Repurposing old drugs for Alzheimer’s treatment

We’ve talked a lot so far about clinical trials of several so-called disease-modifying drugs for Alzheimer’s disease. These are medications designed to slow and perhaps even stop the progression of Alzheimer’s. Most of them so far have been drugs that remove beta-amyloid from the brain or block its production. Most of the clinical trials of these anti-amyloid drugs have been disappointing, although recently aducanumab (Aduhelm) was approved by the FDA based on effectiveness of removing amyloid plaques from the brain and possible mild slowing of cognitive decline.  Several newer anti-amyloid drugs also look promising in early Alzheimer’s.  Trials of drugs designed to remove tau, the abnormal protein in neurofibrillary tangles, have all failed except for one that had possible benefit on two cognitive scales in more advanced subjects. The development costs to design, test, produce and do clinical trials on new drug candidates are enormous, costing millions if not billions of dollars.

A parallel effort is under way to find out if some drugs already in use for other disorders might have benefit in Alzheimer’s disease.  These are almost all inexpensive generic drugs, so the pharmaceutical industry is not interested in pursuing these leads. One drug company buried its own study that showed a small but significant decrease in the prevalence of Alzheimer’s in patients taking the company’s drug for rheumatoid arthritis. The drug was nearing the end of its patent protection, so the company apparently deemed that there was no financial incentive to do the studies necessary to get FDA approval to market the drug for Alzheimer’s. (In fairness, the benefit was modest at best and might not have held up after a rigorous phase 3 trial.)

Funding for these repurposing studies has mostly come from government grants, academic medical institutional support, and non-profit foundations like the Alzheimer’s Association.  According to a recent open-access review paper, in February 2020 there were 53 clinical trials underway involving 58 FDA-approved agents. Most of these old drugs (78%) were tested for possible disease-modifying activity, and the rest were tested for relief of symptoms associated with Alzheimer’s including confusion, anxiety, depression, psychosis and seizures.  The drugs included agents normally used to treat cancer, cardiovascular problems, psychiatric or neurologic problems, inflammation, and diabetes. No breakthroughs have emerged from all these trials yet, but most are still on going. Two recent studies suggest that some of these drugs may be effective in certain subsets of Alzheimer’s disease.

Levetiracetam (Keppra) is a popular drug used to treat a variety of types of epileptic seizures, and it is generally well tolerated. A little over 20% of people with Alzheimer’s have clinical seizures, most commonly complex partial seizures involving change in behavior or confusion without overt jerking of the limbs or signs of a generalized seizure. However up to 50% of people with Alzheimer’s have subclinical seizures, abnormal electroencephalograms (EEG) showing a pattern of epileptic changes but with no overt clinical signs of a seizure. A recent study published in JAMA Neurology looked at the effects of a small dose of levetiracetam on several cognitive tests. As shown in the figure below, tests of executive function showed improvement but only in those subjects who had abnormal EEGs with evidence of electrical seizure activity.  In other words, levetiracetam may be able to improve some cognitive problems in Alzheimer’s patients, but only in those for whom the cognitive impairment is associated with subclinical seizures.

Another very recent paper published in Nature Aging used a clever technique to screen existing drugs for the ability to flip a switch in the transcription of the APOE-4 gene in cell lines derived from APOE-4 transgenic mice and APOE-4 positive humans. The most effective drug at flipping this APOE-4 transcription switch was bumetanide (Bumex), a diuretic commonly used in older patients with severe edema and/or heart failure.  In the APOE-4 transgenic mice, beta-amyloid in the brain was markedly reduced after treatment with bumetanide (see figure below). Also the level of brain excitability returned to normal after exposure to bumetanide. A more accessible discussion of this complicated paper can be found in Alzforum.

In order to see if these laboratory effects of bumetanide might have real-world value in preventing Alzheimer’s, two large medical-record databases of patients over 65 were searched, one in California and one in New York. Compared to age and sex matched controls, the patients taking bumetanide in the California group were 35% less likely to have Alzheimer’s. In the New York group, the effect was even stronger. The patients taking the drug were 65% less likely to have Alzheimer’s than controls. A clinical trial is planned to start in 2022.

What these two new studies have in common is that each drug works only in a subset of people with Alzheimer’s.  The cognitive benefit of levetiracetam is found only in patients with seizure activity on their EEGs. Any benefit from bumetanide in slowing or preventing Alzheimer’s will likely be found only in those who carry the APOE-4 gene.