Zaven Khachaturian

Even Elan and Wyeth's bapineuzumab, once hailed as one of the most promising drugs in medicine, is looking more star-crossed than starlike. Disappointing—and, to some critics, dubious—Phase II data showed only modest cognitive improvements in only one third of the patients. Still, eight Phase III trials have been bankrolled.

What all of these compounds (and 13 of the 20 other late-stage contenders) have in common is that they target amyloid beta, a peptide whose dense, sticky plaques are AD's main hallmark. Amyloid has been hypothesized as the sole cause of AD ever since its sequence was decoded in 1984. But with pharma's apparent failure to produce a disease-modifying drug with this amyloid-targeting approach, that theory, and the field itself, is in grave disarray. "Everyone is waiting on bapineuzumab," Michael Gold, Glaxo's vice president of neurosciences, told Businessweek. "If that one fails, then everyone will say we have to rethink the amyloid hypothesis."

That rethink has, in fact, long been the focus of researchers laboring in the shadows. Claude Wischik, at Britain's University of Aberdeen, has been pursuing the tau protein, whose neurofibrillary tangles are AD's other distinctive hallmark. By contrast, the metal-ions theory of Ashley Bush, who runs an AD lab at the Mental Health Research Institute of Victoria, Australia, has gone from fringe to buzz.

For years, Wischik, Bush, and other alternative theorists have railed against what they call the "church of the holy amyloid" and amyloid bias in federal funding, publication, and venture capital. "It is extraordinary that a disease of such importance has been hijacked for so long by a narrow school of thought," says Wischik, whose original claim to fame was his 1986 sequencing the tau protein. "What is particularly tragic is that this has led to a correspondingly narrow perspective in the pharmaceutical industry."

Key opinion leaders, not surprisingly, scoff at these accusations. Says William Thies, medical and scientific officer at the Alzheimer's Association: "The success or failure of the amyloid-targeting drugs doesn't mean the paradigm in Alzheimer's is changing. That's how science works. It's a naturally evolving process as we learn more and more about the underlying pathology. "

But the science isn't working, says Zaven Khachaturian, who was head of AD research at the NIH in the '80s and '90s when the correlation between amyloid deposits in the brain and the symptoms of AD was institutionalized as causal—a leap too far that nonetheless found support from important advances tagging AD-related genes. "All this provided legitimacy and excitement to the field," Khachaturian recalls. "But it also established a 'scientific orthodoxy' about the hypothesis, acceptance of ideas without questioning the premises, and dismissing opposing views."

Despite such resistance, a series of discoveries over the past decade have significantly undermined the amyloid claim. Some even argue that far from being toxic, the naturally occurring protein may be neuroprotective. This argument gained force last year when San Francisco biotech Medivation reported unprecedented benefits in Phase II results of dimebon, capped by the confounding news that this potentially disease-modifying compound appeared to raise amyloid levels in the brain. Says Khachaturian: "There is a growing recognition that the neurodegenerative processes [of AD] involve several alternate paths, but there the consensus ends. We need a new unifying theory." Lacking that, he fears that "drug companies will walk away from Alzheimer's as untreatable."