The amyloid hypothesis isn't dead, but it seems to be limping a bit in the race for an Alzheimer's cure.
Some researchers who predicted 5 years ago that an antiamyloid disease-modifying therapy was imminent are now reevaluating that optimism—including the geneticist who first suggested the pathologic link between amyloid plaque deposition and Alzheimer's disease.
“I accept that criticism of myself; it's definitely what I thought,” John Hardy, Ph.D., said in an interview. “Everything is taking a lot longer than I thought it would, there's no question about that.”
In 1991, Dr. Hardy, a professor of neuroscience at University College London, postulated that β-amyloid deposition was the root of a pathologic cascade that resulted in Alzheimer's disease. The concurrent discovery that a mutation in the amyloid precursor protein (APP) gene caused early-onset Alzheimer's, coupled with the association of plaque deposition and early Alzheimer's in Down syndrome patients, added weight to the theory (Trends Pharmicol. Sci. 1991;12:383–8). A new research boom was born.
The protein was a near-ideal therapeutic target because there are many ways to get at it: immunotherapy to break up existing plaques, compounds to prevent formation of the sticky bamyloid-42, and antiaggregants to prevent the protein from clumping into neurotoxic plaques.
Dr. Rachelle Doody, director of the Alzheimer's Disease and Memory Disorders Center at Baylor College of Medicine, Houston, said that the failure of antiamyloid drugs illustrates not a failure of the theory, but the failure of specific drugs and possibly the failure of drug companies to follow a comprehensive and logical phase II plan.
Dr. Doody's argument reflects the disappointing results of the first phase III trials of antiamyloid therapies. Tramiprosate, a β-amyloid antagonist, was the disappointment of 2007; tarenflurbil, a gamma-secretase modulator, this year's downer. And although bapineuzumab, a passive immunotherapy, made it to phase II last summer, positive findings in its phase II trial were slim. A post hoc analysis showed that some patients with mild to moderate Alzheimer's, with no genetic risk factors, had cognitive improvement after getting the vaccine. Apparently, the finding was enough for Elan Pharmaceuticals Inc., and Wyeth Pharmaceuticals, although maybe not for Dr. Hardy.
“The data right now are neither positive nor negative. At this point, the only thing we can say about bapineuzumab is that it's not going to be a miracle therapy,” he said.
A long-term follow-up study of patients enrolled in the early AN-1792 immunotherapy trial “doesn't look great for amyloid, either,” he said. The AN-1792 trial was halted early, in 2002, when some of the patients developed encephalitis after getting the vaccine. The follow-up, published last summer, showed that the vaccine did clear plaques, but that clearance didn't affect cognition or survival. In fact, the authors said, “Seven of the eight immunized patients who underwent postmortem assessment, including those with virtually complete plaque removal, had severe end-stage dementia before death” (Lancet 2008;372:216–23).
Dr. Hardy doesn't think that slow progress on antiamyloid drugs negates the theory's basic truth, though—at least for a subset of patients. “There's no doubt at all that the amyloid hypothesis explains the disease in families with mutations of the amyloid precursor protein and presenilin genes. A much more open debate is whether the same process is at work in the typical Alzheimer's patient.”
But drug companies must target this larger population in order to create a financially successful therapy, and the lack of progress has them fidgeting, Dr. Hardy said. “Every drug company is worried now and wondering if they should widen to other therapies, including tau-targeted drugs. And to this I say, ‘Yes, of course you should have other strings for your bow.’”
The essential mystery of amyloid further complicates things, Dr. Hardy said: The protein has not yielded up all its secrets, despite years of research. “The thing that keeps me up at night is that we don't really know if amyloid has a function. It could be that amyloid is a response to vascular damage. We all ignore the fact that amyloid deposition occurs to a large extent in the vasculature. There must be a reason for this, and it could have something to do with vascular repair.”
That worry also plagues Mark A. Smith, Ph.D., a professor of pathology and Alzheimer's researcher at Case Western Reserve University, Cleveland. “We have said for a long time that amyloid is doing something important in the brain. It could be acting like a vascular sealant in areas of injury. It forms structural scaffolding for blood vessels, and if you start getting rid of that scaffold, you'll see problems in the blood-brain barrier.” This reaction probably caused the brain inflammation seen in the AN-1792 trial, he said.
Dr. Smith, who is a paid consultant for several companies investigating non-amyloid related therapies, is among a minority of researchers who resist the amyloid theory, although the overwhelming focus on amyloid has virtually drowned their opinion. In fact, he said, the amyloid research momentum is so strong right now that only more high-profile failures will begin to temper it.
“People still can't believe it's not working, and they're waiting for the results of the phase III vaccine trial,” as well as new data on β-secretase inhibitors, theorized to reduce the buildup of plaque-forming AB-42. “I think at this point, the research community is so totally invested in amyloid that we need to either get something else that works or have an honest, sober, the-party's-over discussion of why amyloid-targeted therapies are failing—and get if off the table.”
In an interview, Dr. Doody said that “Companies want their drug to be labeled as a disease-modifying agent as soon as possible; the implication is that it can then be priced at a higher rate. And because they are going for that, they are designing phase II trials that are long and costly but don't give them all the information they need.”
Ideally, by the time any agent, including a potential disease-modifying compound, finishes phase II, there should be clear evidence that it is both safe and effective in the primary end point. “Neither tramiprosate nor tarenflurbil had a clear signal in phase II, and neither did bapineuzumab, although it at least had some signal. Another phase II study for bapineuzumab would have been nice to further clarify this proposed subpopulation of interest,” she said.
Companies could also modify their research track to prove first that a drug confers symptomatic benefit, and then examine its possible disease-modifying properties. That is the path Medivation Inc. is following with dimebon—the only bright note in late-stage clinical trials this year. The obscure Russian antihistamine, thought to boost mitochondrial function, succeeded where the antiamyloids failed, significantly improving cognition, behavior, and function in Alzheimer's patients, although it did not modify disease progression.
“Dimebon probably is a disease-modifying drug, but proving this requires long-term studies,” said Dr. Doody, primary investigator on the phase II trial. “But many pharmaceutical companies fear that a drug will be priced too low if they go for symptomatic approval first without the disease-modifying work up front.”
Dr. Doody follows research on dozens of potential Alzheimer's drugs, only a portion of which are antiamyloids. But she agreed that these compounds grab the lion's share of attention. “The amyloid story gets articulated over and over again because a lot of people in academia feel most comfortable with a story that's already been told. But there's no a priori reason that any one of these approaches should work better than another.”
In fact, researchers might be looking at antiamyloids through the wrong end of the lens, said Dr. Marwan Sabbagh, chief medical and scientific officer of Sun Health Research Institute, Sun City, Ariz. Rather than a one-step cure, the compounds may be best used in primary prevention. “The problem is, we may be approaching it too late,” he said in an interview. “By the time you clinically manifest dementia, it might be too late for the drugs to help, even if they clear the plaques.”
Using antiamyloids to remove the protein before it aggregates might interrupt the pathologic cascade that Dr. Hardy envisioned, decreasing the excitotoxicity, the tangles, and the inflammation that kill neurons.
“The ideal future for an amyloid-based approach would probably be a combination of immunotherapy to break up existing plaques, and secretase inhibitors to prevent the formation of additional plaques,” said Dr. Sabbagh, who is an investigator on the phase III bapineuzimab trial. Putting this to practical use will require some big advancements in early detection—probably imaging techniques—that can identify patients with a small plaque burden and little or no cognitive impairment. “The question is: Will we be able to advance the field enough to use them in that manner?”
FULL TEXT
