More Options for the Treatment of Type 2 Diabetes

Irl B. Hirsch, MD, Editor

The late 1990s will be remembered as the time of an explosion in the introduction of new diabetes medications.l

Before the spring of 1995, the pharmacological management of type 2 diabetes was relatively simple. First, we tried diet and exercise, which by itself inevitably failed. We then tried oral agents, and we had several sulfonylureas from which to choose. Eventually, we noted secondary failure to the oral hypoglycemic agent, although it was often difficult to determine whether the drug failed to act or whether the patient was following diet and exercise instructions in a less-than-perfect way. In the end, most patients with long-standing type 2 diabetes required insulin to control hyperglycemia, since there were no further choices for pharmacological therapy.

Used correctly, insulin can be quite effective,^1,2 although, as Dr. Claresa Levetan points out in this issue of Clinical Diabetes (p. 94), obtaining the best possible results from insulin therapy requires more than just a prescription.

Diabetes treatment in the United States dramatically changed with the introduction of the biguanide metformin (Glucophage®) in the spring of 1995. We now had a new agent (for us, anyway), giving us another tool for improving glycemic control in our patients. Soon afterwards, the a-glucosidase inhibitor acarbose (Precose®), and later the thiazolidinedione troglitazone (Rezulin®) were introduced.

This spring, the benzoic acid derivative repaglinide (Prandin™) will join the other diabetes drugs on the American market. (A review of repaglinide by Dr. Wayman Wendell Cheatham appears in this issue on page 70.)

In 1997, the per-capita cost of outpatient pharmaceutical therapy for Americans with diabetes was $736. This is 3.2 times higher than that of Americans without diabetes,³ although one must appreciate that this also included the costs of medications to treat diabetes-related complications and other co-morbidities. With better treatment of hyperglycemia, we eventually could see a decrease in the costs of diabetes, in terms of both money and human suffering.

Including insulin, we now have six classes of drugs from which to choose for the treatment of type 2 diabetes. This is a 200% increase in the past 3 years. What is more amazing is that the pharmaceutical industry continues to develop new agents within existing drug classes and even more classes of drugs for diabetes.

This is a good time to reflect about the implications of these developments. First, what is the role of each drug? Which patients should be using each drug, either alone or in combination? How can we identify the niche for each agent in individual patients?

Are there unknown potential side effects with these new drugs? Certainly, the reported hepatotoxicity with troglitazone⁴ came as a surprise to many, since it was described after the drug received initial approval from the Food and Drug Administration.

Should we base our choice of drugs on costs alone? Should the presence of a dyslipidemia, hypertension, heart disease, diabetic nephropathy, or obesity play a role in the choices we make from our new and growing arsenal?

All of these factors should be taken into account. In this issue, Dr. John R. White, Jr., describes the available agents in a concise, scholarly, and nonbiased view (p. 58). Many of us are confused by all of these new choices and would prefer a "cookbook" approach to their use. But clearly, there is no consensus about which drugs should be used under which circumstances.

One concern is that clinicians will use combinations of drugs that have not been tested. For example, is it at least possible that there could be more hepatotoxicity with the combination of metformin and troglitazone? Until proven otherwise, I would avoid this combination. Yet we know that others are using it already.

I also avoid the use of triple therapy, since, to my knowledge, there have been no studies of three drugs from different classes. Theoretically, agents with three different mechanisms of action might be very useful. But one never knows what dangers lurk in certain drug interactions.

Ideally, we need randomized, clinical trials to guide us in identifying the proper role of each agent, both alone and in combination. As of this writing, no such trials exist, although we hope to gain some insight from the United Kingdom Prospective Diabetes Study (UKPDS). This large trial has shown that, as initial therapy, sulfonylureas and metformin are as effective as insulin and that monotherapy has a 60% failure rate in 6 years.⁵ Unfortunately, this information is not at all surprising. We await UKPDS results related to the study’s microvascular and macrovascular endpoints.

Finally, one could take a more strict view about our current choices of agents and conclude that we should rely only on randomized, clinical trials with clinically meaningful endpoints to determine which agents we use for the treatment of type 2 diabetes. One could consider HbA_1c levels a surrogate endpoint, since it is obviously not a direct measurement of diabetes-related complications. Limiting ourselves for a moment only to the trials observing these more important outcomes, the data suggest that the best drug is insulin.

The Kumomoto study, which assessed microvascular disease and neuropathy,² and the DIGAMI study, which observed cardiovascular events and mortality after acute myocardial infarction,⁶ both showed positive outcomes with the use of insulin therapy. Furthermore, the University Group Diabetes Program actually showed an increase in cardiovascular events with the sulfonylurea tolbutamide and the biguanide phenformin, although conclusions were extremely controversial.⁷ There are now plausible mechanisms for why sulfonylureas may be dangerous in some patients with coronary artery disease.⁸ Indeed, in the DIGAMI trial, the only group with a benefit from the insulin therapy were those patients who were taking oral agents before their infarction and were at low risk for cardiac events.

More direct clinical data are necessary. For now, my recommendation is that all factors must be considered when deciding which agent is best. The primary goal should be maintaining target HbA_1c levels, although there may be exceptions for some patients.⁹ Hopefully, the confusion about how best to use the many diabetes agents will resolve in the near future.

References

¹Genuth S: Insulin use in NIDDM. Diabetes Care 13:1240-64, 1990.

²Ohkubo Y, Kishikawa H, Araki E, Miyata T, Isami S, Motoyoshi S, Kojima Y, Furuyoshi N, Shichiri M: Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized, prospective 6-year study. Diabetes Res Clin Pract 28:103-17, 1995.

³American Diabetes Association: Report: Economic consequences of diabetes mellitus in the U.S. in 1997. Diabetes Care 21:296-309, 1998.

⁴Rezulin package insert, Parke-Davis, Morris Plains, N.J.

⁵United Kingdom Prospective Diabetes Study Group: United Kingdom Prospective Diabetes Study 24: a 6-year, randomized controlled trial comparing sulfonylurea, insulin, and metformin therapy in patients with newly diagnosed type 2 diabetes that could not be controlled with diet therapy. Ann Intern Med 128:165-75, 1998.

⁶Malmberg K: Prospective, randomized study of intensive insulin treatment on long-term survival after acute myocardial infarction in patients with diabetes mellitus. DIGAMI (Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction) Study. Br Med J 314:1512-15, 1997.

⁷Sackler AM: The unsettling UGDP controversy. JAMA 243:1435-36, 1980.

⁸Engler RL, Yellon DM: Sulfonylurea K-ATP blockade in type II diabetes and preconditioning in cardiovascular disease: time for reconsideration. Circulation 94:2297-2301,1996.

⁹Skyler JS: Glucose control in type 2 diabetes. Ann Intern Med 837-38, 1997.