Chemotherapy for Alzheimer's?

A provocative opinion piece in JAMA Neurology offers new hope for the treatment of Alzheimer’s disease. University of California Santa Barbara neuroscientist Kenneth Kosik begins by acknowledging that recent drug trials have all failed, this despite the fact that that were designed based on the latest understanding of the pathophysiology of Alzheimer’s disease. The most spectacular recent failures were studies of two separate monoclonal antibodies that were expected to selectively attack amyloid deposition in the brain. This approach was so intriguing that scientists haven’t given up on it yet, despite the lack of success so far—just a few weeks ago the drug company Biogen announced the results of a Phase I trial of yet another monoclonal antibody, this one evidently quite promising. But Kosik has a different idea.

Kosik comments that the typical response to the drug trial failures is either to question whether our understanding of the mechanism by which Alzheimer’s is produced is correct or to conclude that treatment of the disease will need to begin before symptoms develop. He suggests that perhaps where we have gone wrong is in believing that Alzheimer’s disease is homogeneous. In fact, he argues, there are many genetic variants. The best known is the APOE alleles, with the APOE4 allele conferring particularly high risk. And of course there are the relatively rare familial forms of Alzheimer’s disease, which involve mutations in any of several chromosomes. Perhaps if we identify a person’s genome, we can find a particular therapy that will work for that person.

This is the approach that is increasingly used in the treatment of cancer, with a handful of notable successes. In non-small cell lung cancer, for example, there are several mutations typically found in younger patients with no history of smoking. Drugs targeted against these mutations have produced excellent results in patients with the relevant mutation: the drugs Erlotinib (Tarceva) and Crizotinib (Xalkori), when used in appropriate patients, can convert what was previously a uniformly lethal disease (usually within a year) into a chronic illness. Will the same strategy work for Alzheimer’s disease?

Maybe. It’s certainly worth pursuing. And Kosik also has ideas about how to proceed with this kind of targeted, individualized treatment—he recommends what are known as “N of 1 trials,” tests of a potentially useful drug in a single patient rather than the conventional strategy of testing a drug in a large group of people. The proposed strategy would both revolutionize the treatment of a particular disease (Alzheimer’s) and change the search process for new drugs. It’s not likely to change the epidemiologic reality any time soon, but it’s more promising than many radical new ideas. Let's hope NIH or other funding agencies have the wisdom to provide the support needed for this innovative work to go forward.