Bucindolol trial yields neutral and negative outcomes for CHF patients
Treat blacks with beta-blockers until we have more evidence’
Surprising new findings revealed at the American Heart Association meeting in Atlanta last fall from the BEST trial include both neutral and negative outcomes that the beta-blocker bucindolol has had for some CHF patients. When the drug was administered to patients with advanced CHF, no benefit was seen, and subgroup analyses revealed that blacks, 23% of the study population, in general did not benefit and may even have been harmed by the drug.
BEST (Beta-blocker Evaluation of Survival Trial), sponsored by the National Heart, Lung and Blood Institute (NHLBI) and the Department of Veterans Affairs, randomized nearly 3,000 mainly NYHA Class III and IV heart failure patients on standard therapy to the nonselective beta-blocker bucindolol or placebo and followed the patients for a mean of two years.
Unlike previous beta-blocker studies, BEST placed a significant emphasis on the enrollment of women. While women over age 60 comprise approximately half of the 3 million Americans who suffer from heart failure, they typically comprise less than a fifth of the patients enrolled in heart failure studies. One of the goals of BEST was to achieve an enrollment of at least one-third women in order to evaluate the effectiveness of beta blockade in both men and women with CHF.
Although findings showed that bucindolol (marketed by Intercardia of Research Triangle Park, NC) reduced deaths by 10% overall, those data were not statistically significant; patients with more advanced disease in particular did not benefit.
Carvedilol and metoprolol studies have had clearly positive results. What went wrong with BEST? Discussion at the meeting focused on several issues:
1. Most of the other beta-blocker trials focused on Class II patients, with some Class IIIs and just a few Class IVs. "Carvedilol is not recommended in Class IV or unstable patients even with lesser symptoms," wrote Michael H. Crawford, MD, in the December issue of Clinical Cardiology Alert.
2. Bucindolol is nonselective like carvedilol, but bucindolol does not have the alpha-blocking properties that carvedilol has.
3. The relative lack of effect in blacks who comprised about a quarter of the patients may have skewed the results of an otherwise positive trial. The Scandinavian metoprolol trials and the carvedilol studies done in Southeast Asia included few black patients.
A medical mystery
"But the answer is not yet clear," says Josh Hare, MD, assistant professor of medicine and associate director of the heart transplant program at Johns Hopkins Hospital in Baltimore. "This is one of the puzzles in medicine that we have to confront from time to time. I don’t think anyone knows the answer. All we can do is look at the results and put them in the context of other studies we’ve done and try to come up with a uniform picture." He says the answer could be something completely intangible and adds that investigators can go on for decades developing drugs for specific reasons, then only much later come up with new discoveries that show that the reasons they were using them were wrong all along.
"We think these [beta-blocker] drugs work by blocking the beta receptor, and we know there are different beta receptors — 1, 2, and 3," says Hare. "And we also know there are alpha receptors." Drugs such as metoprolol are selective for the beta 1 receptor. Bucindolol is nonselective, but there’s another nonselective drug, carvedilol, that is known to work. The reason for the BEST results may have to do with something investigators don’t know right now. We have to have a level of humility as scientific researchers," he says. "We have to accept the fact that there are some things we don’t understand. Also, these studies are imperfect."
Hare says statistical chance is a given, and even with studies that are done on thousands of patients and cost millions of dollars, it’s still possible to come up with an incorrect answer. "It may be as simple as that the investigators and the patients enrolled in this particular trial were unlucky and that the results they came up with were incorrect by chance. If the study were to be repeated in its exact form, it’s conceivable that the answers would be different."
He says in many other areas in which multiple studies have been done, every once in a while a study will come up with an opposite and different answer from the body of studies. "We have to examine the totality of evidence, therefore, from all the different trials, and the totality of evidence about beta-blockers is that they work. This one may be just a fluke."
Subgroup analysis and unexpected results
Is there something about the black patient that makes a difference? "This is the only study to my knowledge to date that shows such a difference, and I would not want to conclude on the basis of these data that blacks don’t respond to beta-blockers," Hare says. He adds that the BEST trial was not designed to look at differences between races.
"History is replete with examples where you come up with some kind of observation from a subgroup analysis to try and help you give yourself an answer to an unexpected result," he says. "And when those observations are put to the test in subsequent trials, they have been proven wrong."
Hare says if the medical community feels that there is a difference between blacks and Caucasians, the only way to get a definitive answer is to test that concept prospectively in another study.
"Based on the history of clinical trials, it would not come as a great surprise if the subgroup analysis from this study was not borne out in a subsequent study. The results may even be reversed." Hare says it is concerning to have an outcome like this, because it may lead to anxiety on the parts of both patients and physicians as to what to do. "For now, you do nothing on the basis of the BEST trial. You continue to treat blacks with beta-blockers until such time as there is true evidence that they don’t benefit from it."
Eric Eichhorn, MD, BEST study co-chairman and director of the cardiac cath lab at the Dallas VA Medical Center, says bucindolol was chosen for the BEST study because at the time the trial began, there were only two beta-blockers on which any data existed. Do the drug’s properties explain the results of the trial? The subanalysis results suggest that bucindolol was not beneficial and may even have had an adverse effect. "The drug itself does not explain that though, because if it were the drug, I would have expected the same effect in Caucasian patients as we got in black patients," he says.
Do blacks respond differently to beta-blockers than Caucasians? "Maybe," Eichhorn says. "We don’t know for sure. This was a subgroup analysis, and whenever you do a subgroup analysis, you find something by chance. Would we have had a better response with another agent such as carvedilol or metoprolol? Maybe. Unless we did a head-to-head comparison, we won’t know."
Eichhorn says the results are in part explained by the fact that the study group was a much sicker group of patients than in the other major trials, Metroprolol CR/XL Randomized Intervention Trial (MERIT) and Cardiac Insufficiency Bisopro-lol Study (CIBIS). MERIT included 5% blacks, and CIBIS included less than 1%.
"Our placebo mortalities were higher," he says. "We were pushing beta-blockers to the limits. By the time patients get that sick, there’s nothing left to salvage. It’s often too late."
"There was a quirky response in blacks," Eichhorn says. "If we were to go to a less sick population of blacks, my guess would be that they would respond to the drug." He says there are unpublished data that support the fact that blacks benefit from this type of therapy, but there are so few blacks in the other major trials, with such little follow-up, the confidence intervals for making any sort of mortality claim about that subgroup is impossible. "There are not enough data out there on blacks." Eichhorn says they plan to go back and look at their data to see if there’s something that explains the outcome.
"The data at least raise the suspicion that blacks with advanced CHF may not respond as well to bucindolol as Caucasians, but we don’t know for sure. The bottom line is that this study is hypothesis-generating, but not conclusive.
"What’s interesting is that the beneficial response we got was in the same type of patients that were studied by MERIT and CIBIS — male Caucasians," Eichhorn says. We need to re-examine this issue." He says that in practice he has found his black patients with mild to moderate CHF benefit when taking beta-blockers. "If they show good response in ejection fraction and symptoms, I tend to leave them on that therapy."