Pharma Efforts to Bar Placebo Responders from Trials – WSJ
Fri, 25 Jun 2004
A clinical trial report in The Lancet debunks the value of Aricept, a popular drug prescribed for Alzheimer’s patients. The findings of a three-year placebo controlled trial show that: “Although the patients taking the drug did have slightly higher scores on mental tests, after three years they did not differ from the placebo group in their rates of being put in a nursing home or becoming disabled. There were also no significant differences between the groups in behavioral or psychological symptoms or in the emotional well-being of the people taking care of the patients.” See: British Study Sees Scant Value in Alzheimer’s Drug Aricept, NY Times, June 25, 2004, A-20. http://www.nytimes.com/2004/06/25/health/25alzh.html
As disappointing as such findings are for patients whose hopes had been raised about the drug’s effectiveness, for those with financial stakes in the marketing of new drugs, such set- backs cut profits. Placebo-controlled clinical trials were declared “the gold standard” by the FDA and pharmaceutical companies who were loath to test their new drugs against existing ones, lest the (always cheaper) existing drugs proved more effective than the new. But, as clinical trial after clinical trial reveals, today’s new drugs can’t even perform better than a sugar pill.
Since 1962, US law requires drug manufacturers to submit proof to the FDA–not only of a drug’s safety, but its effectiveness for specific medical purposes, through “adequate and well-controlled” studies. But recent revelations about clinical trials–specifically, concealed drug hazards and failed efficacy–raise serious questions about the credibility of trials that were submitted to the FDA to gain marketing approval.
If FDA-approved drugs fail the placebo comparison in clinical trials conducted by independent scientists, how valid are the claimed findings from pre-approval trials?
Furthermore, one wonders how many of those enrolled in a clinical trial who responded to placebo were included in the data analysis of those trials? Clearly, the FDA’s drug approval process isn’t identifying ineffective drugs, and is approving such drugs in contradiction to the US Food and Drug Administration Act.
The suit filed by NYS Attorney General, Eliot Spitzer, against GlaxoSmithKline, charging fraudulent concealment of data, has sent shivers down the spines of pharmaceutical CEOs.
See: The Guardian. Glaxo changes tack after Spitzer assault, by Heather Stewart, June 19, 2004, http://www.guardian.co.uk/business/story/0,3604,1242494,00.html
See Glaxo internal memo, 1998 : http://www.ahrp.org/risks/SSRI0204/GSKpaxil/pg1.html
Comments by AHRP board member, David Cohen, Ph.D., addressing the placebo effect in psychotropic drug trials, are followed by an article in The Wall Street Journal.
Leila Aboud reports in the Wall Street Journal (below) that Eli Lilly and Pfizer plan to fund UCLA-based research using brain scans “to identify patients who respond strongly to placebos, who could then be excluded from clinical studies of new antidepressants.” Undoubtedly, such research primarily serves commercial, not scientific or ethical imperatives.
Placebo response has always been high, especially in trials of anti-anxiety and anti-depressant drugs, where it regularly approaches 50%; but it is substantial even in studies of patients with acute manic or schizophrenic psychosis, where it averages around 25%.
The spin put on publications from investigators and trial sponsors with a direct stake in the tested drug’s performance, however, tended to distract from this high placebo response. It also obscured the large variability of drug response among individual patients and study samples. It led government agencies, health care professionals, and consumers to exaggerate the benefits of drugs and downplay their harmful effects.
However, revelations and independent studies concerning “publication bias” – the concealment of vital information from published reports of clinical trials by sponsors, researchers, and authors – have altered the landscape. It now looks like the major differences between placebos and SSRIs – hugely profitable “specific medications for specific disorders” – may amount to the following: SSRIs have brand names, enjoy huge advertising support, and are more harmful.
In this light one can understand the jitters from drug makers. Now that their control over reporting clinical trial results is beginning to erode, they are attempting to modify clinical trials themselves in their favor. The reason is that placebo responders “are the people who ruin clinical trials for drug companies.” Steven Paul, the head of research and development at Lilly is quoted as saying: “The placebo response has plagued psychiatric clinical trials.”
Paul’s quote aptly describes the point of a clinical trial in today’s interest-conflicted and profit-driven psychiatry: not a robust scientific technique to establish whether a drug’s healing effects surpass those resulting from suggestion and expectation, but a hurdle to gain FDA approval and reap multi-billion dollar profits.
The WSJ article states that drug “companies dismiss the criticism that they will use the information [about placebo responders] to gain approval of ineffective drugs. They say the techniques would be used only in early trials, not for trials done for FDA approval.” But many questionable strategies that help tested drugs look positive are now used to gain FDA approval. Why wouldn’t drug makers use new strategies? Second, Drug makers already enjoy considerable design latitude for early trials and rarely use strict placebo controls. The issue of placebo response arises specifically in Phase III randomized, controlled clinical trials submitted for FDA approval, as the FDA requires at least two such trials.
The placebo effect remains crucial to understand any self-reported or observed improvement in a drug trial, especially for disorders that cannot be diagnosed with objective bodily markers. Much of what appears as drug-induced improvements in symptom rating scales may be part of the “placebo effect.” Because the nature of the placebo effect has proven exceedingly difficult to identify, its benefits have been easily claimed by others.
All efforts to understand the nature of placebo response should be encouraged, though one can question the simplistic mindset of some researchers attempting to distinguish placebo responders on the basis of brain scans. There is no doubt that physiological responses underlie placebo responses. However, as Charles Medawar and Anita Hardon hint in their book, Medicines Out of Control? (London, 2004), the placebo effect includes many factors that do not fit either in the user-drug or doctor-patient relationship. These factors include hope, faith, belief, susceptibility to influence from advertising, as well as dependency on the power of organizations.
However, if the spotlight needs to shine on the chemistry of those who improve their distress because of a complex response triggered by ingesting a sugar pill, then we might entertain recruiting some other study participants. Perhaps equal attention should be directed to doctors as placebo responders, with brain scans of physicians used to determine why such a high proportion of them continue to prescribe drugs that research regularly shows to be only slightly more effective than placebos.
David Cohen, PhD, Secretary AHRP
THE WALL STREET JOUNRAL
Drug Makers Seek to Bar ‘Placebo Responders’ From Trials
By LEILA ABBOUD
June 18, 2004; Page B1
They are the people who ruin clinical trials for drug companies: placebo responders, who get better on sugar pills. Drug makers want to get rid of them, especially in trials of depression drugs, where placebos can have a particularly powerful effect. The problem: Nobody knows who they are.
Now drug companies are looking for ways to identify patients who respond strongly to placebos, who could then be excluded from clinical studies of new antidepressants.
With funding from Eli Lilly & Co. and Pfizer Inc., scientists at the University of California, Los Angeles, are beginning to unravel the mysteries of the placebo effect in depressed patients. Armed with brain-imaging technology, these researchers believe that placebo responders share the despairing mood that other depressed people feel. But the responders may not have the physical symptoms often accompanying severe depression, including disruptions of sleep and thinking. In a way, they may not be as severely depressed — and it may be possible to spot them ahead of time.
Another technique for finding placebo responders is being tried by Pfizer, which has found that such people have different DNA in key areas of the genome that are linked to depression. So Pfizer is exploring whether genomics could also one day be used to pick out placebo responders.
The placebo effect has existed since snake oil and bloodletting. Scientists noticed that big placebo pills worked better than small ones, brightly colored ones were more potent than dull and sham surgeries — entailing anesthesia and an incision but no actual repair — were the most powerful of all.
Many antidepressant trials fail because the placebo effect can be so powerful and highly variable. Somewhere between 30% to 50% of patients in depression trials get better when given fake pills, and that number has increased and become more volatile over time, making it more difficult to prove that a drug works. In comparison, only about half of patients taking antidepressants find their symptoms relieved by 50% or better.
“The placebo response has plagued psychiatric clinical trials,” says Steven Paul, head of research and development at Lilly. If drug companies could sort out the placebo responders, it “would greatly help us in the short term in drug development.”
Psychiatric medicines are huge sellers. Antidepressants racked up more than $17 billion in world-wide sales in 2003 and were the third-largest class of drugs by sales.
To get permission from the Food and Drug Administration to market a new drug, its maker must show that it works better than a placebo in at least two large, controlled studies. The drug companies say that weeding out placebo responders would allow them to do much smaller, quicker, cheaper trials to find out if an experimental antidepressant actually worked.
“It would tell us early on in development whether we have an effective agent or not,” Lilly’s Dr. Paul says. Such quick answers would prevent the industry from spending years researching new kinds of antidepressants that turn out to be ineffective.
Many clinical trials have found that placebos are almost as effective as the actual antidepressants:
Drug Brand Name Number of Trials Response* to Drug Response* to Placebo
Fluoxetine Prozac 5……………. 8.3 ……………7.34
Paroxetine Paxil 12……………. 9.88………….. 6.67
Sertraline Zoloft 3……………. 9.96 ………….. 7.93
Venlafaxine Effexor 6…………… 11.54 ………….. 8.38
Citalopram Celexa 4……………. 9.69 ………….. 7.71
*Participants’ average improvement, measured in points on a standard scale used to quantify the severity of depression
Source: Prevention & Treatment
The cost of failure can be immense. After a decade of study, Merck last year abandoned its effort to develop a new type of antidepressant when it couldn’t do better than a placebo in tests; Merck didn’t reveal how much the effort cost.
But some clinical-trial experts worry that excluding placebo responders to facilitate drug development could lead to the approval of ineffective drugs. Kay Dickersin, a Brown University professor who teaches courses on clinical trials, says such winnowing “allows bias to enter in” and constitutes “a subtle manipulation” of trial results.
The companies dismiss the criticism that they will use the information to gain approval of ineffective drugs. They say the techniques would be used only in early trials, not for trials done for FDA approval.
It’s too early for the FDA to have weighed in on the issue.
New York State Attorney General Eliot Spitzer recently threw a spotlight on antidepressants when he sued GlaxoSmithKline PLC, charging that the company hid clinical trials from the public in which its antidepressant Paxil worked no better than placebos in children and adolescents. Glaxo has asserted it didn’t hide the trial results.
Why are placebos so potent in depression trials? Some blame technical flaws in the way clinical trials are done, such as including people who are not clinically depressed and using inexact ratings to measure complicated feelings. Scientists speculate that human factors may also play a role: Patients who like the researcher or nurse doing the study may get better even without actual medication.
The biggest complication may turn out to be intrinsic to antidepressant trials. Study subjects are divided into two groups. One gets the drug and regular meetings with researchers to track their progress, while the placebo group gets sugar pills as well as the regular meetings.
There’s the rub: People in the placebo group are actually getting treatment of a kind. They still have the doctors’ visits with someone paying attention to their distress and trying to address it. That medical interaction, and the hope engendered by simply being in a study, may be responsible for their improvement even when on a fake antidepressant.
Pfizer and Lilly, unlikely allies, are together ponying up more than $1 million to fund the research at UCLA. Psychiatrist Andrew Leuchter, vice chairman of the university’s Department of Psychiatry and Biobehavioral Sciences, will lead a study aimed at developing tools to spot people who seem to get better without drugs. The 11-week study will examine 60 patients, mapping their brain waves four times, recording their sleep patterns, and checking their mental agility at solving puzzles.
“The placebo responders look just the same as the drug responders,” Dr. Leuchter says. “But in some subtle ways they may be different.”
Dr. Leuchter believes that placebo responders, although they appear just as sad and down as other depressed patients, actually have less severe physical symptoms. That’s why tests related to sleep and mental agility may be able to pick them out, he says.
Some doctors worry that making the trials more restrictive would make the test subjects even more unlike the depressed people who actually show up in their offices seeking treatment. “If you exclude a whole bunch of placebo responders,” says Duke University psychiatrist P. Murali Doraiswamy, “the results won’t apply to the patients I see in my practice.”
FAIR USE NOTICE: This may contain copyrighted (C ) material the use of which has not always been specifically authorized by the copyright owner. Such material is made available for educational purposes, to advance understanding of human rights, democracy, scientific, moral, ethical, and social justice issues, etc. It is believed that this constitutes a ‘fair use’ of any such copyrighted material as provided for in Title 17 U.S.C. section 107 of the US Copyright Law. This material is distributed without profit.