Plaque-Attacking Antibody Fails to Help Alzheimer’s Patients


by Jim Schnabel

August 9, 2012

An antibody-based drug that reduces amyloid beta (Aβ) “plaques” in the brain and was once touted as a potential blockbuster therapy did not help Alzheimer’s patients in two large clinical trials, according to the drug’s corporate sponsor. Further details of the studies have not yet been released; however, the top-line findings suggest that plaque-clearing is simply the wrong strategy in patients who already have dementia. Aβ-reducing strategies may yet show some effect when applied before dementia has set in; or when targeted to more toxic, non-plaque forms of Aβ. Even so, this latest clinical failure of a potential Alzheimer’s therapy underscores the fact that the research community still doesn’t know what drives the dementia phase of the disease.

“We don’t know exactly what we should be targeting,” says John Hardy, a professor of neuroscience and long-time Alzheimer’s researcher at University College London.

The drug in question, bapineuzumab, is a solution of antibodies that fasten tightly to Aβ—the primary constituent of the insoluble brain deposits known as amyloid plaques. Infused into patients’ bloodstreams, these antibodies are meant to filter slowly into the brain, peel off plaque material bit by bit, and dispose of it within waste-gobbling cells called microglia. Studies of bapineuzumab-treated patients as well as patients treated with an active vaccine against Aβ have shown that these strategies often do reduce the amount of amyloid plaque in the brain. The antibodies also have been shown to clear Aβ plaques in genetically engineered “Alzheimer’s mice.” However, the plaques in such mice are not accompanied by the profound neurodegeneration seen in Alzheimer’s. For that and other reasons, researchers have never been certain whether such plaques are a cause of the disease or just a byproduct.

The two failed trials of bapineuzumab had enrolled a total of about 2,400 patients. One group had a high-risk Alzheimer’s gene variant, known as apo-E4, and the other did not. Based on an earlier, smaller study, the drug’s sponsors had hoped to see an effect of bapineuzumab at least in the non-E4 group, which on average had less plaque buildup.

Why bapineuzumab failed to show an effect in these trials is a question with many possible answers. One is that the antibodies do reduce the concentration of harmful forms of Aβ in the brain, but just not enough to slow the progress of the disease. Sam Gandy, an Alzheimer’s researcher who directs the Center for Cognitive Health at Mount Sinai School of Medicine in New York City, notes that bapineuzumab in previous studies has reduced plaques by only about 15-25 percent, as measured using amyloid-sensitive brain scans. That might not have been sufficient to change the course of the disease. The current trial sponsors are expected to release brain-imaging data showing patients’ plaque reduction amounts later this year. Data on cerebrospinal fluid levels of tau, another Alzheimer’s associated protein, are also expected later this year. “If this treatment took patients in the right direction, we would expect tau levels in the cerebrospinal fluid to have dropped a bit,” Hardy says.

The wrong target?

Plaque-clearing strategies originated in the mid 1990s, but since then, researchers have begun to focus on smaller, soluble, harder-to-detect aggregates of Aβ, known as oligomers, as the more harmful form of the protein. Bapineuzumab may bind to Aβ oligomers as well as plaques—a mouse-antibody version has been reported to do so—but its clinical failure might have been due partly to a relative lack of effect against this more toxic form of Aβ. Ideally, researchers in clinical trials would be able to see whether anti-Aβ treatments such as bapineuzumab lower levels of oligomers in the brain, but the technology to do that doesn’t yet exist. “Even today, we have no way of tracking oligomeric Aβ in the brain [in living patients],” notes Gandy.

Even the clearance of oligomeric Aβ may come too late to help Alzheimer’s dementia patients, in whose brains other, late-stage degenerative processes may already be at work. For example, there is evidence that tau aggregates—long linked to Alzheimer’s and other neurodegenerative diseases—help to kill brain cells during the dementia phase of Alzheimer’s. This “tauopathy” process may be triggered by Aβ, but might also be able to keep going, in an infection-like manner, even after Aβ levels are reduced. “If the fire has already started, removing the match won’t do any good,” says Hardy.

Still some reasons for hope

Given the large and growing population of Alzheimer’s patients, an Alzheimer’s drug that really works could earn its maker a vast sum of money. Yet a corporation also has to risk a vast sum of money to develop such a drug. Bapineuzumab’s sponsors, for example, are now writing off about a half-billion dollars in clinical trial costs and related investments. With so many clinical failures to date, and no scientific consensus on what drives Alzheimer’s once dementia has set in, many drug companies are going to be reluctant to take such risks.

Still, the clinical side of Alzheimer’s research has not been a complete failure. Solanezumab, an anti-Aβ antibody solution that is similar to bapineuzumab, has just completed a large scale clinical trial. There is no strong reason to suppose that its Aβ-binding effect or its clinical impact is much different than bapineuzumab’s, but until its trial results are released later this year, researchers still have room for hope.

More importantly, scientists have secured the funding ($100 million) for a medium-scale trial of another anti-Aβ antibody—crenezumab—this time to prevent or delay dementia in cognitively normal people. In principle, the prevention or delay of dementia requires only a reduction of Aβ, and is thus more achievable than the reversal of established dementia. The drug will be tested in people who carry a gene mutation that causes Aβ aggregates to be overproduced and normally triggers a form of Alzheimer’s dementia in early middle age. Results of the trial should be available in about five years. “We really need to have these trials in people in whom we know the Aβ cascade is operating,” says Hardy. Success could lead to similar trials in ordinary elderly people who don’t have an Alzheimer’s-causing mutation but show early warning signs of the disease on brain scans and blood tests.

Yet another hopeful development concerns IVIG, an FDA-approved mix of human antibodies that is pooled from donor blood and typically used to treat immune-related conditions. Small trials of IVIG in Alzheimer’s patients have shown a stabilization of cognitive signs—the treated patients tend to stop getting worse—as well as a reduction of brain shrinkage and other secondary measures of dementia. IVIG’s maker, Baxter Laboratories, is now sponsoring a trial in about 400 patients, the results from which should be available in early 2013. Positive results could lead to swift FDA approval of IVIG as an Alzheimer’s treatment, since it is already approved for other uses. “One of the possible advantages of IVIG that has been proposed is that it recognizes oligomeric Aβ,” says Gandy. The IVIG mix may also contain antibodies against tau aggregates, as well as antibodies that serve to quiet brain inflammation, which is a known cause of damage in Alzheimer’s.

IVIG would have to be administered intravenously on at least a monthly basis, at a few thousand dollars per dose, and its supplies—which depend on human blood donations—probably could never be adequate to serve widespread demand for Alzheimer’s treatment. But success for this product would at least trigger a race to understand its mechanisms of action, which could then be replicated by cheaper, more easily produced drugs.

Finally, a study published earlier this year found that another FDA-approved drug, bexarotene, a distant relative of vitamin A that is used for treating a rare skin cancer, works well at clearing Aβ from the brains of mice. The drug seems to have its effect by boosting the production of apo-E, a protein that normally carries loose fat molecules for disposal within waste-gobbling microglial cells. Because Aβ tends to cling to fat molecules, it too is borne along for disposal by apo-E. Small trials of the drug against Alzheimer’s are planned, and if any of these shows a positive effect, widespread “off label” prescribing of the drug in Alzheimer’s and pre-Alzheimer’s patients could follow. In principle, this could happen even without pharmaceutical company sponsorship, large-scale clinical trials, or specific FDA approval for Alzheimer’s. Bexarotene’s patent is due to expire, clearing the way for generic substitutes, in 2016.