The Expanding Role of New Medications in the Treatment of Type 2 Diabetes Mellitus
We examined the prevalence of use and cost impact of DPP-4 inhibitors and GLP-1 receptor agonists in people with diabetes.
It is estimated that 26 million people in the United States have diabetes mellitus.1 Diabetes mellitus describes a number of metabolic conditions, but the most common forms are type 1 diabetes mellitus, a condition in which the body destroys insulin cells, requiring the person to receive insulin injections to properly control their blood glucose levels; and type 2 diabetes mellitus (T2DM), whereby the body is unable to either produce or respond to insulin in a manner sufficient to maintain normal blood glucose levels. Our analysis will focus on T2DM, which accounts for 90% to 95% of all cases of diabetes in the United States.1
Over 1.4 million new cases of diabetes are diagnosed each year among people over the age of 40 years—virtually all are T2DM. Risk factors for T2DM include older age, obesity, and inactivity.1 As the prevalence of these risk factors will increase with the aging of the baby boomer generation, the prevalence of T2DM is expected to increase dramatically as well. T2DM places a tremendous burden on the US healthcare system. People with T2DM have twice the mortality risk of people of similar age without T2DM. The Centers for Disease Control and Prevention reports that diabetes (types 1 and 2 combined) places a burden of over $174 billion annually on the US economy ($116 billion in direct medical costs and $58 billion from disability and productivity loss).1
The goal of treatment for a diabetic patient is to maintain blood glucose levels at normal or near-normal levels. This is essential in the prevention of diabetic complications, as increased blood glucose levels are associated with loss of vision (retinopathy), kidney failure, and cardiovascular disease.2 The recommendations for medical management of diabetes have evolved with the introduction of new treatment options, particularly incretin-based therapies—dipeptidyl peptidase-4 (DPP-4) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists.
The incretin response, or activity, is reduced in adults with T2DM, resulting in excess blood glucose. DPP-4 is an enzyme that breaks down incretins and, by inhibiting DPP-4 production, incretin levels in the blood increase, resulting in the reduction of excess blood glucose.3 GLP-1 receptor agonists increase the levels of cyclic AMP inside cells by binding to GLP-1 receptors found in the pancreas which, when blood glucose concentrations are elevated, results in insulin secretion. As blood glucose concentrations decrease and approach normal levels, insulin secretion decreases.4
We will examine how utilization and prescribing patterns have evolved for patients with existing diabetes and those newly diagnosed, paying particular attention to the role of GLP-1 receptor agonists and DPP-4 inhibitors in members who were historically treated with biguanides and/or sulfonylureas.
METHODS
We conducted 3 related analyses using deidentified CVS Caremark pharmacy benefit claims from the funded commercial population from 2008 to 2012. For all analyses, members were required to be at least 40 years old and members with prescription fills for insulin were excluded to eliminate people with type 1 diabetes mellitus from our sample. Our first analysis was a description of members with prevalent diabetes between 2008 and 2012. The cohort used for this analysis consisted of all members who had at least 1 prescription fill for a diabetes medication between 2008 and 2012 (we refer to this as the Prevalence Cohort).
In the second analysis we examined patterns of medication use among members who were new users of diabetic medication. This cohort consisted of members using a diabetic medication between 2008 and 2012 who did not have a prescription fill for that medication in the 6 months prior to their first prescription fill between 2008 and 2012. In order to establish treatment-naïve status, we required these members to be continuously eligible for benefi ts for 6 months prior to their first fill of diabeticmedication (we refer to this as the Naïve Cohort). In the third analysis, we described trends in the use of DPP-4 inhibitors and GLP-1 receptor agonists as an adjunct treatment to biguanides and sulfonylureas. For this, we identified a cohort of members who had at least 1 prescription fill for either a biguanide or a sulfonylurea in 2011 and no prescription fills for either a DPP-4 inhibitor or GLP-1 receptor agonist that year. We required these members to be continuously eligible from 2011 to 2012, over the age of 40 years in 2011, and have no prescription fills for a DPP-4 inhibitor and/or a GLP-1 receptor agonist 6 months prior to their fi rst biguanide/sulfonylurea fill in 2011. We identified members of the 2011 cohort using biguanides and sulfonylureas who added (or switched to) a DPP-4 inhibitor or GLP-1 receptor agonist in 2012. Switching away from a sulfonylurea to a GLP-1 receptor agonist and/or a DPP-4 inhibitor was defi ned as the last fill of sulfonylurea before the first fill of either a GLP-1 receptor agonist or a DPP-4 inhibitor, allowing for a 7-day overlap. We refer to this as the Uptake Cohort.
In these analyses, cost was defined as the total paid charges for the medication by combining the payer and member portions of the health plan’s allowable charges. Our statistical analyses were conducted using SAS 9.1 (SAS Institute, Cary, North Carolina).
RESULTS
Among plan members 40 years and older, we found that 9.6% used diabetes medications and on average, 1.2% started diabetes therapy each year during our 5-year observation period. Between 2008 and 2012, the prevalence of diabetic medication use increased by 1.4 percentage points, from 8.7% to 10.1% (
Figure 1
). There was little change in incidence of diabetic medication use between 2008 and 2012, as it fluctuated between 1.0% and 1.3%. The average age in our Prevalence Cohort was 62.2 years (standard deviation [SD]: 11.8 years), and 49.1% were male. Members of our Naïve Cohort were younger than those in the Prevalence Cohort, with an average age of 58.6 years (SD 11.1), and 52.3% were male.
In
Table 1
, we describe trends in medication use for the Prevalence Cohort. Between 2008 and 2012, utilization of biguanides as monotherapy or in combination increased by 5.3 percentage points, from 46.1% to 51.4%. During this same period, utilization of sulfonylurea as monotherapy or in combination declined by 3.9 percentage points, from 27.7% to 23.8%. Utilization of DPP-4 inhibitors as monotherapy or in combination increased from 8.3% in 2008 to 14.2% in 2012, and utilization of GLP-1 receptor agonists increased by 1.0 percentage point, from 3.5% in 2008 to 4.5% in 2012.
In
Table 2
, we detailed the observed trends in cost of pharmaceutical treatment for people with diabetes from 2008 to 2012. Between 2008 and 2012, the average annual cost of providing pharmaceutical care for a person with T2DM increased from $667 to $741 (11.1%). During this period, the cost of the those drugs most commonly prescribed for T2DM—biguanides, sulfonylureas, and their combination—declined by over 33%, driven by a generic dispensing rate in excess of 96% for these drug classes. Conversely, the observed increase in cost of care for people with T2DM was driven by the increased annual cost of medication in less commonly prescribed drug classes: the DPP-4 inhibitors, GLP-1 receptor agonists, amylin analogue, and meglitinide agents. Of these, only the DPP-4 inhibitors and GLP-1 receptor agonists had a utilization rate in excess of 2%, and the increase in cost between 2008 and 2012 for DPP-4 inhibitors was 44% and for GLP-1 receptor agonists was 75%.
Between 2008 and 2012, approximately 85% to 90% of people naïve to diabetic medications initiated treatment on monotherapy (Table 3). Most commonly, these people with diabetes were started on biguanide monotherapy(63.8% of utilizers during the study period). For biguanides, the utilization rate as monotherapy increased from 55.5% in 2008 to 67.0% in 2012. During the same period, utilization of sulfonylurea monotherapy decreased from 14.2% to 10.8%. Between 9.9% and 14.7% of people naïve to diabetes therapy were started on more than 1 medication over the study period, most commonly on a combination of sulfonylurea and biguanide, or a combination of sulfonylureas, biguanides, and at least 1 additional drug (
Table 3
).
In our Uptake Cohort, we examined the addition of DPP-4 inhibitors or GLP-1 receptor agonists in people who were naïve to the use of those medications in 2011. This is illustrated in
Figure 2
, where, for the sake of clarity, we have scaled our fi gure to represent a hypothetical cohort of 1000 people with T2DM. In 2012, 570 of the 1000 hypothetical cohort members (57%) were on biguanides, 130 (13%) were on sulfonylureas, and 300 (30%) were on both. Less than 1% of those who were naïve to DPP-4 inhibitor or GLP-1 receptor agonist therapy in 2011 were changed to one of these medications as their primary therapy in 2012. Among those on biguanide monotherapy, 5.4% had a DPP-4 inhibitor added to their regimen and 1.6% had a GLP-1receptor agonist added in 2012. For those on sulfonylurea monotherapy, 6.0% had a DPP-4 inhibitor added and 1.2% had a GLP-1 receptor agonist added in 2012. Among the 300 (30%) who were on a combination of biguanide or sulfonylurea in 2012, 28 (9.3% of the subgroup members) had a DPP-4 inhibitor added, 8 (1.9% of the subgroup) had a GLP-1 receptor agonist added, and <1 (0.3% of the subgroup) added both. In this subgroup, there were also 5 members (less than 1.3%) who had their sulfonylurea replaced with DPP-4 inhibitors, GLP-1 receptor agonists, or both. It should be noted that of our 1000 hypothetical cohort members, 914 remained only on biguanides, sulfonylureas, or a combination of those 2 medications—fewer than 95 (or <9.5%) of patients naïve to the use of DPP-4 inhibitors or GLP-1 receptor agonists started therapy on one of those medications during the year.
Impact of New Agents and Pipeline Therapies
In these analyses, we found a prevalence of diabetic medication use of 9.6% among our plan members over the age of 40 years, with an additional 1.2% initiating therapy on diabetic medication each year. At least 70% of these members were on a biguanide, sulfonylurea, or a combination that includes these medications. We estimate that the annual cost of medication for these members was $741 in 2012, a sum that has been increasing at an average rate of 2.7% per year since 2008 despite a reduction in cost of biguanides and sufonylureas.
The new incretin-based therapies—DPP-4 inhibitors and GLP-1 receptor agonists—represent an important option to augment fi rst-line therapies for people with T2DM. Prior to the availability of these medications, first- and second-line treatment options for T2DM were primarily biguanides, sulfonylureas, basal insulin, and thiazolidinediones.5 These new therapies are considered to offer comparable glycemic management with fewer side effects than traditional therapies, including hypoglycemia, weight gain, and edema.6,7 Incretin mimetics have also been reported to improve β-cell function and reduce cardiovascular risk factors.8,9 Recent guidelines from the American Association of Clinical Endocrinologists/ American College of Endocrinology and American Diabetes Association/European Association for the Study of Diabetes support incretin-based therapies as a fundamental treatment option, suggesting that DPP-4 inhibitors and GLP-1 receptor agonists are helpful in treatment of patients with T2DM who are on other glucose-lowering medications and those naïve to therapy. The guidelines also suggest GLP-1 receptor agonists are preferred over the DPP-4 inhibitors because of their superior impact on glycated hemoglobin (A1C) levels.7,10 An additional benefit that has been reported for the GLP-1 receptor agonists is modest weight loss for people with diabetes.7-10
It should be noted, however, that incretin-based therapies have signifi cant shortcomings. GLP-1 receptor agonists require subcutaneous administration, which can be a challenge for elderly diabetics. There is also the concern of a potential link between people being treated with GLP-1 receptor agonists or DPP-4 inhibitors and pancreatitis and pancreatic cancer.8,9,11 It must also be recognized that the enhanced efficacy of the incretin based therapies comes at a higher fi nancial cost. The cost of pharmaceutical treatment of people with diabetes has been rising since 2008 in spite of a decline in the cost of biguanides and sulfonylureas. This has been largely due to the adoption of incretin-based therapies that have a higher cost than these first-line therapies. We estimate that in 2012, the cost of incretin-based medications added $282 per utilizer per year to the cost of treatment for the average person with T2DM. Extrapolating this to the total population of people with T2DM in the United States,this represents an annual burden of $6.6 billion to the US healthcare system. Given the trend toward increased use of these medications, we would expect to see the economic burden increasing by over $2.9 billion annually over the next few years. However, at least a portion of the increased costs for medication may be offset by lower healthcare expenditures for diabetics who benefi t from improved glycemic control and fewer side effects. Pharmaceutical manufacturers have continued to invest heavily in the development of incretin-based therapies. In 2013, 3 new medications were approved for the US market: Nesina (alogliptin), Kazano (alogliptin/metformin), and Oseni (alogliptin/pioglitazone) by Takeda Pharmaceutical.12,13 Alogliptin, an active ingredient in each of these medications, is a new DPP-4 inhibitor.13 Several new incretin-based therapies are also pending US Food and Drug Administration approval or are in phase 3 clinical trials in the United States (
Table 4
).
It is a limitation to our analysis that we excluded members who were on insulin. We recognize that this eliminates from our analysis a substantial subgroup of people. However, we felt it to be very important that we eliminate from our analysis people with type 1 diabetes, or other non—type 2 conditions that would reduce the applicability of our results to people with T2DM. It is likely that this approach led to a very modest underestimation of the absolute prevalence and incidence of diabetic medication use (we suspect that our prevalence estimate is at most 1.5 points below the true prevalence and the incidence estimate is at most 0.2 points below the true incidence). However, comparisons between the medications would be affected only to the extent that insulin use would be associated more with one medication than another. As there is no reason to expect that to be the case, these comparisons are likely to be unbiased.
Incretin-based therapies have proved to be effective in the treatment of people with T2DM who are not responsive to fi rst-line therapy with biguanides and sulfonylureas. As medical therapies for T2DM advance, so do the complexities of recommended treatment options. The role of incretin-based therapies will likely evolve and continue to challenge formulary managers and payers weighing effectiveness and cost in management of people with T2DM.
