CLINICAL DIABETES EDITORIAL First, Do No Harm Irl B. Hirsch, MD, Editor Advances for the treatment of diabetes and its complications evolved in the 1990s unlike in any other decade since the discovery of insulin. It is difficult to prioritize the importance of the various recent advances. The importance of glucose control was unequivocally demonstrated for both type 1 and type 2 diabetes. Effective treatments for nephropathy and cardiovascular disease have improved the lives of millions of patients with diabetes throughout the world. Appropriate screening for complications has been shown to decrease the burden of diabetic retinopathy and lower-extremity amputation. The first insulin analog (insulin lispro, Humalog) was introduced in 1995, and more are soon to follow. However, perhaps the most important new therapy for the treatment of type 2 diabetes was the introduction of the "insulin sensitizers." Although metformin (Glucophage) is not a new drug and for that matter does not represent a new class of oral agents for American patients with diabetes (phenformin, another biguanide, was removed from the market in 1977), its introduction was understandably greeted with much enthusiasm. The excitement was further supported by the results of the United Kingdom Prospective Diabetes Study (UKPDS), which demonstrated that obese patients with newly diagnosed type 2 diabetes randomized to metformin therapy had improvements in diabetes-related end points and all-cause mortality.1 It is therefore not surprising that the introduction of troglitazone (Rezulin) in 1997 was seen by many as a critical advance for the treatment of type 2 diabetes. It was thought that an agent that could treat insulin resistance, one of the primary pathophysiological etiologies of type 2 diabetes, might not only be the ideal therapy for all patients with this disease, but potentially could also be used to prevent type 2 diabetes in high-risk populations. With much fanfare, troglitazone, the first thiazolidinedione (TZD), was brought to the U.S. market in spring of 1997. Initially, it was indicated only for patients with poorly controlled type 2 diabetes who were receiving insulin therapy. Its indications were then broadened to include use as combination therapy with sulfonylureas and as monotherapy. Initial reports also suggested it was beneficial for the treatment of polycystic ovary disease, a state of severe insulin resistance and hyperandrogenism.2 The possibility that this drug could prevent type 2 diabetes was so compelling that it was included in the National Institutes of Health (NIH)-funded Diabetes Prevention Program (DPP). However, it soon became clear that there were problems with this drug, despite that fact that we were initially told it was safe and that "there were few reports of drug-related adverse events. . . and no weight gain . . . ."3 Reports of severe hepatotoxcity created concern for some and alarm for others. In late 1997, liver failure resulting in death or requiring transplantation was reported by the Food and Drug Administration (FDA). The first of several changes to the package insert for liver function monitoring was announced, and the drug was eliminated from study in the DPP after being withdrawn from the British market. Clinicians also noted that severe weight gain, in excess of 20 lb, was seen in some patients. The risk of this weight gain over many years could not be answered without long-term follow-up. Nevertheless, the troglitazone controversy divided the diabetes community unlike anything else had for many years. How could anything with so much potential to be beneficial be a problem? At the very least, many argued that the potential benefits of troglitazone outweighed any risks. The FDA agreed with this philosophy. Liver function monitoring was increased, and the indication for monotherapy was eventually eliminated. After all, when used in combination with sulfonylureas or insulin, the drug was often extremely effective at improving HbA1c levels, and severe hepatotoxicity was quite rare. For monotherapy, there were other drugs with proven track records that could be substituted. By 1999, two new TZDs were introduced. Both in the initial registration studies and in the first 6 months after launch, it became clear that the two new agents were safer than troglitazone. Finally, in March 2000, the American Diabetes Association withdrew its support of troglitazone. One week later, the drug was removed from the market in the United States. Immediately thereafter, it was also removed from the Japanese market. We have learned a lot from the troglitazone experience. First, the
removal of this drug further enforces my opinion that we should always be cautious with
any new therapy until randomized controlled trials (RCTs) prove significant benefit with
"hard end points." After the use of insulin dramatically demonstrated that
people with type 1 diabetes could survive certain death from ketoacidosis, the major
outcome used in all drug studies was the surrogate end point of blood glucose control. By
the 1980s, we had a more sophisticated test for this with HbA1c, but the
important outcomes of blindness, end-stage renal disease, Next, I have learned to be even more skeptical about what I read in newspapers, see on television, and especially what I find on the Internet. It eventually became impossible to differentiate correct information from propaganda, and the motivation for the latter was not always apparent. The behavior of some of my colleagues, both in primary care and endocrinology, was perhaps the most interesting observation. For some, there was a sincere belief that this agent had specific benefits not present with the alternatives, despite the lack of data to support this contention. For others, I wonder how the aggressive marketing campaign influenced how the drug was perscribed. Is it also at least possible that financial incentives to hospitals or physicians may have biased the use of this agent? I know of many physicians who were opposed to troglitazone and, for that matter, the newer thiazolidinediones, not only because of a lack of RCTs showing improved outcomes, but also because of the high cost and aggressive marketing of these agents. So how do we move forward from here? Personally, I will not change the way I use TZDs. It is premature to use these agents for the prevention of type 2 diabetes in high-risk individuals. Although these drugs may eventually be used for this indication, we need RCTs to prove effectiveness and safety. For the treatment of type 2 diabetes, I will not use these agents as monotherapy until we have data to prove benefits on more important end points than HbA1c. Ideally, we need cardiovascular outcomes, such as reduction of myocardial infarction and stroke, but even end points such as an improvement of microalbuminuria and retinopathy would be welcomed. Because our other available agents are cheaper, cost-effectiveness data must also be considered when using any agent as monotherapy. Any new class of agents used for the treatment of type 2 diabetes will also need to be considered differently than in years past, since the UKPDS has shown that our currently available agents are safe and effective. Although unlikely, we don't actually know if the weight gain seen with TZDs will have a detrimental effect after years of use. On the other hand, I use TZDs, and in general, I suspect they are underutilized. I feel there are best used in combination, particularly with insulin therapy. When glycemic control is not achieved with relatively large doses of insulin, it is time to treat the other fundamental problem of type 2 diabetes—insulin resistance. Although metformin can also be used, many patients at this stage have already failed metformin. Because so many patients have poor glycemic control, TZDs would be appropriate for hundreds of thousands, if not millions, of patients in the United States alone. At least for now, I will take my chances using this newest class of agents in combination with other drugs, even without the RCTs to show their efficacy and safety, because of the indisputable proof that hyperglycemia is toxic. As physicians, we must always remember to first, do no harm. REFERENCES 1U.K. Prospective Diabetes Study Group: Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 352:854-65, 1998. 2Dunaif A, Scott D, Finegood D, Quintana B, Whitcomb R: The insulin-sensitizing agent troglitazone improves metabolic and reproductive abnormalities in the polycystic ovary syndrome. J Clin Endocrinol Metab 81:3299-306, 1996. 3Saltiel AR, Olefsky JM: Thiazolidinediones in the treatment of insulin resistance and type II diabetes. Diabetes 45:1661-69, 1996. Copyright © 2000American Diabetes
Association For Technical Issues contact webmaster@diabetes.org |