According to the drug company giant Pfizer, 2.4 million Americans have taken donepezil (Aricept) since it was approved by the Food and Drug Administration in 1996 for use in mild to moderate Alzheimer’s disease. The British think tank, NICE (National Institute for Health and Clinical Excellence), which evaluates the effectiveness of drugs and devices, reported last month that the drug’s usefulness is very limited. If it’s helpful to anyone, it’s mainly those with somewhat more advanced Alzheimer’s who stand to benefit. By contrast, the American Academy of Neurology says that “considering” Aricept is the “standard” approach for mild to moderate disease (RS Doody et al, “Practice Parameter: Management of dementia: An Evidence Based Review. Report of the Quality Standards Subcommittee of the American Academy of Neurology,” Neurology 2001; 56: 1154-66.) And the Alzheimer’s Association, the leading non-profit advocacy group in the field, while falling short of endorsing the use of Aricept and related drugs, recommends their use “if clinically indicated, to treat cognitive decline.” (California Workgroup in Guidelines for Alzheimer’s Disease Management: Alzheimer’s Association of Los Angeles, Riverside and San Bernardino Counties, 2002).
The latest dispute over anti-Alzheimer’s drugs began last March when, after performing an exhaustive review of every study published on the effect of the drugs on cognition, on daily functioning, or on behavioral symptoms, NICE issued a report concluding that neither Aricept nor its newer cousins, Rivastigmine and Galantamine, were worth the money the National Health Service (NHS) spent on them. They clearly did something—they produced a statistically significant improvement in various test scores, but the real life benefit was miniscule. When NICE released its report, family members of Alzheimer’s patients, physicians, and of course the drug manufacturers, were outraged. They were convinced, based in many cases on anecdotal experience, that the drugs worked. NICE responded by asking the drug companies to produce evidence that there were in fact subgroups of people who benefited substantially from this entire class of drugs, known as cholinesterase inhibitors.
Last month (January, 2006) NICE released a new report in which it once again asserted the drugs are of no meaningful use in people with mild Alzheimer’s. They concede there may be enough benefit in people with moderate Alzheimer’s to warrant trying them. In other words, even after bending over backwards to find some justification for recommending the use of Aricept and similar drugs in people with early Alzheimer’s, NICE just couldn’t find any.
The view that Aricept, with US sales in 2004 of just under a billion dollars, probably shouldn’t be used in most of the patients for whom it’s prescribed, has so far been largely ignored in the U.S. It hasn’t been picked by the New York Times or the Washington Post or by the news section of leading American medical journals such as the Journal of the American Medical Association. And this isn’t the first time we in America have chosen to pay little attention to fairly compelling evidence that Aricept doesn’t work well. In 2004, another British study, this one a randomized controlled clinical trial, tested whether Aricept could delay nursing home placement, a claim made in many advertisements for the drug. The study, which compared 283 patients who got Donepezil with an equal number who did not, found the two groups had exactly the same chance of admission to a nursing home (AD2000 Collaborative Group, “Long-Term Donepezil Treatment in 565 Patients with Alzheimer’s Disease: Randomised Double-Blind Trial,” Lancet 2004; 363:2105-10.) It was published in a leading journal, the Lancet, and did generate a response by the Alzheimer’s Association, journals such as the Journal of the American Geriatrics Society, and also by the New York Times, but many practicing physicians remained unaware of its existence. The study has been criticized on a variety of technical grounds. But why accept the results of earlier, small, non-randomized studies and reject the results of a much better, although less than perfect, randomized study? Only a fervent desire to believe the conclusions of the study that professed to show Aricept keeps people out of nursing homes can explain this phenomenon.
The irony is that when the cholinesterase inhibitor Tacrine was first alleged to help in Alzheimer’s disease, back in 1986, the American neurologic community was extremely skeptical. How could a drug that worked by preventing the destruction of a neurotransmitter in the brain (acetylcholine) be useful in a disease in which the nerve cells that used that neurotransmitter had already been destroyed? And the skeptics seemed justified when the experiments behind the first enthusiastic reports were found to be if not fraudulent, at least questionable. After years of further study, rigorous tests established that the drug did have a statistically significant effect. Whether it had a clinically meaningful effect was another matter.
Even when Tacrine was approved by the FDA in 1994, and then, in record time, its successor, Aricept, was also approved, many Alzheimer’s researchers were dubious that the drugs was actually helpful. John Growdon, a prominent researcher at the Massachusetts General Hospital in Boston, wrote diplomatically in an editorial in the New England Journal of Medicine that “the expectation of benefit…exceeds reality.” Yet use of cholinesterase inhibitors in the US has become more and more widespread. In 2004, Aricept was ranked #52 in the top 200 drugs by sales in the US ($.92 billion, up from a rank of 62 and sales of $0.7 billion in 2003). Why?
Alzheimer’s is perhaps the most feared disease of old age. Called “the gray plague,” it makes once vibrant people debilitated and dependent (see my book, Tangled Minds: Understanding Alzheimer’s Disease and Other Dementias; NY: Dutton, 1998). It’s not surprising that family members of those afflicted are desperate for a treatment. It’s not surprising that physicians are eager to have something to offer—and one after another, the various proposed remedies have proved worthless or even harmful, including estrogen and Rofecoxib (Vioxx). But should we be spending nearly a billion dollars a year to indulge our wishful thinking? Or would we all be better off spending the billion dollars on research for other drugs which might actually be effective? Or perhaps we could invest the billion dollars in improving the quality of nursing homes and assisted living facilities that claim to specialize in the care of people with Alzheimer’s so as to find ways to promote dignity and well-being in some of our most vulnerable and most demeaned citizens. The British seem to recognize that an unchecked appetite for drugs that might be beneficial has consequences when our resources are unlimited. Americans are loath to accept this view. Maybe we have something to learn from our English cousins.