An Antidepressant to Prevent Alzheimer’s?

by Jim Schnabel

July 21, 2014

There is bad news and good news in Alzheimer's research these days. One piece of bad news is that the pharmaceuticals industry has begun to abandon its three-decade-long quest for drugs that slow or stop established Alzheimer's dementia. Every candidate treatment that has been tested so far has failed to show convincingly that it works in people whose brain degeneration has advanced that far.

The good news is that the prevention of Alzheimer's dementia-or at least, the delaying of the disease's symptoms-remains a viable option. In principle, a preventive drug could be taken throughout middle age to inhibit the Alzheimer's process in its early stage, just as one-a-day statins inhibit atherosclerosis and thus prevent heart attacks and strokes.

A bonus would be to find an existing "off-the-shelf" drug, approved for some other purpose, that also happens to work as an Alzheimer's preventive. Such a drug would likely be off-patent and thus relatively cheap, and it would have a well established safety profile. Although few if any pharmaceutical companies would want to invest heavily in an off-patent drug, even a little evidence that the drug works would likely draw government sponsorship, at least, for the needed clinical trials. Moreover, neurologists could prescribe such a drug "off label" for dementia prevention, in advance of full regulatory approval for that use, if it is already approved for other uses.

One off-the-shelf drug, bexarotene, which is used to treat a rare T-cell cancer that affects the skin, is already being investigated for its possible Alzheimer's-preventing effect. In the May 15, 2014, issue of Science Translational Medicine, researchers reported promising findings with another off-the-shelf drug, the antidepressant citalopram (Celexa).

SSRIs and amyloid beta

Citalopram belongs to a class of antidepressants known as SSRIs (selective serotonin reuptake inhibitors) because of the way that they boost serotonin signaling in the brain. Boosting serotonin signaling appears to treat depression. Recent studies in Alzheimer's-model mice suggest that it also reduces the production of the neuronal protein amyloid beta.

Amyloid beta is currently the prime target of candidate Alzheimer's preventives. Aggregates of amyloid beta-from hard-to-detect "oligomers" made of a few copies of the protein, to thousand-copy polymer fibrils that are easily visible under a microscope-are thought to accumulate in the aging brain for decades before triggering the classic Alzheimer's cascade of neurodegeneration and dementia. People who carry a gene variant that halves amyloid beta production appear to have near-zero Alzheimer's risk; conversely, people with gene variants that increase amyloid beta production, or its aggregation-proneness, have increased risk. Thus many scientists suspect that reducing the rate of production of amyloid beta or boosting its rate of clearance from the brain, starting when people are in their 60s or 70s, could keep Alzheimer's dementia at bay indefinitely.

In a study published earlier this year, European researchers found evidence that a rise in serotonin signaling reduces amyloid beta production by boosting the activity of the alpha secretase enzyme. The latter is one of a group of enzymes that can cut amyloid beta's precursor protein, APP, at different sites to release different fragments. Alpha secretase's cut helps makes fragments other than amyloid beta: thus, more alpha secretase activity means less processing of APP into amyloid beta.

In the new study, researchers tested serotonin-boosting citalopram on Alzheimer's mouse models and on healthy humans. In the Alzheimer's mice, they found that the drug reduced concentrations of amyloid beta in the fluid between brain cells by up to 25% a day after treatment. Chronic treatment (28 days) reduced the appearance of new fibril-filled amyloid beta "plaques."

In the 23 healthy humans (age 21-50) who were evaluated, citalopram reduced levels of newly produced amyloid beta in cerebrospinal fluid by roughly 40% over two days of evaluation following treatment (60 mg delivered in two 30 mg doses, two hours apart). That suggests that chronic treatment might lower amyloid production enough to reduce Alzheimer's risk considerably.

Is it safe enough?

Statins aren't without side effects, but they are widely considered safe enough to be taken daily by tens of millions of healthy people to prevent heart attacks and strokes. Some doctors have suggested making them generally available without a prescription.

Are SSRIs such as citalopram that safe? Certainly SSRIs are among the most commonly prescribed drugs in the world, and millions of psychiatric patients tolerate them well enough to keep taking them for years. But could they be taken for years and even decades by psychiatrically and cognitively normal middle aged people, to lower amyloid beta production meaningfully and thus ward off Alzheimer's, without causing unacceptable side effects?

That's an issue that still has to be resolved. "Common side effects [of citalopram] in my clinical experience are nausea, loss of appetite, and hypernatremia [low sodium level]," says Ben Underwood, a consultant psychiatrist in Cambridgeshire, UK. "There is also an association with increased risk of bleeding."

In some patients citalopram causes a heart rhythm disturbance called QT interval prolongation, which in principle can lead to a serious, even fatal arrhythmia, although Underwood says that in practice the QT interval effect is "thankfully not reflected in a high incidence" of these dangerous arrhythmias. He adds that older people tolerate citalopram less well than younger people, and have a greater risk of arrhythmia; on the whole, however, the drug is "still reasonably well tolerated," Underwood says.

Another potential side-effect of citalopram, apparently more noticeable to younger people who take it, and discussed anecdotally on web forums, is a disturbance of normal sexual function. One group of researchers even found evidence recently that citalopram adversely affects sperm production. Some studies suggest that this side-effect is underreported; others that it is exaggerated.

Earlier this year, researchers reported on a clinical trial of citalopram in elderly Alzheimer's dementia patients. The purpose of the trial was not to treat the underlying disease process (lowering amyloid beta is now widely thought to be ineffective once the disease has progressed to full-blown dementia). Instead, the citalopram-commonly prescribed off-label for anxiety-was meant to reduce the agitation that is common among Alzheimer's patients and can seriously complicate patient care. The trial enrolled 186 patients, who were treated with the drug for nine weeks.

Although the drug did show an ability to reduce treated patients' agitation over the nine week trial period, it also was associated with the usual side effects. One reported side-effect that was not usual was a "[w]orsening of cognition"-which few if any healthy people would willingly risk, especially if their aim is to avoid dementia. However, that "side-effect" might not have been real. At the start of the nine-week trial and the finish, the treated patients actually scored about the same on a standard cognitive test known as the MMSE, as expected. The appearance of worsening was due to the fact that the placebo group unexpectedly scored a bit better at the end. Anton Porsteinsson, a researcher at the University of Rochester who was lead author of the study, notes that before the trial began the placebo group had a slightly lower average MMSE score than the treatment group, so the resulting shift was "most likely what we call a drift towards the mean, and not necessarily evidence of cognitive toxicity."

Porsteinsson and his colleagues used a maximum daily dose of 30 mg of citalopram, which seemed well tolerated: about 90 percent of treated subjects completed the trial-essentially the same completion rate seen in the placebo group.

However, in the amyloid-reduction trial, the 40% drop in amyloid levels was achieved with a much higher, 60 mg dose, spread out over two hours. A 60 mg dose is currently considered too high for chronic use among the elderly, particularly because of the QT interval effect-which recently prompted the FDA to recommend a maximum citalopram dose of "20 mg per day for patients older than 60 years of age."

A big question now is whether a chronic, daily SSRI dose that is low enough to be safe in terms of side-effects can also reduce amyloid beta production sufficiently. Sheline and her colleagues have started a new trial that hopefully will answer that question, this time using escitalopram (Lexapro). The latter is identical to citalopram except that it includes only one of the two mirror-image structures (stereoisomers) of the drug molecule. The idea is that the included stereoisomer is the more active one, and thus escitalopram is more selective and potent, bringing fewer side effects (not everyone agrees that this is the case).

"One of the goals of the trial is to determine the minimal dose necessary to achieve lowering of CSF amyloid concentrations," says Yvette Sheline, a researcher at the University of Pennsylvania who was lead author of the amyloid-lowering study. "If we can demonstrate that SSRIs produce a sustained effect, rather than just an acute effect, then we would plan a longer-term [Alzheimer's] prevention trial."

Porsteinsson agrees that citalopram/escitalopram has the potential for long-term use to prevent Alzheimer's. "With a little more fleshing out of the effect and the mechanism of action, this could become a prime candidate for a large prevention study," he says.