ALLHAT revisited

Susan Jeffrey

March 03, 2003

Mon, 03 Mar 2003 20:45:00

Houston, TX - Immediately upon their release in December last year, results from the hypertension arm of the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) were widely publicized in the consumer press. The notion that an old, cheap drug could come out looking better overall than its newer, more expensive counterparts was appealing to many and good news from the point of view of those paying for antihypertensive medications.

However, while few within the cardiology community would quarrel with the assertion that using a diuretic to treat hypertension is a good ideagiven that hypertension treatment guidelines have always said to start with themwhat to make of other findings, particularly the less-than-stellar showing of the ACE inhibitor in that study against all expectations, is less clear. In the weeks after release of the ALLHAT data, heartwire revisits the trial to get some sense of how the findings are being received.

ALLHAT Army on the march

Dr Curt D Furberg

Dr Curt D Furberg (Wake Forest University-Baptist Medical Center, Winston-Salem, NC) chaired the Steering Committee for ALLHAT. "I think we've been very pleased with the coverage we've had from the media, and we are just developing plans for how we can monitor changes in prescriptions."

However, they expect it is too soon at this point to see any change. After the doxazosin arm of ALLHAT was stopped two years ago because the alpha blocker was found to be less effective than chlorthalidone, it took a year and half for any change to show in prescribing habits, Furberg noted. "But again, it's not clear whether it was because of the results that we found or because doxazosin went off patent. But if you add up the generic doxazosin and the branded Cardura® (Pfizer), the use of doxazosin is down quite a bit, a quarter, maybe a third."


We've been very pleased with the coverage we've had from the media.


"It's not going to happen overnight," he said. "At least there have been changes in knowledge, and now we're waiting for the behaviors to change."

Dr Barry Davis (Source: University of Texas-Houston)

To help change those behaviors, dissemination plans are "in the works, although they're not as well formulated as we might like them to be," says Dr Barry Davis (University of Texas-Houston), director and principal investigator for the ALLHAT Clinical Trials Center. In general, though, "they would consist of trying to bring the results to doctors in the field in various ways and to try to do some patient activism, based on hypertension in general and the results of this trial in particular."

This latter avenue is hampered somewhat by the fact there is "no great patient activist network for hypertension," he noted. As far as doctors are concerned, though, they plan to present further results from ALLHAT at most of the major meetings over the next year and still have their "ALLHAT Army," more than 600 investigators, many of whom have volunteered to help with dissemination activities, he said.

Responding to the questions

Both Furberg and Davis acknowledged they have been getting some questions about their results from doctors in the community. In an upcoming issue of the Journal of the American Medical Association, for example, ALLHAT investigators respond to correspondence sent to the journal after publication of both the hypertension and lipid-lowering arms of the trial.


People have good questions about various aspects of ALLHAT, and we've been trying to respond to them.


"People have raised questionsprimarily people who are close to the pharmaceutical industry," Furberg said. Davis conceded that "people have good questions about various aspects of ALLHAT, and we've been trying to respond to them. We have other plans in the works for presenting further analyses of the data to address various questions."

One analysis that has already been done is a heart failure validation study, perhaps in response to some speculation raised about the diuretic masking cases of heart failure that were seen more commonly with both the ACE inhibitor and calcium channel blocker. They expect to present this study at the upcoming American College of Cardiology meeting in Chicago, he said.

Chlorthalidone now diuretic of choice?

Dr Eric J Topol

In a recent Thumbs up, thumbs down program for, Drs Robert M Califf (Duke University Medical Center, Durham, NC) and Eric J Topol (Cleveland Clinic Foundation, OH) discussed some of the more complex issues surrounding ALLHAT. Both Califf, who was chair of the trial's Data and Safety Monitoring Board, and Topol said they have changed their practice based on ALLHAT findings inasmuch as they have adopted use of the diuretic studied in this trial, chlorthalidone, over the more widely used agent hydrochlorothiazide.

"We're using [chlorthalidone] specifically here [at Cleveland Clinic] as a result of the trial, so it's had an impact," Topol said. "The doses of course are less than hydrochlorothiazide, with just 12.5 mg instead of 25 mg to start with, and going up to 25 mg instead of 50 mg, so it makes a bit of adjustment there."

Dr Robert M Califf

Asked by Topol whether he felt that the two diuretics should be considered similar, Califf said, "The answer I always give, regardless of which class you're talking about, is 'As far as I know, there's not a major difference, but there could be.' We have examples where drugs in the same class don't get the same results, so I'm looking around to find chlorthalidone these days."

In ALLHAT, the diuretic came out equal to lisinopril and amlodipine in the primary end point of cardiovascular death or nonfatal infarction and appeared superior to lisinopril in reduction of stroke and heart failure and to amlodipine in the reduction of heart failure. However, in such metabolic end points as blood-glucose levels and new-onset diabetes, both lisinopril and amlodipine proved superior to chlorthalidone.

Topol pointed out that these undesirable metabolic outcomes with the diuretic should not be underestimated. "In some ways, you're picking your secondary outcomes, because someone's making a value judgment that getting diabetes is not as bad as heart failure," he said.


In some ways, you're picking your secondary outcomes, because someone's making a value judgment that getting diabetes is not as bad as heart failure.


Califf called this a "fair critique" but thought that given the cost differential and the expected outcomes, the majority of patients would opt for the cheaper agent.

However, since most of these patients will end up on more than one drug, where to go next after a diuretic fails to achieve blood-pressure targets becomes a more pressing question. Califf told heartwire that not enough is known about combination therapy, "but I think there will be trials now of adding a calcium channel blocker vs adding an ACE inhibitor, and of course the ARBs are still undefined."

"Nobody really knows exactly how to do these combination studies, to me that's a fascinating area of clinical trial design, and the only way to find out how to do the studies is to do a few and see what happens," he added.

ACE inhibitors: HOPE's dashed?

One of the hardest findings to swallow for many is the fact that lisinopril did not apparently live up to the high expectations set for this class of drugs by the Heart Outcomes Protection Evaluation (HOPE) with another ACE inhibitor, ramipril (Altace® - King Pharmaceuticals).

"I think the majority of people would have predicted that the ACE inhibitor would have been the winner of this trial in many waysbetter tolerated, less heart failure, better quality of life, and prevention of stroke, and it simply didn't do that," Califf said during the Thumbs program. He pointed out, though, that the "soft underbelly of HOPE all along has been the question of whether the patients enrolled in HOPE were being adequately treated with regard to their other risk factors."

That two other large ACE-inhibitor trials, EUROPA and PEACE, are still ongoing, "after many thousands of people enrolled, and well into the follow-up phase, it does make you wonder," he said. Both studies are scheduled to report later this year.

Dr Salim Yusuf

In an interview with heartwire , Dr Salim Yusuf (McMaster University, Hamilton, ON), coprincipal investigator of HOPE, addressed some of the issues raised about how ALLHAT reflects on the results of that trial.

"I think HOPE and ALLHAT are complementary, not contradictory," Yusuf said. Blood pressure for ALLHAT patients at study outset was 146 to 147 mm Hg systolic; in HOPE, it was 139 mm Hg. HOPE therefore answers a different question, he says: Once blood pressure is down to accepted levels, what does adding an ACE inhibitor do?

"HOPE looked at the effect of an ACE inhibitor on top of a diuretic or on top of a beta blocker or on top of a calcium channel blocker, and in each one of those cases, there was benefit," he said. "If you want to use a diuretic, that's great, and I think it's a good idea, but what HOPE is telling you is that an ACE inhibitor on top of that is of benefit."


If you want to use a diuretic, that's great . . . but what HOPE is telling you is that an ACE inhibitor on top of that is of benefit.


He also pointed to "marked differences" between the two populations studied in these trials. The ALLHAT population comprised about 35% African Americans; in HOPE that number was about 5%, and ACE inhibitors are thought to be less effective in African Americans. In HOPE, about 80% had coronary artery disease, seen in less than 25% of ALLHAT patients. These factors resulted in a change in the ratio of event rates between the studies, with many more of the ALLHAT group at higher risk of stroke rather than heart disease, Yusuf said.

Whether there is any real difference between ramipril and lisinopril is just not known, he said, but he did point out that the dose of lisinopril in ALLHAT may not have been optimal, something that is not clear from the publication. "We know at the end of the trial, but we don't know the dose of lisinopril during the trial." In HOPE, he says, "we wanted to get to a full dose straightaway, with in a month, irrespective of blood pressure."

Results of SECURE, a substudy of HOPE, confirmed that 2.5 mg and 10 mg per day of ramipril produced similar blood pressures, but the effect on atherosclerosis prevention and left ventricular mass was much more evident with the higher dose.

"I don't have a problem to say start with diuretics, but I think most people who did ALLHAT will say you very quickly need to add additional drugs. And at this stage, in people who are high risk irrespective of blood pressure, I'd add an ACE inhibitor."

ANBP-2: Direct contradiction of ALLHAT?

Since the ALLHAT results were released, a second large hypertension trial was published in the New England Journal of Medicine, the Second Australian National Blood Pressure Study (ANBP2). Findings in the new trial raise more questions about what conclusions can be drawn about both diuretics and ACE inhibitors. The main result was opposite to those seen in ALLHAT: hypertensive patients started on the ACE inhibitor enalapril appeared to have better clinical outcomes overall than those begun on the diuretic hydrochlorothiazide.

The primary end point of any cardiovascular event or death from any cause was significantly reduced by 11% with enalapril over the diuretic, with a more pronounced effect in men than in women, despite similar reductions in blood pressure. Rates of nonfatal cardiovascular events and MI were also reduced by the ACE inhibitor. However, a similar number of strokes occurred in each group, with more fatal strokes seen with enalapril, the researchers reported.

In an accompanying editorial, Dr Edward Frohlich (Ochsner Clinic Foundation, New Orleans, LA) explored some of the possible differences between the trials. First, whether or not there are differences between the two diuretics or the two ACE inhibitors is not known, and the effects of other drugs added to get patients to goal increases the complexity of comparing the trials, he writes. It is possible that the diuretic looked better in ALLHAT because it lowered blood pressure in that population better than lisinopril, but if blood pressures were equal, perhaps the ACE inhibitor would have the advantage, Frohlich speculated. As pointed out with HOPE, ACE inhibitors may work differentially in blacks and whites, and again, unlike ALLHAT, 95% of those in ANBP2 were white.

Primary and secondary outcomes were also defined differently in the two trials. The primary outcome in ANBP2 was total fatal and nonfatal cardiovascular events, which favored enalapril, whereas in ALLHAT, the treatment groups were similar in the primary outcome of death from coronary causes or nonfatal MI, but when combined with secondary cardiovascular events, outcomes favored chlorthalidone.


Treatment of the individual patient with hypertension is complicated, requiring time and judgment.


Frohlich suggests that for uncomplicated essential hypertension, a diuretic followed by an ACE inhibitor if blood pressure is not controlled "makes sense." Other agents may come into play with other comorbidities such as diabetes, history of MI, or heart failure.

"Treatment of the individual patient with hypertension is complicated, requiring time and judgment," he concludes. "In choosing between a diuretic and an ACE inhibitor, the physician can make a reasonable selection by reviewing the patient's history."

No need for "spin"?

Finally, while ALLHAT investigators were wary of "spin" from the drug companies involved in the trial obscuring their message, there doesn't seem to have been a lot of evidence of this in the meantime.

"I think the drug companies have handled it very smartly," Califf told heartwire . "The biggest thing for them is that treatment of hypertension by and large is inadequate, so any noise that's made about hypertension will increase the use of all the classes of drugs. People have idiosyncratic side effects and if it takes more than one drug, you're going to end up with all classes of drugs being used more often if hypertension is treated more effectively."

Drug companies are most concerned about what is bad for them, he says. "Pfizer and any company with a calcium channel blocker must feel good on a relative basis, and the ACE inhibitors didn't look as good as expected, but they've got a pretty strong argument that since the second drug with an ACE inhibitor is usually a diuretic and since that was prohibited in ALLHAT, then maybe the ACE inhibitors were disadvantaged."


I don't think there's a loser in ALLHAT, really.


Added to this argument are the new findings from ANBP2, Califf added. "So, I don't think there's a loser in ALLHAT, really."

"The most interesting question to me is whether the message about the diuretic will be driven home enough that people will really change practice as to what's used first, when there's not a sales force out there selling diuretics," he said.

Little marketing in the ACE inhibitor market?

Tony Butler, an analyst for Lehman Brothers, New York, told heartwire they haven't seen significant fallout for these companies from release of the ALLHAT findings for a variety of reasons. First among these is that current hypertension guidelines call for diuretics to be used as first-line antihypertensive therapy, he says, but he pointed out, as many clinicians have, that most patients end up on more than one drug over time.

"It's clear that the trial, which was designed during the early 1990s, did not test the effects of polypharmacy as it tends to be practiced today," Butler said. "We think most cardiologists would certainly agree that prescribing more than one agent from various therapeutic classes is the standard of care. Thus, the demonstration that a diuretic prescribed alone is as good as any other single agent is interesting but hardly enough to alter the practice of medicine that's evolved over the past decade."


Most cardiologists would certainly agree that prescribing more than one agent from various therapeutic classes is the standard of care.


Nor have they seen in the months since the ALLHAT publication any prescribing changes to indicate a significant change in behavior. According to prescription tracking information obtained from Scott-Levin, calcium channel blockers, for example, still occupy 18% of prescriptions in this area. Of these prescriptions, 20% include a diuretic, and 60% have some other coprescribed medication, Butler says.

The impact of the ALLHAT findings on ACE inhibitors will be hard to judge, given current trends in the marketing of ACE inhibitors as more of these agents come off patent, Butler said. "For the most part, despite Altace and a modest amount of Accupril® (quinapril, Pfizer), there's not a lot of marketing in the ACE-inhibitor market, so I'm not clear that one study actually drives a conclusion that changes behavior," he said, particularly, again, given the design of the study. "But I am suggesting that if you correlate that with a reduction in the marketing of ACE inhibitors, you might assume the combination may drive a fall-off in the total class." However, they have not yet seen such an effect, he noted.

Finally, good results have recently been reported with A2-receptor blockers, including losartan (Cozaar® - Merck) in the LIFE and RENAAL studies and valsartan (Diovan® - Novartis) in congestive heart failure, Butler added. "The question is then whether those agents do not get greater focus because there are three different marketers supporting some very strong outcomes with those agents."

No losers

As for the drug companies themselves, none appear to feel like a "loser" in the trial, although all made a very big point of supporting the use of diuretics in this population.

At Pfizer, news that amlodipine (Norvasc®) came out neck and neck with the other agents in the primary outcome was extremely welcome, Dr Michael Berelowitz (vice president for worldwide medical, cardiovascular and metabolism group) told heartwire .

"We had two agents as comparators in the study, we had provided financial support for the study, and it was a big opportunity for us to see that comparative efficacy and safety of Norvasc in a huge, NHLBI-sponsored and managed clinical program," he said. "Norvasc came out equivalent to a proven effective antihypertensive agent in the primary outcome and in all of the prespecified secondary outcomes." Some end points, such as stroke, favored amlodipine.

"When you consider that among a group of cardiologists who had had concerns about the safety of calcium channel blockers in general and Norvasc in particular, not only were we safe, but in fact there was a decreased noncardiovascular mortality in favor of Norvasc," Berelowitz said.

However, he acknowledged that there was more heart failure with amlodipine, as there was with lisinopril and doxazosin, compared with chlorthalidone. "Diuretics are indicated for the treatment of heart failure, so to have more heart failure than a diuretic is to some extent perhaps not unexpected."

AstraZeneca provided the lisinopril used in ALLHAT, marketed under the name of Zestril®. The company's director of public affairs, Gary Bruell, told heartwire that diuretics have always been first-line therapy but added, "Lisinopril has provided physicians with an effective and safe treatment option for hypertension, angina, and heart failure, that's all we can say, and ACE inhibitors will continue to play an important role in hypertension."


Ultimately, it's what's best for an individual patient, and a physician will decide that.


"Ultimately, it's what's best for an individual patient, and a physician will decide that. If diuretics aren't working, the next step is to add another agent," he said. He added, though, that "a large proportion of patients in ALLHAT were African American, and ACE inhibitors normally aren't the most effective way to treat that population." Merck & Co, which makes a second lisinopril product, Prinivil®, declined to comment on ALLHAT except in general terms. "I know they're the same compound, but we're going to leave it to AstraZeneca to comment," a spokesperson said.

King Pharmaceuticals (Bristol, TN), whose wholly owned subsidiary Monarch Pharmaceuticals markets both ramipril and chlorthalidone, finds itself in the odd position of potentially benefiting with its chlorthalidone product Thalitone® from ALLHAT results but faces a possible challenge to ramipril because of the ACE inhibitor findings.

John Lambrechts, senior director of investor relations, told heartwire that it is the company's position that lisinopril and ramipril are very different agents and that ALLHAT findings should not detract from the benefit of ramipril seen in HOPE. "These are very different studies," he said. He also pointed to the fact that ACE inhibitors were not used in ALLHAT as they would be in "real life," most often with a diuretic.

As for their chlorthalidone product, Lambrechts says they have not seen much difference in sales at this point but added that the company may consider producing a combination product of chlorthalidone and ramipril.

Related links

1. [Education > HeartBeat; Jan 15, 2003]

2. [Heartwire > News; Dec 31, 2002]

3. [Heartwire > News; Dec 17, 2002]

4. [Heartwire > News; Dec 17, 2002]

5. [Heartwire > News; Nov 26, 2002]

6. [Heartwire > IndustryPulse; Nov 15, 2002]

7. [Heartwire > News; Oct 10, 2002]



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: