Trials demonstrate clot-busting superiority
Not just faster, but better results
The U.S. Food and Drug Administration (FDA) approved reteplase late in 1996 for management of acute myocardial infarction (AMI) in adults, including the dissolution of blood clots obstructing coronary arteries, reduction of the incidence of congestive heart failure, improvement in ventricular function post-AMI, and reduction of mortality associated with AMI.Reteplase competes directly with Genentech Inc.’s tissue plasminogen activator (t-PA, or alteplase, marketed as Activase) and with streptokinase. T-PA and reteplase at about $2,200 to $2,600 per dose cost 10 times more than streptokinase. Reteplase is the first new thrombolytic agent approved solely on the basis of comparative clinical trials rather than the standard placebo-controlled trials. That’s because, with effective agents such as t-PA and streptokinase available, it’s no longer ethical to administer placebos to heart attack patients. As a result, the manufacturer conducted two open-label Phase III trials comparing reteplase to t-PA and one larger double-blinded Phase III study comparing reteplase to streptokinase.
RAPID I and II compare drug to t-PA
The two reteplase vs. t-PA studies are known as RAPID I and RAPID II (Reteplase and Alteplase Patency Investigation During Myocardial Infarction) and enrolled 606 and 324 patients, respectively. These trials used arterial patency at 90 minutes as the primary endpoint. RAPID I demonstrated that reteplase caused a statistically significant increase in the number of patients who achieved complete patency as compared to t-PA and RAPID II reinforced that finding.Although it’s an important measure of a drug’s performance, patency alone is no longer considered adequate for approval of thrombolytics — mortality is the gold standard. Thus, Boehringer Mannheim conducted the INJECT study (International Joint Efficacy Comparison of Thrombolytics) with the primary endpoint of mortality. However, for this 6,010-patient trial conducted entirely in Europe at 208 centers, the company chose to compare reteplase with streptokinase, not t-PA as in the RAPID trials. Company officials claim that comparison was selected because streptokinase is the thrombolytic of choice in Europe as opposed to the United States where t-PA captures 75% of the market.
INJECT compared mortality rates between patients treated with streptokinase and reteplase. At 35 days, the reteplase group had a 9% mortality rate as compared with a 9.5% rate in the streptokinase group. The difference of a half a percentage point was not significant. However, because INJECT was a comparative trial, the finding met the criteria for approval as set out by the FDA.
Reteplase did significantly reduce congestive heart failure, a prospectively defined secondary endpoint of the INJECT trial, as compared to streptokinase, but it also caused a statistically significant increase in the rate of hemorrhagic strokes.
INJECT proved only 50% of mortality benefit
Comparative trials can be difficult to design and interpret because they seek to establish equivalence rather than superiority — the aim of most placebo-controlled trials (i.e., the experimental drug is better than sugar water). Alastair Wood, a professor of medicine and pharmacology at Vanderbilt University in Nashville, TN, and an FDA panel member, says that statistical analyses deemed critical in placebo-controlled trials are "turned on their head" when applied to equivalence trials. The results can be confusing.For example, the INJECT study was designed to prove that reteplase retains at least 50% of the mortality benefit of streptokinase. That doesn’t sound like a stunning clinical victory to most people. In fact, according to the estimate of one FDA reviewer, the data suggest that reteplase retains about 65% of the efficacy of streptokinase. Panel members criticized 50% as too low. Wood argues that new thrombolytic agents should have to prove that they retain at least 75% of the efficacy of comparative agents.
But a higher comparative efficacy threshold would dictate that clinical trials be four or five times as large as the 6,000-patient INJECT study, already huge by ordinary product license application standards.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content