Drugs for the Earliest Stage of Alzheimer’s Disease


by Guy McKhann, M.D.

April 29, 2013

The development of a new pharmacologic approach to a disease is a long and extremely expensive process. Estimates suggest that it takes 12 years and costs around $1.2 billion from the first research on a drug to its final clinical acceptance. That cost estimate can be much greater if the pharmaceutical company includes the cost of its drug failures as part of the calculation. Even large companies like Merck or Glaxo Smith Kline can only support a few drugs for expensive Phase 3 trials. Brain drugs must compete with drugs for other diseases such as cancer or heart disease. Not surprisingly, decisions are influenced by recent successes or failures. Thus if a drug for Alzheimer’s disease or stroke has failed in a trial, that company is gun-shy on further trials in brain drugs. That is exactly what is happening—big pharma has decreased its neuroscience operations and clinical trials of drugs for brain diseases.

A further complication is determining if a drug is effective. This step requires clinical trials in which a drug is compared with existing therapy. But what if there is no effective existing therapy, as is the case with Alzheimer’s disease? Then the drug must be shown to improve upon the natural history of the disease. In other words, the drug is compared with a population receiving no drug.

The situation with Alzheimer’s disease is more difficult because our concepts of what we are trying to treat have changed. Alzheimer’s disease involves a progressive degeneration of nerve cells in the cerebral cortex. The disease starts years before people show signs; it is clinically defined as a severe loss of memory, a dementia. By this latter stage many nerve cells are lost, and the chances of a drug having beneficial effects are slight. Thus the emphasis has shifted to trying to identify those in the earlier stages of the disease, when the damage to the brain is just getting started. A recent updating of our concepts about Alzheimer’s disease divided the disease into three stages: a preclinical phase; an interim phase of minimal cognitive impairment (MCI); and a dementia phase.

In the earliest (preclinical) phase, patients are functioning quite well but have demonstrable impairment on tests of cognition, and abnormalities on tests of the cerebrospinal fluid (CSF) and on brain scanning. If doctors are going to introduce a therapy at this stage, what would be the criteria of success? Previously the FDA required both a change in performance on cognitive tests as well as a change in ability to function. That latter criterion could take years to establish in these mildly affected individuals.

An editorial in The New York Times states the problem quite clearly, but in my opinion overemphasizes the risk for subjects who do not have Alzheimer’s possibly getting drugs with side-effects. The challenge to those who treat Alzheimer’s is to develop sensitive cognitive tests that demonstrate early impairment, and diagnostic tests that identify proteins that accumulate in the brain of those with Alzheimer’s. Although it has not been reflected in producing effective therapies, there is much going on in Alzheimer’s research. That research is a dynamic, evolving field, with many investigators capable of taking on the problems of defining early diagnosis.

A challenge to those in clinical neuroscience is to devise ways to simplify this process of demonstrating effectiveness of therapies, so that clinical trials involve smaller numbers of subjects, have defined predictable outcome measures, are of shorter duration, and are less expensive. For example, the population being studied can be simplified by being more homogenous, with a better defined clinical course and defined outcomes. Generally, in the development of a drug there is a jump from animal results to large trials in people. The results in animals may hint at clinical efficacy, but the animal model is often so far from the human disease or the therapeutic approach in animals so impractical in the human that using that data as the basis of an expensive human trial is hazardous. There is a crucial step, or question, missing: “Does the damn stuff do anything in the human?” That is the necessary first step in developing drugs for the first phase of Alzheimer’s disease.

[Ed.’s note: The latest Cerebrumarticle by Steven E. Hyman examines the underfunding for new medications for mental disorders. He surveys the reasons for its retreat and the changes necessary to meet the growing demand.]