Do Not Use Before April 2005 Rivastigmine (EXELON)—A Third Drug For Alzheimer’s Disease

The Food and Drug Administration (FDA) approved a third drug, rivastigmine (EXELON), in April 2000 for the treatment of mild to moderate dementia of the Alzheimer’s type. Rivastigmine was preceded by tacrine (COGNEX) approved in September 1993 and donepezil (ARICEPT) cleared by the FDA in November 1996. Rivastigmine is produced by Novartis Pharmaceuticals of East Hanover, New Jersey.

The editors of The Medical Letter on Drugs and Therapeutics, an independent source of drug information written for physicians and pharmacists, concluded their October 2, 2000  review of this newest drug by saying, “…there is no convincing evidence that rivastigmine markedly improves quality of life in patients with Alzheimer’s disease or substantially alters progression of the disease.” The Medical Letter editors had already concluded in earlier reviews of tacrine and donepezil that there is no evidence that either leads to substantial functional improvement or prevents progression of Alzheimer’s (see the August 1997 issue of Worst Pills, Best Pills News). We agree with The Medical Letter editors’ opinions on all three of these drugs.

Tacrine sales slumped to the point where the drug’s manufacturer, then Warner-Lambert (now merged with Pfizer), announced that effective June 12, 1997  it suspended advertising to physicians and discontinued patient support programs. The drug is still available, however. Tacrine is associated with liver toxicity.

In the September 1999 issue of Worst Pills, Best Pills News we reported that the British equivalent of the FDA had required new safety warnings on donepezil’s labeling. The U.S. labeling was changed on February 28, 2000 with the addition of the following warnings:

Voluntary reports of adverse events temporally associated with Aricept that have been received since market introduction that are not listed above, and that there is inadequate data to determine the causal relationship with the drug include the following: abdominal pain, agitation, cholecystitis, confusion, convulsions, hallucinations, heart block, hemolytic anemia, hyponatremia [low blood sodium levels], pancreatitis [inflammation of the pancreas] and rash.

All three drugs work by inhibiting an enzyme that breaks down acetylcholine, a brain transmitter, a deficiency of which is thought to play a role in Alzheimer’s disease.

The FDA approved rivastigmine on the basis of two clinical trials comparing rivastigmine to an inactive dummy pill (placebo). The effect of the drug was measured using two different rating scales. The first was the Alzheimer’s Disease Assessment Scale (ADAS-cog). The ADAS-cog examines selected aspects of a patient’s cognitive performance including memory, orientation, attention, reasoning, language and habits. Scores range from 0 to 70, higher scores indicating greater cognitive impairment.

The second rating scale used was the Clinician Interview Based Impression of Change (CIBIC) scale, which uses information from both the patient and physician and is a broad assessment of behavior, general pathology, cognition, and activities of daily living. CIBIC scores range from 1 to 7 with scores of 1, 2, or 3 equaling marked, moderate, or minimal improvement, respectively. A score of 4 is no improvement and scores of 5, 6, or 7 mean minimal, moderate, or marked deterioration, in that order.

One of these trials was published in the March 6, 1999 issue of the British Medical Journal. It involved 725 patients in Europe and North America and lasted 26 weeks. The patients were randomized to receive daily low dose (1 to 4 milligrams) or high dose treatments (6 to 12 milligrams) with either rivastigmine or placebo. 

On the ADAS-cog rating scale, using scores from all the patients who started the study and receiving high dose treatment, the difference from the start of the study between those taking rivastigmine and the placebo was only 1.6 on the 70 point scale. This difference was not reported as statistically significant. However, using less rigorous analyses of the data there was a statistical difference between the treatments on the ADAS-cog rating scale.

The difference in scores on the CIBIC rating scale between those receiving high dose rivastigmine and placebo was less than half a point (0.47) on the 7 point scale at the end of the 26 week study. This difference was statistically significant.

In patients similar to those enrolled in pre-marketing clinical trials, the known benefits of rivastigmine are: (1) a small positive change in ADAS-cog score after 26 weeks compared to no treatment; and (2) a small positive change in the CIBIC rating score after 26 weeks compared to no treatment. This is the regulatory meaning of the phrase: rivastigmine is an effective treatment for mild to moderate dementia of the Alzheimer’s type. In everyday language treatment connotes a cure or remedy. This is very different from the regulatory definition.

In clinical trials conducted before rivastigmine was approved, nausea, vomiting, diarrhea, abdominal pain and loss of appetite were the most common adverse effects. Weight loss occurred in some patients. Approximately 20 percent of patients were unable to tolerate high dose rivastigmine because of adverse reactions. The liver toxicity that is seen with tacrine has not been reported with rivastigmine.

Why are we seeing a tide of minimally effective, perhaps injurious, and wasteful “me too” drugs that only raise the cost of care of Alzheimer’s disease with a minimally detectable benefit being approved by the FDA? Part of the reason is the 38-year-old law that the FDA must follow to approve new drugs.

In 1962, the Kefauver-Harris Amendments to the Food, Drug and Cosmetic Act established an effectiveness standard for the approval of new drugs. This was in addition to a 1938 requirement to test new drugs for safety before marketing. The 1962 law did not say that a new drug had to be effective before it was cleared for sale. It said there had to be “substantial evidence” of effectiveness, defined in the law as:

“... evidence consisting of adequate and well-controlled investigations, including clinical investigations, by experts qualified by scientific training and experience to evaluate the effectiveness of the drug involved, on the basis of which it could fairly and responsibly be concluded by such experts that the drug will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, suggested in the labeling or proposed labeling thereof."

Notice that this law does not require that a new drug be safer or that there is substantial evidence that it is more effective than drugs already on the market. In fact, using this standard it is possible that a new drug can be approved that is less effective and less safe than drugs already on the market. This is clear from the recent epidemic of newly approved drugs that had to be removed from the market for safety reasons.

Over time the default for substantial evidence has evolved to be a statistical showing that the new drug is more effective than no treatment at all, in the form of a placebo, in two clinical trials. If a company can show that its new drug meets this standard, and has been tested for safety, the FDA must allow the company to market the drug. It is the law.

This 38-year-old standard does not allow for answering the question that is most important for patients and physicians: What drug is the safest and most effective for a given condition? The FDA’s hands are tied, they do not have the legal authority to require a comparison between a new drug and older drugs as a condition for approval. Only Congress can raise the bar for drug approval by changing the law and every time Congress has raised the bar over the last century patients have benefited from safer and more effective drugs.

What You Can Do

If you are considering rivastigmine treatment for a family member, before starting the drug carefully discuss with your doctor the benefits that can be expected from it. Our well-known “five-year rule” says you should not use a new drug until it has been on the market for at least five years because pre-market testing does not provide a large enough “statistical universe” to make it likely that hidden hazards will show up. The five-year rule clearly seems to apply in the case of rivastigmine.