Introduction
Type 2 diabetes mellitus (T2DM) is very common in older individuals. Indeed, US data from the Centers for Disease Control and Prevention in 2011 showed that the percentage of diagnosed diabetes was more than 13 times higher in people aged 65–74 years when compared with those younger than 45 years of age; the overall percentages for these two populations were reported to be 21.8% and 1.6%, respectively [
1]. Furthermore, the most recent International Diabetes Federation Diabetes Atlas estimates that the global burden of diabetes in older individuals will increase greatly by 2030 (between 42% and >50%) in people over the age of 60 [
2].
There are a number of treatment options available for patients with diabetes; however, many of them are associated with significant clinical adverse events (AEs). Metformin, which is considered the standard initial therapy for T2DM in conjunction with diet and exercise regimens [
3], is associated with gastrointestinal AEs and, rarely, lactic acidosis [
4]. It is also contraindicated in patients with compromised renal function: the package insert states that metformin treatment “should not be initiated in patients ≥80 years of age unless measurement of creatinine clearance demonstrates that renal function is not reduced” [
4], which is a common comorbid condition in older patients. Moreover, serum creatinine levels may be unreliable in an older patient with low muscle mass. Similarly, sulfonylureas, which are often prescribed as an initial therapy in patients who are not overweight, or as an additional therapy in patients who have failed to achieve adequate glycemic control with metformin alone, are associated with hypoglycemia and weight gain [
3]. Hypoglycemia is of particular importance in older patients because drug-induced hypoglycemic episodes may go unrecognized and may be confounded by cognitive dysfunction. Hypoglycemic episodes are more common and often more serious in older patients receiving oral antidiabetic drugs and can lead to complications such as falls and hip fractures, as well as even more serious cardiovascular events and death [
5,
6]. These AEs and complications all limit the long-term use of sulfonylureas. This is particularly true in older adults who often have underlying health problems and may also be receiving treatment for other comorbid conditions.
Alogliptin is a dipeptidyl peptidase-4 (DPP-4) inhibitor that has shown efficacy as a treatment for T2DM, either as monotherapy in conjunction with diet and exercise or as an adjunct therapy to other diabetic treatments such as sulfonylureas, metformin, thiazolidinediones, or insulin [
7‐
11]. In particular, in an exploratory retrospective pooled analysis of one phase 2 and five phase 3 studies, alogliptin was shown to be consistently efficacious and demonstrated a good safety profile in patients aged between 65 and 80 years. These benefits were seen when alogliptin was administered both as monotherapy and as an add-on therapy, and was not found to increase the incidence of hypoglycemia, weight gain, or other AEs when compared with younger patients [
6].
In this exploratory, post hoc analysis of a prospective study specifically designed to evaluate the efficacy of alogliptin and glipizide, a sulfonylurea, we focused on patients aged 65–90 years over a 52-week period [
12]. The purpose of the study was to explore the benefits of alogliptin in achieving glycemic control without the risk of weight gain, hypoglycemia, or both, as compared with glipizide in this population of patients with T2DM.
Discussion
Alogliptin demonstrated similar efficacy to glipizide in lowering HbA1c in this older population of patients with T2DM, as previously reported in the original clinical study [
12]. This exploratory analysis has extended the data from the original study and demonstrated that significantly more patients in the alogliptin group achieved an HbA1c goal of ≤7.0% without hypoglycemia and weight gain when compared with the glipizide group. In general, in this older population of patients with T2DM, alogliptin was well tolerated and acceptably safe with significantly less hypoglycemia compared with glipizide; full tolerability data have been reported previously [
12].
The assessment of HbA1c is a vital part of the clinical assessment of glycemic control. Most guidelines recommend a target HbA1c of 7.0%, but emphasize the need to individualize treatment targets, especially in older patients with comorbidities [
14]. Assessing other clinical outcomes, particularly safety, alongside HbA1c is now recognized as an important part of the personalized management of T2DM [
14]. Indeed, the use of a patient-centered approach for the treatment of patients with chronic diseases is one of the key factors in evidence-based medicine. This approach takes into consideration individual patient preferences, needs and values [
3]. It is particularly appropriate for patients with T2DM, especially older patients who often have comorbid conditions that are contraindicated for certain treatment options.
The extent to which HbA1c levels should be corrected in elderly patients is still in question. Data from three short-term studies, the Action to Control Cardiovascular Risk in Diabetes (ACCORD) [
15], the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified-Release Controlled Evaluation (ADVANCE) [
16], and the Veterans Affairs Diabetes Trial (VADT) [
17], have shown no reduction in primary cardiovascular endpoints with tight glycemic control when compared with standard glycemic control in older patients. Indeed, the ACCORD study showed a 22% increase in total mortality with the intensive therapy regimen in patients aged 40–79 years. The cause of the unchanged or even negative effect of intensive therapy on cardiovascular endpoints is unclear but older age is a risk factor for severe hypoglycemia [
18], and older patients are at greater risk of hypoglycemia-associated morbidity, which could be a contributory factor [
19]. This increased risk of severe hypoglycemia is thought to be due to a combination of age-related comorbid conditions present in this population as well as a lack of awareness of the symptoms of hypoglycemia, such as cognitive dysfunction, weakness, and unsteadiness, which are often mistaken for other conditions. Hypoglycemia in this population can lead to significant morbidity, including serious vascular events such as stroke and myocardial infarction [
19]. Therefore, avoiding—or at least reducing—hypoglycemia risk in the older patient is particularly important and will impart significant clinical benefits. Indeed, the importance of avoiding hypoglycemia in the elderly is uniformly recognized as a key feature in the treatment of this patient population. The recently updated Canadian Diabetes Association (CDA) guidelines recognize the importance of hypoglycemia avoidance over achieving glycemic targets in some elderly patients, including those who are frail or demonstrate cognitive impairment [
20]. These guidelines suggest that, in these patients, sulfonylureas should be used with caution, further highlighting the need for individualized treatments.
Weight gain has been associated with increased cardiovascular risk in patients with T2DM and is, therefore, another important clinical outcome that should be carefully controlled [
21]. The benefits of DPP-4 inhibitors in terms of weight neutrality and no increased risk of hypoglycemia have been recognized in the recent position statement published by the ADA and the European Association for the Study of Diabetes (EASD) [
3]. The use of composite analyses, such as the ones used in this exploratory study, allows for the assessment of more than 1 clinically relevant outcome. This is important in the decision-making process for selecting treatment options for T2DM, particularly in the older patient, because often there is a need to focus on and consider more than 1 therapeutic goal. A number of studies comparing the efficacy of DPP-4 inhibitors and sulfonylureas on composite outcomes, comprising HbA1c targets, without hypoglycemia or weight gain have now been published. Similar to the results found in the current analysis, sitagliptin and saxagliptin, in addition to metformin, were found to result in more adult patients with T2DM achieving HbA1c targets without hypoglycemia or weight gain compared with glipizide [
22,
23]. Vildagliptin and linagliptin have also been shown to help more T2DM adult patients inadequately controlled with metformin reach a composite endpoint of HbA1c <7.0% with no hypoglycemia and no weight gain when compared with another sulfonylurea, glimepiride [
24,
25]. Although these studies were all performed in adult patients and were not specifically conducted in older patient cohorts, one study [
24] that examined the effectiveness of antidiabetic drugs in different age groups found that vildagliptin was superior to glimepiride in achieving an HbA1c target of <7.0% with no hypoglycemia and no weight gain in patients of all ages, including those aged 60 to < 70 and 70 to <80 years.
In our original study [
12], alogliptin and glipizide resulted in similar reductions in HbA1c at Week 52 from baseline using the last observation carried forward (the primary endpoint), as well as reductions in observed HbA1c values throughout the 52-week period. However, in both groups, reductions from baseline to Week 52 in the primary endpoint were only small; a finding that was not predicted. In an exploratory analysis performed on the data from the per-protocol set in the original study, a comparison between patients who completed the study and those who received rescue medication was performed [
12]. This analysis found that while patients who completed the study (i.e., who did not receive rescue medication) demonstrated reductions in HbA1c within the magnitude expected with alogliptin or glipizide, those who did receive rescue medication actually showed an increase in HbA1c levels. Therefore, it would appear that the inclusion of rescued patients, with the last observation carried forward, may have been a contributing factor to the unexpectedly small reductions in HbA1c from baseline to Week 52 [
12]. A further possible explanation was provided by the post hoc analysis of drug-naïve versus washout patients. As observed in patients who completed the study, drug-naïve patients showed predicted reductions in their levels of HbA1c. However, patients entering the study after the 4-week washout period demonstrated a negligible change in HbA1c levels. Therefore, the relatively short washout period in patients who had already received antidiabetic monotherapy may have been a second contributory factor toward this unexpected finding.
The composite data from this current exploratory analysis further support the primary data previously published. In these data, the alogliptin group demonstrated considerably fewer hypoglycemic episodes and as well as significant body weight decreases, compared with the glipizide group [
12]. Indeed, while there were small but significant weight reductions observed in the alogliptin group, weight increases were observed in the glipizide group [
12]. These findings are important, particularly in older patients who are at risk of further complications associated with hypoglycemia.
Although the numbers of patients with a baseline HbA1c of ≥8.0% in the alogliptin and glipizide groups who achieved both composite endpoints were not significantly different, the sample sizes were very small, making the interpretation of results difficult. In addition, it is possible that patients with higher baseline HbA1c levels may be less responsive to treatment and therefore less likely to achieve HbA1c ≤7.0% without hypoglycemia or weight gain.
There are study limitations that warrant discussion. For example, the titration of glipizide was limited to 10 mg in this population of patients. This was due to the age of the patient population, as older patients are more at risk of complications, including hypoglycemia, with higher doses of glipizide. Limiting the titration avoided any imbalances in hypoglycemia that could have been attributed to inappropriately high doses of glipizide in this older population. Another potential limitation relates to the post hoc nature of this report; this type of analysis is an appropriate exploratory examination of the primary data further supporting the initial findings, and one that is often performed after the initial study is complete. Lastly, although the results from this study provide important data on the treatment of older patients with T2DM, interpretation of the results must be limited to those between the ages of 65 and 90 years, as per the primary objective of the study.
Diabetes is a complex disease with many contributory factors. Treatments that can target more than one aspect of the condition will offer patients additional benefits. For example, in a systematic review and meta-analysis of 53 studies, 17 of which reported endpoint lipid levels, DPP-4 inhibitors were found to significantly reduce total cholesterol levels compared with controls [
26]. Other studies have also demonstrated improvements in lipid levels with DPP-4 inhibitors [
27‐
32], with these agents reportedly being more effective at reducing lipid levels than alpha glucosidase inhibitors [
32]. Within class differences have been observed between the DPP-4 inhibitors, with alogliptin showing a greater reduction in total cholesterol and LDL levels than sitagliptin [
28]. The efficacy of vildagliptin, sitagliptin and saxagliptin for reducing lipids varies, with one study showing vildagliptin to be more effective at reducing total cholesterol than sitagliptin and saxagliptin [
30], and another showing vildagliptin to be more effective than sitagliptin at reducing total cholesterol and triglycerides [
29]. Although further studies are still required, these differences between and within drug classes suggest some treatments may be of more value to certain populations of patients with T2DM than others, depending on their characteristics.
Sulfonylureas have been available for many years and the efficacy, safety and tolerability profile of this class of drug is well established. While DPP-4 inhibitors also have a good safety profile, the recent observation that linagliptin may result in liver toxicity demonstrates that additional findings are still being reported with this newer class of drugs [
33]. When considering the value of individual treatments for different patient populations, both efficacy and safety should be considered. For example, elderly patients often have multiple comorbidities and although the well-established safety and tolerability profile of sulfonylureas is beneficial in this population, the risk of hypoglycemia with these agents may be of particular concern. In addition, cost also plays a part in the treatment decision. Cost is one of the main disadvantages of DPP-4 inhibitors [
34]. This is particularly true in countries, such as Germany, where DPP-4 inhibitors are not reimbursed [
35]. It is in these situations that value-based assessments are important. It is possible that, on balance, in elderly patients who are at high risk of hypoglycemia and may also have other comorbid conditions, DPP-4 inhibitors are likely to offer more value than the sulfonylureas. As such, more and more focus is being placed on personalized medicine in older patients with T2DM, addressing not only glycemic control but also other factors that might increase morbidity and mortality in this population [
36‐
40]. In light of concerns that aggressively lowering HbA1c may be harmful in older patients with diabetes, especially those who are frail or have multiple comorbidities [
15,
41], flexible glycemic targets are more desirable in this patient group. By examining composite endpoints that encompass not only indicators of efficacy but also issues of concern for both clinicians and patients, such as hypoglycemia and weight gain, the current analysis attempts to provide an insight into the overall success of glycemic control and its adverse effects in older patients. Such an approach may also offer a way of differentiating the various treatment options available for T2DM in older patients. Treatment options for T2DM are shown to have similar efficacy in terms of glycemic control; hence, additional clinical outcomes may add value and play a more important role. Data such as these can only help clinicians in getting closer to achieving a more personalized approach to T2DM treatment in this high-risk population of patients.
Acknowledgments
Sponsorship and article processing charges for this study were funded by Takeda Global Research & Development Center, Inc., Deerfield, IL, and Takeda Global Research and Development Centre Ltd., London, United Kingdom. Manuscript writing and editorial assistance in the preparation of this manuscript was provided by Sarah Baldock, Gillian Gummer and Pin Lu (Rx Communications, UK), with funding from Takeda Pharmaceuticals North America, Inc. The sponsors were involved in the design, methods, analysis and preparation of this paper.
The authors wish to thank the patients and investigators at the 110 study sites across 15 countries for their participation in this global study. Thanks must also go to Hung Lam (Takeda Global Research & Development Center, Deerfield, IL), who assisted with the statistical analyses for this study.
All named authors meet the ICMJE criteria for authorship for this manuscript, take responsibility for the integrity of the work as a whole, and have given final approval to the version to be published. Morgan Bron, Penny Fleck, and Craig Wilson contributed to the study design, analysis and interpretation of the data, and critical review of the manuscript.