Combination Therapy in Type 2 Diabetes John Buse, MD

Diabetes management in the 21st century has benefited greatly from the availability of new drugs and technologies, such as innovative modalities for glucose monitoring and education that have been added to the formulary since 1990. These have enabled physicians to help their patients with diabetes achieve the recommended hemoglobin A1C targets that are important for reducing the risk of complications. Although the societies that propose optimal glucose targets differ slightly in their recommendations, the goals are identical: optimal glucose control for patients with diabetes. Table 1 illustrates recommended glucose levels as determined by the American Diabetes Association (ADA) and the American College of Endocrinology (ACE).

Table 1. Glycemic Goals of Therapy^1,2

Goal	ADA	ACE
Fasting plasma glucose (FPG)	90–130 mg/dL	<110
2-hour postprandial plasma glucose (PPG)	<180	<140
A1C	<7%	<6.5%

Experience in clinical practice suggests that the ACE recommendations for glucose levels are impossible to achieve consistently in a substantial percentage, perhaps even a majority of patients with diabetes. The best solution is for patients and physicians, as a team, to aim for the lowest levels possible without the occurrence of unacceptable adverse events (AEs) such as hypoglycemia.

Pathophysiology

Each class of drugs currently available deals with one part of the pathophysiology of the disorder. Because of these mechanistic differences, combination therapy allows patients to achieve levels of glycemic control that are impossible with single agents.

In normal people, blood glucose in the fasting state is maintained in the range of 60 to 90 mg/dL. This results from a balance of hepatic glucose production with glucose utilization, predominantly by the brain. Most other tissues in the body primarily rely on free fatty acids as an energy source in the fasting state, although there is some glucose uptake and utilization as well. After eating, carbohydrate molecules, protein, and fat are absorbed from the intestinal tract; as glucose levels rise above 60 mg/dL, there is a progressive recruitment of ß cells within the pancreas to secrete insulin. These insulin molecules pass to the liver via the portal vein, where their primary role is to suppress the hepatic glucose production. The insulin molecules that pass through the liver circulate throughout the body, stimulating the uptake of glucose in the insulin-responsive tissues, muscle, and fat. Effectively, what occurs in the postprandial state is that the body "shunts" carbohydrates from the intestinal tract into muscle and fat for storage. The brain's utilization of glucose does not increase or decrease with meals; the liver is the ultimate provider of food for the brain during the fasting state.

Type 2 diabetes is characterized by a decrease in insulin sensitivity in the liver, muscle, and adipose tissues, and impaired ß-cell function.³ Together, these lead to the major metabolic defects responsible for type 2 diabetes, including⁴:

• Glucose absorption is increased—patients tend to overeat; enzymes and transporters involved in carbohydrate processing are upregulated; and, in general, the stomach empties faster
• Hepatic glucose overproduction—the liver's production of glucose is not shut off in the face of hyperglycemia or in response to meals
• Decreased glucose clearance into muscle and fat—glucose uptake is not stimulated which results in a decrease in glucose utilization and storage in muscle and fat
• The pancreas fails in its ability to produce insulin progressively; this synergizes with insulin resistance in the liver, muscle, and fat to produce further hepatic glucose overproduction and further decreases in glucose clearance from the circulation

These multiple pathophysiologies provide a number of different targets for treatment (Figure 1).

Sulfonylureas and the newer meglitinide drugs stimulate insulin production by the pancreas; subcutaneous insulin is a supplement to endogenous production. With these approaches, we now have the capacity to essentially normalize insulin delivery.

Figure 1. Treatment of type 2 diabetes.

• When combined with agents that work in the liver, such as metformin, clinicians can come close to normalizing hepatic glucose production
• When combined with approaches that work in muscle and fat tissues, such as exercise and the glitazones, we can essentially normalize glucose utilization
• When combined with physical activity, the restriction of calories, or the use of
  α-glucosidase inhibitors, we can normalize the flux of glucose from the intestinal tract

In summary, to a large extent we can now deal with multiple pathophysiologic defects simultaneously in patients with type 2 diabetes.

Therapeutic agents

Insulin Secretagogues
The agents that are available today can be divided into 2 categories: those that augment the supply of insulin, and those that enhance insulin's effectiveness.³

The oldest agents used to treat type 2 diabetes include the sulfonylureas, which stimulate increased pancreatic insulin secretion.⁵ The most common side effect of these agents, including glipizide, glyburide and glimepiride is hypoglycemia. Other AEs are uncommon, and include nausea, vomiting, and skin reactions.³ All the drugs in this class reduce A1C levels by 1% to 2%.³

Biguanides: Metformin
Metformin's main action is to reduce hepatic glucose production with a net result of decreasing A1C by 1% to 2%.³ In approximately one-third of patients, AE's include diarrhea and nausea, though the majority of patients tolerate it adequately, particularly if it is started at about 500 mg/day and titrated slowly to the maximum effective dose of 2000 mg/day. The major concern regarding metformin is the rare association with lactic acidosis. By following the guidelines in the prescribing information and avoiding its use in patients with renal insufficiency and heart or liver failure, the risk of lactic acidosis can be virtually eliminated. New, sustained-release formulations are associated with less nausea, which allows greater tolerability in patients for whom nausea is the chief AE. There are also a number of combination formulations with rosiglitazone, glyburide, and glipizide.

Enthusiasm regarding the role of metformin in diabetes care was magnified in 1998 with the publication of the United Kingdom Prospective Diabetes Study (UKPDS). Figure 2 demonstrates data from the overweight subgroup, which was randomized either to conventional therapy with diet and exercise, insulin or sulfonylureas, or to metformin therapy. Although metformin was not associated with greater improvement in glycemic control, there was a statistically significant reduction in cardiovascular events compared to patients who were maintained on diet and exercise alone.⁶ Metformin was also associated with less weight gain and hypoglycemia than insulin or sulfonylureas. Arguably, because of this superior performance in the UKPDS, metformin should be the foundation of therapy for type 2 diabetes, at least in those who tolerate the medication, do not have contraindications, and who can adhere to twice-daily dosing.

Figure 2. Diabetes-related deaths: UKPDS overweight subgroup.

Thiazolidinediones: Pioglitazone, Rosiglitazone
The first drug in this class, troglitazone, was withdrawn from clinical use in March 2000, as a result of rare but severe drug-related liver toxicity.⁷ This class of drug acts primarily to enhance insulin sensitivity in adipose tissue, and secondarily, on muscle. It also reduces hepatic glucose production and seems to stabilize ß-cell dysfunction.⁷ The newer glitazones, pioglitazone and rosiglitazone, produce A1C reductions of 1% to 2%,³ either as monotherapy or in combination. The major AEs are related to anemia, weight gain, and fluid retention, which presents as congestive heart failure (CHF) in some patients. Liver-function test monitoring is recommended at baseline, every 2 months for the first year, and intermittently thereafter, and at least in this context, there seems to be no issue with liver safety for these newer compounds. Because these drugs work slowly, titration frequency should be no more than every month, with maximal results from a particular dose being seen in about 6 months.

Drugs in this class have a unique mechanism of action mediated by changing activity of a family of nuclear receptors called "peroxisome proliferator-activated receptor," or PPAR.⁸ They seem to exert their effects predominantly to lower glucose and modify lipid levels by changing fat cell metabolism. They are also associated with a decrease in hepatic glucose production and an increase in insulin sensitivity in muscle, which results in improved glucose uptake.⁸

Although there are some differences between pioglitazone and rosiglitazone, they are more or less equivalent in their glucose-lowering efficacy, either in mono- or combination therapy.

There are also data with both agents to suggest that they may have an effect of stabilizing β-cell dysfunction, with the possibility that they may reduce or halt the progressive nature of glycemic dysregulation in type 2 diabetes.

Figure 3. Insulin resistance: cardiovascular correlates.9

Insulin resistance is a pathogenic factor in the development of a broad spectrum of clinical conditions other than type 2 diabetes (Figure 3), including⁹:

• Hypertension
• Dyslipidemia
• Atherosclerosis
• Central obesity
• Endothelial dysfunction
• Coagulation/fibrinolytic defects
• Oxidation/inflammation

A variety of studies have demonstrated that the glitazone drugs, perhaps through their effect of improving insulin sensitivity, are associated with: improvement in glycemic control, reduction in blood pressure, reduction of visceral fat (despite increasing total fat mass), an increase in high-density lipoprotein (HDL), a reduction in small-dense low-density lipoprotein (LDL) particles, improvement of endothelial function, partial reversal of the procoagulant state, and reduction of CRP and other markers of inflammation.

These findings provide substantial promise that techniques aimed at reversing components of the insulin resistance syndrome will reduce the burden of CVD in patients with type 2 diabetes.

In initiating glitazone therapy, it is important that patients understand the potential for AEs.

• They need to understand the safety afforded by having ALT levels monitored regularly
• They should be informed about the possibility of edema and weight gain, and agree to adhere to a prospective plan to evaluate and manage these potential effects
• They should be taught how to check for edema, and to restrict their sodium intake if edema develops
• They should be instructed to call or come in for evaluation if edema is progressive, in order to consider modification of the treatment plan or the use of diuretic therapy

The risk of edema can be minimized by patient selection, avoiding use in patients with Class III or Class IV heart failure and by starting with a low dose in those treated with insulin or with pre-existing edema.

α-glucosidase inhibitors: acarbose and miglitol
Acarbose was the first α-glucosidase inhibitor in clinical use, followed by miglitol. These drugs delay absorption of carbohydrates by blunting the rise in postprandial glucose. They are associated with flatulence, abdominal discomfort, and diarrhea in up to 50% of patients. These AEs can be minimized by starting with a small dose once daily followed by slow titration.

Results from the STOP-NIDDM trial¹² reported in The Lancet in 2002 showed that patients randomized to acarbose who had impaired glucose tolerance (IGT) had approximately a 25% reduction in their risk of developing diabetes (32% vs 41%). Although these are very preliminary results, the patients in the acarbose arm also had a reduction in the rate of developing new cases of hypertension and myocardial infarction, or any cardiovascular event.¹² These results indicate that the α-glucosidase inhibitors deserve attention as effective agents with potential CVD benefits and only nuisance side effects.

Insulin secretagogues: sulfonylureas and glinides
The sulfonylureas have been the backbone of therapy for decades. The newer agents, nateglinide and repaglinide, though nonsulfonylurea compounds, also exert their glucose lowering action after binding to the sulfonylurea receptor by increasing insulin secretion.

There are differences in the nature of the interaction with a sulfonylurea receptor among some agents relating to the precise nature of the interaction. Figure 4 demonstrates the effects of repaglinide and nateglinide on insulin release in an isolated perfused organ system. Repaglinide rapidly stimulates insulin secretion but has a tail in its action with measurable insulin output for at least 20 minutes after the drug is washed out of the system. Nateglinide has a similar rapid increase in insulin secretion, and a more rapid decrease when withdrawn. As a result, nateglinide works fairly exclusively in the postprandial state and is associated with very low risk of hypoglycemia. Repaglinide appears to have efficacy in reducing glucose for a period of time longer than one would predict based on its pharmacologic half-life of approximately 1-2 hours, perhaps explaining how it seems to have substantial impact to normalize fasting glucose.

Figure 4. Nateglinide-stimulated insulin secretion: fast on and fast off.13

Sulfonylureas and ischemic preconditioning
Ischemic preconditioning is a protective mechanism in the heart. In a European study reported in 1999,¹⁴ patients with diabetes underwent balloon angioplasty, which was performed with recording of ST segment elevation—an index of heart injury. The balloon was inflated, ST-segment measurements were taken, and the balloon subsequently deflated. The ischemic preconditioning mechanism normally would result in a reduction in the amount of ST elevation with successive episodes of ischemia. Figure 5 illustrates the results of the double-blind, placebo-controlled evaluation of glimepiride and glyburide and their effects on cardiac response to ischemia. In the patients with diabetes treated with placebo and with glimepiride, ST segment elevations were reduced. However, patients in the glyburide arm lose this ischemic preconditioning. The sulfonylurea receptors are present in the heart and vascular smooth muscle; various agents in this class may have differential effects in these tissues.

Figure 5. Ischemic preconditioning: are all sulfonylureas the same?

The risk of hypoglycemia does not seem to be exclusively related to the half-life of a drug, but to the residence time on the sulfonylurea receptor and drug metabolism. Glyburide and chlorpropamide seem to be associated with a higher risk of hypoglycemia than glipizide, glimepride, repaglinide, and nateglinide, and as a result should probably be avoided in an era where more stringent glycemic targets are being aggressively pursued.

Insulin Therapy

Diabetes therapy should address both postprandial and basal requirements. Figure 6 illustrates the normal physiologic response of glucose and insulin to meals highlighting the need for both basal and meal-time insulin. Meal insulin release occurs in response to nutrient ingestion; basal insulin is continuously secreted over a 24-hour period to maintain fasting and interprandial glucose levels. In the past, clinicians and diabetes educators had to make do with various insulin formulations that did not have adequate pharmacokinetics to duplicate the profiles illustrated in Figure 6. However, within the past few years, new insulin analogs that provide more physiologic profiles have been developed.

Figure 6. Normal glucose physiology: 24-hour profile

When a patient with type 2 diabetes requires insulin in addition to an oral antihyperglycemic agent, a simple and effective method for making this transition is as follows³:

• The oral agent should be continued at the same dose
• Long-acting insulin (NPH or glargine) should be given as a single dose, generally starting at approximately 10 U, in the evening
• The insulin dose is adjusted according to the results of self-monitoring of FPG to achieve target levels (generally 90 mg/dL to 130 mg/dL)
• The dose may be increased by 4 units weekly, if FPG is >180 mg/dL, or by 2 units weekly if FPG is >140 mg/dL

Successful control with less nocturnal hypoglycemia
A study that was presented at the 2002 American Diabetes Association meeting included 700 patients who had failed sulfonylurea, metformin, or sulfonylurea and metformin combination therapy and had A1C levels greater than 7.5%. Over the course of 24 weeks, patients were randomized either to glargine insulin or NPH insulin at bedtime. (Figure 7) There was somewhat less nocturnal hypoglycemia associated with the glargine arm of the trial, and 57% of both treatment groups reached A1C levels of ≤7.¹⁵

Figure 7. Treatment to Target study: insulin glargine vs NPH insulin added to oral therapy of type 2 diabetes.15

Rapid-acting insulin analogs
The rapid-acting insulin analogs, lispro and aspart, are similar in action (Figure 8). The major decision about which one to prescribe is perhaps best based on the delivery system: which insulin pen system is most convenient.

Figure 8. Are lispro and aspart different?

Intensive management strategy
An accepted strategy for intensive management is combination therapy, as illustrated in Figures 9 and 10.

Figure 9. Intensive management strategy.

Figure 10. Treatment algorithm.

References

1. American Diabetes Association. Position Statement. Standards of medical care for patients with diabetes mellitus. Diabetes Care. 2003;26(suppl):S33-S50.
2. American College of Endocrinology Consensus Statement on Guidelines for Glycemic Control. Endocrine Practice. 2002;8(suppl):5-11.
3. Medical Management of Type 2 Diabetes. American Diabetes Association. 1998. Alexandria, Va.
4. Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest. 1999;104:787-794.
5. Sonnenberg GE, Kotchen TA. New therapeutic approaches to reversing insulin resistance. Curr Opin Nephrol Hypertens. 1998;7:551-555.
6. United Kingdom Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-853.
7. Boyle PJ, King AB, Olansky L, et al. Effects of pioglitazone and rosiglitazone on blood lipid levels and glycemic control in patients with type 2 diabetes mellitus: a retrospective review of randomly selected medical records. Clin Ther. 2002;24:378-396.
8. Camp HS, Li O, Wise SC, et al. Differential activation of peroxisome proliferator-activated receptor-gamma by troglitazone and rosiglitazone. Diabetes Care. 2000;49:539-547.
9. American Diabetes Association. Consensus development conference on insulin resistance. Diabetes Care. 1998;21:310-314.
10. Gegick CG, Altheimer MD. Comparison of effects of thiazolidinediones on cardiovascular risk factors: observations from a clinical practice. Endocrine Practice. 2001;7:162-169.
11. Lister CA, Boam D, Bretherton-Watt D, et al. Rosiglitazone increases pancreatic islet area, number and insulin content, but not insulin gene expression. Diabetologia. 1998;41(suppl 1):A169.
12. Chiasson JL, Josse RG, Gomis R, et al. Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomised trial. Lancet. 2002:359:2072-2077.
13. Leclercq-Meyer V, Ladriere L, Fuhlendorff J, Malaisse WJ. Stimulation of insulin and somatostatin release by two meglitinide analogs. Endocrine. 1997;7:311-317.
14. Klepzig H, Kobert G, Matter C, et al. Sulfonylureas and ischaemic preconditioning; a double-blind, placebo-controlled evaluation of glimepiride and glibenclamide. Eur Heart J. 1999;20:439-446.
15. Riddle M, Rosenstock J. Treatment to target study: insulin glargine vs NPH insulin added to oral therapy of type 2 diabetes. Successful control with less nocturnal hypoglycemia. Diabetes. 2002;51(suppl 2):A113. Abstract 437-P.