospital discharge rate for adults with diabetes was 1.7 times higher among blacks than whites (265.6 vs. 160.5 per 1000 patients)15. The racial disparity in hospitalization rates for ambulatory care sensitive conditions is also consistent for African American versus non-Hispanic white patients with diabetes aged ≥65 years (2006 data)16. A review of studies that looked at ethnic disparities in diabetes-related amputations found no pattern of disparity17, but in the most recent studies it cited, clear differences in rates of amputations between African American and non-Hispanic white patients were shown18^,19. Ethnic variations in diet and physical activity likely contribute to the disparities seen in modifiable risk factors among minorities. The longitudinal Dallas Heart Study found lower rates of physical activity among African Americans and Hispanics compared with non-Hispanic whites when accounting for income, education, age, sex, body mass index (BMI), diabetes, hypertension, and hyperlipidemia20. A recent observational study also showed lower rates of dilated eye examinations among ethnic minorities compared with non-Hispanic white patients with diabetes21^,22. The CDC data for 2010 showed that 64% of both African Americans and non-Hispanic whites with diabetes had received a dilated eye exam in the past year, but only 55% of Hispanic subjects had one23.

Concurrent evaluations of the impact of SES and race/ethnicity on the incidence of diabetes and diabetes complications are limited. In the Boston Area Community Health (BACH) survey (2002–2012 data), the 7-year incidence of diabetes was 128% higher in African Americans and 67% higher in Hispanics than in non-Hispanic white subjects24. When the data were adjusted for SES (income, occupation, and education), however, these racial/ethnic differences were attenuated24. In the BACH study, differences in bio-geographical ancestry that were based on the composition of specific genetic loci were consistent with self-reported race/ethnicity. These effects were also attenuated when data were adjusted for SES. The BACH study concluded that racial/ethnic disparities in the incidence of T2DM were potentially mediated by SES24.

Limited access to healthcare in low socioeconomic populations is often an explanation for racial/ethnic differences in diabetes complications. An observational study from 1995 to 1998 found that ethnic disparities in diabetes complications still exist in patients with uniform healthcare25. Karter et al. concluded that socioeconomics may, therefore, only partially explain the racial/ethnic disparities in the prevalence of diabetes complications such as myocardial infarction, stroke, congestive heart failure, low extremity amputation, and ESRD25.

Oral agents and non-insulin injectables for the treatment of T2DM among racial/ethnic minorities

Despite the disproportionate rates of T2DM and the potential for differences in outcomes between racial minorities and non-Hispanic white populations, few clinical trials to date have enrolled sufficient numbers of ethnic patients to allow racially specific analysis of the effects of treatments.

For this review, we performed a literature search for clinical trials that evaluated glucose-lowering agents in racial/ethnic adult minority patients for the treatment of T2DM. For race/ethnicity, we used the terms African American, African, Hispanic, and Asian. We searched for clinical trial results using these terms by drug class and specific drug then revised the results qualitatively based on actual relevance of the data presented, and counted studies in which the drug was given as a primary intervention in patients with T2DM, insulin resistance, or IGT. We excluded trials that did not include analyses by sub-populations. The corresponding number of randomized patients for each race/ethnicity (African American, Hispanic, and Asian) was also counted. Terms and results are summarized in Table 1. The clinical trials evaluated in our review, thus, included subanalyses of patients based on race/ethnicity. Although observational studies and meta-analyses were not included in the quantified publication analysis, they were evaluated and covered in this narrative review.

Table 1. Results of a detailed search of clinical trials publications with subanalyses of patients by race/ethnicity.

		Publications according to race/ethnicity, n
Medication class	Generic name	Publications with African American pts, n	Total pts*, N/African American pts, n	Publications with Hispanic pts, n	Total pts*, N/Hispanic pts, n	Publications with Asian pts, n	Total pts*,N/ Asian pts, n
Sulfonylureas†	–	5	1120/352	3	1120/401	7	1120/367
α-glucosidase inhibitors	Acarbose	0	665/0	0	665/0	4	665/665
	Voglibose	0	3489/0	0	3489/0	7	3489/3489
	Miglitol	1	780/345	1	780/384	1	780/50
Dipeptidyl peptidase-4 inhibitors	Sitagliptin	0	1879/0	0	1879/0	9	1879/1879
	Saxagliptin	0	1138/0	0	1138/0	2	1138/1138
	Linagliptin	2	1497/267	0	1497/0	3	1497/1230
	Vildagliptin	0	1032/0	0	1032/0	4	1032/1032
	Alogliptin	0	880/0	0	880/0	3	880/880
Biguanides	Metformin	5	1450/7620	4	1146/7620	7	1443/7620
Thiazolidinediones	Pioglitazone	0	1572/0	3	1572/328	8	1572/1243
	Rosiglitazone	1	1239/151	2	1239/162	4	1239/37
Glinides	Nateglinide	0	230/0	0	230/0	1	230/230
	Repaglinide	0	230/0	0	230/0	1	230/230
Glucagon-like peptide 1 receptor agonists	Exenatide	0	1199/0	1	1199/102	5	1199/1002
	Exenatide extended-release	0	–	0	–	0	–
	Liraglutide	0	1277/0	0	1277/0	3	1277/1277
	Lixisenatide	0	702/0	0	702/0	2	702/702
Sodium glucose co-transporter 2 inhibitors	Dapagliflozin	0	608/0	0	608/0	3	608/608
	Canagliflozin	0	–	0	–	0	–
	Empagliflozin	1	498/11	0	498/0	1	498/288
Miscellaneous agents	Bromocriptine	0	–	0	–	0	–
	Colesevelam	0	439/0	2	439/439	0	439/0

*Total number of patients randomized to the study, including other races/ethnicities.

†Specific drug names were not searched in addition to the class name ‘sulfonylurea’.

Overall, the majority of publications that fit these criteria (i.e., included subanalyses of patients based on race/ethnicity) pertained to Asian patient populations studied in their own countries (described below as native Asian); these included clinical studies that report the efficacy of the α-glucosidase inhibitors acarbose and voglibose; the dipeptidyl peptidase-4 (DPP-4) inhibitors sitagliptin, saxagliptin, vildagliptin, and alogliptin; the thiazolidinedione pioglitazone; the glucagon-like peptide-1 (GLP-1) receptor agonists exenatide, liraglutide, and lixisenatide; and the sodium glucose co-transporter 2 (SGLT2) inhibitor dapagliflozin. Sulfonylureas; the α-glucosidase inhibitor miglitol; the biguanide metformin; and the thiazolidinedione rosiglitazone have been evaluated in African American and Hispanic populations, as well as in Asians. The DPP-4 inhibitor, linagliptin, was evaluated in African American and Asian populations, but not in Hispanic populations. The literature on the GLP-1 receptor agonists exenatide extended-release and lixisenatide; the SGLT2 inhibitors canagliflozin and empagliflozin; the glinides nateglinide and repaglinide, as well as other agents such as bromocriptine and colesevelam among non-white races/ethnicities is more limited.

Biguanides: metformin

Clinical algorithms for the treatment of T2DM outline a multi-faceted approach that combines lifestyle optimization and pharmacotherapy26^,27. The AACE (American Association of Clinical Endocrinologists) algorithm encourages individualized therapies that depend on HbA_1c targets and other patient characteristics. Antihyperglycemic therapy typically begins with metformin. The drug has been relatively well studied among racial and ethnic minorities, and clinical findings have demonstrated similar pharmacokinetic properties in healthy non-Hispanic white, African American, and Hispanic populations when metformin is given with pioglitazone28. Recent observational data for 19,672 patients suggest that African Americans may respond to metformin treatment better than non-Hispanic white Americans29. The efficacy of metformin was compared between African Americans and non-Hispanic whites in six strata of baseline HbA_1c, and African Americans experienced greater reductions at all strata. The researchers evaluated confounders including age, sex, BMI, baseline HbA_1c levels, total time on metformin, and exposure to other diabetes modifications, but could not offer any non-racial explanation for these findings.

However, a prospective study in 18 insulin-resistant non-diabetic African American subjects receiving metformin showed significantly increased arterial stiffness30. Additional larger studies are needed to evaluate whether other clinical endpoints such as CVD and death are similarly affected by metformin in African Americans and other racial/ethnic minorities.

Seven clinical trials consisting of Asian and Asian Indian subjects treated with metformin combination therapies were identified. Results of the completed VISION trial (NCT01541956) comparing upper-titration of metformin monotherapy to early use of combination therapy of vildagliptin and metformin in Chinese patients with T2DM have not been published.

Sulfonylureas

Sulfonylureas, such as glyburide, glipizide, and glimepiride, are commonly used as second-line therapies in T2DM26^,27. However, hard endpoints with sulfonylureas have not been well evaluated in ethnic populations. In the aforementioned prospective study by Stakos and colleagues, glipizide also increased arterial stiffness in insulin-resistant non-diabetic African American patients30. Based on prior research showing decreased insulin sensitivity and reduced hepatic insulin extraction in African Americans31, Osei et al. hypothesized that glipizide could be used to prevent diabetes in obese high-risk African Americans with IGT32. These individuals showed improvements in β-cell function and insulin sensitivity, without symptoms of weight gain or hypoglycemia32.

Seven publications reported on sulfonylurea studies of Asian subjects with T2DM. More recently, a fixed-dose pioglitazone/glimepiride combination tablet was shown to significantly improve glycemic control and lipid profiles in Japanese patients with T2DM33.

Thiazolidinediones

Our literature search revealed six clinical trials of the thiazolidinediones pioglitazone and rosiglitazone among African Americans or Hispanics with T2DM. In the clinical trial mentioned above, pioglitazone in combination with metformin was shown to have similar pharmacokinetic properties in healthy non-Hispanic white, African American, and Hispanic populations28. In a non-randomized study consisting of African Americans with IGT and T2DM, rosiglitazone increased post-glucose challenge hepatic insulin extraction, and glycemic control was improved, but not normalized34. In Mexican American subjects with T2DM, 4 months of pioglitazone improved β-cell function and insulin secretion and suppressed gluconeogenesis during fasting and glucose ingestion35.

Twelve clinical trials have been published since 2003 reporting results from Asian cohorts treated with either pioglitazone or rosiglitazone. Native Japanese individuals with T2DM who were randomized to rosiglitazone or pioglitazone (vs. placebo) demonstrated a greater response to pioglitazone in HbA_1c levels and no difference between the change in fasting plasma glucose (FPG) after 28 weeks of either drug treatment36. However, similar to findings in other patient populations, pioglitazone treatment in Japanese patients was associated with more adverse events relating to edema and weight gain compared with rosiglitazone.

α-glucosidase inhibitors

Miglitol

Two clinical trials evaluated the efficacy of miglitol treatment in Hispanic37 or African American38 patients with T2DM that were previously treated with diet alone or sulfonylurea. African American patients (baseline HbA_1c 8.7%) exhibited a placebo-corrected reduction in HbA_1c from baseline of −1.19% at month 6. Treatment of Hispanic patients (baseline HbA1c 8.7%) with miglitol resulted in a −0.26% least-squares mean reduction in HbA_1c at month 6. Miglitol treatment has been evaluated in one clinical trial consisting of Asian subjects from Japan. During a 12-week administration of miglitol or voglibose, both drugs were found to increase postprandial GLP-1 responses and reduce the level of gastric inhibitory polypeptide (GIP)39.

Acarbose

No clinical trials have evaluated acarbose treatment of African American or Hispanic patients with T2DM. Acarbose has been evaluated in four clinical trials of native Asian patients with T2DM. In native Asian patients inadequately controlled with sulfonylurea (baseline HbA_1c 9.0%), the addition of acarbose was well tolerated and significantly improved glycemia, which resulted in an HbA_1c placebo-corrected reduction of −1.05% (p = 0.0018) from baseline in the intent-to-treat population40.

Voglibose

Voglibose therapy has not been evaluated in African American or Hispanic populations. Seven clinical trials of voglibose have included native Asian subjects. Sitagliptin once-daily monotherapy was found to be superior to thrice daily voglibose for HbA_1c lowering, providing a reduction from baseline (HbA_1c 7.8%) in HbA_1c of −0.7% vs. −0.3% with voglibose at week 12 (p < 0.001). Voglibose was not tolerated as well as sitagliptin, demonstrating more clinical adverse events (AEs), drug-related AEs, and gastrointestinal AEs compared with sitagliptin41.

Dipeptidyl peptidase-4 inhibitors

Sitagliptin

Sitagliptin was the first DPP-4 inhibitor to be approved by the United States Food and Drug Administration (FDA). Clinical studies with sitagliptin have primarily included Caucasian or native Asian subjects. No publications specifically pertaining to African American, Hispanic, or Asian study populations in the United States were identified in our literature search. Sitagliptin has been evaluated as monotherapy in Asian patients (in China, India, and Korea; baseline HbA_1c 8.7%) over 18 weeks42 and as add-on to metformin in Chinese patients (baseline HbA_1c 8.5%) over 24 weeks43, resulting in HbA_1c placebo-adjusted reductions of −1.0% and −0.9%, respectively. In an observational study of Japanese patients with T2DM (baseline HbA_1c 8.0%), 31% of patients taking sitagliptin for 12 months achieved an HbA_1c ≤7.0% and improved FPG and postprandial glucose (PPG), with no impact on body weight44.

Saxagliptin

Several studies included patient populations consisting of 0.2–7.7% African Americans, and one study included 61.0% Hispanic subjects, but these racial subpopulations were not separately analyzed45–52. Among drug-naïve native Asian subjects with T2DM (baseline HbA_1c 8.1%), 24-week treatment with saxagliptin resulted in a −0.84% adjusted mean change in HbA_1c and reductions in both FPG and PPG, with few serious AEs53. In that study, 46% of patients treated with saxagliptin achieved HbA_1c <7.0%.

Linagliptin

Linagliptin was evaluated in 226 African American patients with T2DM in a randomized placebo-controlled trial lasting 24 weeks54. Patients were either treatment naïve or had been treated with one oral glucose-lowering drug. Treatment of patients with linagliptin (baseline HbA_1c 8.7%, body weight 95.8 kg) lowered HbA_1c by a placebo-corrected mean of −0.58% (p < 0.001). Mean weight change in both treatment groups was −1.1 kg, and AEs were observed in similar proportions of both groups. The pharmacokinetics of linagliptin were previously evaluated in African American patients, revealing no racially based differences55.

In a retrospective analysis, data were pooled from Phase 3 studies that had evaluated linagliptin monotherapy or combination therapy in Hispanic patients from Argentina, Brazil, Canada, Mexico, Peru, and the United States (baseline HbA_1c 7.7–8.2%, FPG 8.6–9.2 mg/dL). Linagliptin 5 mg led to significant improvements in HbA_1c (placebo-adjusted difference, −0.63%, p < 0.0001, week 18; −0.58%, p < 0.0001, week 24) and FPG (placebo-adjusted difference, −11.7 mg/dL, p = 0.0028, week 18; −14.1 mg/dL, p = 0.0004, week 24) in Hispanic patients who had been treated with a broad background of glucose-lowering drugs. Low rates of AEs and incidence of hypoglycemia were observed in similar proportions in linagliptin and placebo-treated patients56. Long-term treatment of Japanese subjects with T2DM with linagliptin as monotherapy improved HbA_1c levels with no effect on body weight and few serious AEs57.

Vildagliptin

Although not approved in the United States, vildagliptin is a DPP-4 inhibitor that has been shown to be effective and well tolerated in patients with T2DM. Clinical trials have provided evidence that the pharmacokinetics of vildagliptin in healthy subjects are not affected in a clinically meaningful way by ethnicity (primarily non-Hispanic white, African American, and Hispanic), or by sex58.

Among the four Asian clinical trials identified, the addition of vildagliptin to metformin was well tolerated and improved glycemic control in patients with T2DM59. Yoshioka et al. reported in a prospective study (N = 19) that vildagliptin can improve glycemic control as well as glycemic fluctuations when added to sulfonylurea therapy in Japanese patients with T2DM60.

Alogliptin

The effects of alogliptin treatment have not been assessed for differences according to race/ethnicity in US-based clinical trials. In native Japanese patients with T2DM, alogliptin added to voglibose has been shown to significantly reduce HbA_1c, FPG, and PPG in comparison with placebo over 52 weeks61.

Glucagon-like peptide-1 receptor agonists

Exenatide

In 2005, the FDA approved the first GLP-1 receptor agonist, exenatide, for the treatment of T2DM. Despite its approval almost 10 years ago, few publications report on treatment findings according to race/ethnicity. No publications report on the efficacy of exenatide treatment in African American study populations, and a study (N = 30) by Schwartz et al. consisting of 60% Hispanics and 7% African Americans did not discuss the impact of ethnicity on the findings62. Forti and colleagues reported findings which suggested that exenatide administration in Hispanic patients with T2DM before breakfast and dinner was non-inferior to administration before lunch and dinner63.

The recently reported CONFIDENCE trial in native Asian patients (N = 416) in China compared changes in HbA_1c and β-cell function with exenatide, insulin, and pioglitazone over 48 weeks64. All three regimens resulted in significant improvements in HbA_1c, with numerically superior efficacy in the exenatide group. Measures of β-cell function also showed improvement with all three regimens. Clinical studies with exenatide extended-release therapy in African American, Hispanic, or Asian patients have not been published to date.

Liraglutide

Liraglutide was approved by the FDA in 2010. No publications have reported on racial/ethnic variances in treatment outcomes among African American and Hispanic patients with T2DM. Three publications reported on clinical trials with liraglutide in Asian subjects. A 12-week trial in Chinese patients with T2DM and obesity compared liraglutide as add-on to insulin versus increased insulin dose65. Patients treated with liraglutide (baseline HbA_1c 8.8%, body weight 88.6 kg) displayed a decrease in central body fat (waist circumference, body weight, and BMI), and 67% of patients with no hypoglycemia and no weight gain achieved an HbA_1c ≤7.0% compared with 19% of patients in the increased insulin treatment group.

In a retrospective study, 46 Japanese patients previously treated with insulin or oral glucose-lowering drugs switched to 6 months of liraglutide therapy66. Patients experienced significant decreases in subcutaneous and visceral fat, but no significant change in HbA_1c. When patients who had received pre-treatment pioglitazone (n = 9) were separated from the analysis, however, a significant reduction in HbA_1c was observed in the remaining population.

Lixisenatide

Lixisenatide is the newest GLP-1 receptor agonist. No publications have reported clinical studies with lixisenatide in primarily African American or Hispanic populations. In native Asian patients in China who were inadequately controlled with metformin with or without a sulfonylurea, lixisenatide treatment was well tolerated and significantly reduced HbA_1c levels and 2-hour PPG67. Lixisenatide also significantly improved glycemic control in a study of native Asians from Japan, the Republic of Korea, Taiwan, and the Philippines, who were insufficiently controlled with insulin with or without a sulfonylurea68.

Sodium glucose co-transporter 2 inhibitors

The SGLT2 inhibitors, a recently launched class of anti-diabetes drugs, use an insulin-independent method of regulating glucose homeostasis. Dapagliflozin, canagliflozin, and empagliflozin are approved in the United States for the treatment of T2DM.

Dapagliflozin

No clinical trials have evaluated dapagliflozin treatment of African American or Hispanic patients with T2DM. Three studies that treated Asian subjects with dapagliflozin were identified. In a Phase 3 study, native Japanese patients with T2DM and inadequate glycemic control were treated with dapagliflozin (baseline HbA_1c 7.5%) for 24 weeks, resulting in significant adjusted mean reductions in HbA_1c (−0.41 to −0.45% vs. placebo −0.06%) and FPG, with a low risk of hypoglycemia69. In healthy Japanese subjects and Japanese patients with T2DM, dapagliflozin was well tolerated and demonstrated predictable pharmacokinetic and pharmacodynamic properties70.

Canagliflozin

No clinical trial publications reporting studies with canagliflozin in specifically African American, Hispanic, or Asian subjects with T2DM were identified in the literature search.

Empagliflozin

Empagliflozin is the most recently approved drug in the SGLT2 inhibitor class. Our literature search identified no published trials of empagliflozin with Hispanic enrolled populations. In a subgroup analysis of empagliflozin as add-on to pioglitazone or pioglitazone plus metformin in patients with T2DM, there was no interaction between treatment and race in a population that included African American (n = 11) and Asian patients (n = 288) with T2DM71. A Phase 3 placebo-controlled study is currently recruiting patients to evaluate the efficacy and safety of empagliflozin treatment in hypertensive African Americans with T2DM (NCT02182830).

A global population pharmacokinetic study of five pooled randomized clinical trials suggested that Asian patients (primarily Japanese, Taiwanese, and Koreans) with T2DM may experience slight increases in the rate of oral empagliflozin absorption compared with the overall population72.

Conclusions

Diabetes is a chronic disease with profound consequences on a patient’s overall health and quality of life. The treatment and long-term care of people with diabetes imposes a significant burden on the United States healthcare system.

While the prevalence of diabetes has risen in all racial/ethnic groups over the past 30 years, rates are higher in minority populations, especially African Americans. However, African American and Hispanic patients are not well represented in clinical trials of glucose-lowering therapies. Clinicians treating these patients must inform their treatment recommendations based on extrapolations, and conjecture trial data in predominantly white populations. Based on our review of the literature, the numbers of patients randomized to clinical trials of sulfonylureas and metformin were approximately equally distributed among African American, Hispanic, and Asian subjects, whereas the other glucose-lowering agents were primarily studied in Asian populations in their native countries.

Current treatment guidelines present strong algorithms for the management of hyperglycemia, insulin resistance, and other morbidities associated with chronic diabetes – many of which occur with especially high incidence in African Americans. For all patients, guidelines stress lifestyle changes, and recommend ever more aggressive therapeutic regimens to achieve treatment goals. An expanding armamentarium for the treatment of T2DM targets the multiple pathophysiological defects present in the disease, described as the ‘ominous octet’73. Due to the progressive nature of the disease, combination therapy is commonplace among patients with uncontrolled T2DM. But observational data have demonstrated that African Americans and Hispanics are less likely to intensify treatment than non-Hispanic white patients with T2DM74. Both medication adherence and treatment strategies can contribute to these variances in intensification.

Suggested strategies to reduce disparities in healthcare include interventions to improve healthcare providers’ awareness of racial/ethnic disparities and improve communication by utilizing racially/ethnically stratified clinical performance feedback and improved measures for collecting information from minority patients75. Cultural competency training is generally well received by primary care clinicians, but the expanded knowledge gained from such initiatives can be difficult to translate into action. Interventions to address disparities in diabetes care in minority patients should couple clinician-focused programs with broader, multi-level approaches that target patients and communities to improve diabetes outcomes and promote healthy behavior. In time, the emerging results from studies to explore and eliminate barriers to optimal diabetes care for minority patients will be incorporated into the standard-of-care, with the aim of improving health at both individual and national levels.

Transparency

Declaration of funding

This review was sponsored by Boehringer Ingelheim Pharmaceuticals Inc., Ridgefield, CT, USA. The authors received no direct compensation related to the development of the manuscript.

Declaration of financial/other relationships

K.C.F. has disclosed that he has participated in Speaker’s Bureaus for AstraZeneca, Novartis, and Takeda Pharmaceutical Company; he has received grant and/or research support from Daiichi Sankyo and Boehringer Ingelheim; and he has served as a consultant for Daiichi Sankyo, Novartis, and Boehringer Ingelheim. S.A.N. has disclosed that she has no significant relationships with or financial interests in any commercial companies related to this study or article.

CMRO peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors meet criteria for authorship as recommended by the International Committee of Medical Journal Editors (ICMJE). The authors received no direct compensation related to the development of the manuscript. Writing support was provided by José L. Walewski PhD and Daniella A. Babu PhD of Envision Scientific Solutions, which was contracted and funded by Boehringer Ingelheim Pharmaceuticals, Inc. (BIPI). BIPI was given the opportunity to review the manuscript for medical and scientific accuracy as well as intellectual property considerations.