Why do so many candidate drugs for Alzheimer’s disease fail in clinical trials? The question hung over this week’s annual Alzheimer’s Association International Conference in Paris, where the post-mortem on semagacestat, a drug from pharmaceutical giant Eli Lilly that was pulled late last year, was a hot topic. Researchers are now rushing to study carriers of rare genetic forms of Alzheimer’s disease in search of an answer — and of a therapy that could save not only people affected by these ‘familial’ forms of the disease, but also victims of the more common ‘sporadic’ form that affects millions worldwide.
According to the consensus emerging from the conference, drugs fail because they are given too late, after the accumulation of amyloid-β peptides, which form sticky plaques in the brain, and other physiological changes have already destroyed the patient’s neurons and neuronal networks. At that point, even removing the amyloid-β, the aim of many of the treatments, won’t do much good. “There is already significant neuronal loss, so removing the amyloid is unlikely to reverse things,” says Randall Bateman, a neurologist at Washington University in St Louis, Missouri.
The consensus was built on a mass of data presented at the meeting showing that years, and perhaps decades, before the tell-tale signs of dementia appear, changes in ‘biomarkers’ related to the disease — including concentrations of amyloid-β peptides in the brain, concentrations of tau proteins (which are found in the neurons of patients with Alzheimer’s disease), and changes in the volume of a brain region called the hippocampus — are occurring. “What we currently recognize as Alzheimer’s disease is only the tip of the iceberg. There are a massive number of preclinical cases,” said Reisa Sperling, a neurologist at Harvard Medical School, in the meeting’s concluding plenary lecture on 21 July.
Treating patients in the preclinical or pre-symptomatic stage might help. But who should be treated? Early changes in biomarkers indicate early pathophysiological changes and are statistically convincing in population studies, but some of the data are contradictory and there is still much debate as to how the information can be applied to individual cases.
For some individuals, however, there is no debate. People carrying certain mutations in any of three genes — the presenilin genes PSEN1 and PSEN2 and the gene for the amyloid-β precursor protein APP — can be told with certainty that they will contract the disease (if they live long enough). The mutations, which all affect production of amyloid-β, first drew attention to the crucial role of the protein in the early 1990s.
Researchers in the Dominantly Inherited Alzheimer Network (DIAN), a web of 11 research centres based at Washington University in St Louis and funded with a six-year, US$16-million grant from the National Institute on Aging, are now ready to gamble that they can also show us how to prevent the disease.
Onset is early — afflicting carriers in their 40s or 50s. Because carriers normally show symptoms of the disease at around the same age as their parents did, DIAN scientists have a good estimate of how many years to onset for each person and they then track biomarker and clinical changes on that measuring stick. At the conference, DIAN scientists presented the first aggregated data taken at different centres since the programme was begun in 2008. Perhaps the most significant is that amyloid-β levels in the cerebrospinal fluid were high 20 or even 30 years before expected onset. “We can show the chronology of pathology during the long preclinical phase. That’s information that is critical for a prevention clinical trial,” says DIAN member Anne Fagan at Washington University.
“We should not be afraid of the heterogeneity.”
The early onset in familial cases also means that patients will not be suffering from some of the other ravages of age, such as cardiovascular problems, that confound researchers trying to study the progression of Alzheimer’s disease. “It is the purest form of Alzheimer’s,” says neurologist John Ringman, who heads the DIAN centre at the University of California, Los Angeles.
The ultimate aim of DIAN — therapeutic trials in familial cases of Alzheimer’s — will start next year. Drug companies and other drug-development programmes have already submitted 11 proposals. A selection committee within DIAN, which includes representatives from the US Food and Drug Agency and the National Institute on Aging, will choose the most promising. The trials will be small at first — in the range of five to twenty people — and short, just a few months, to see whether targeted changes in particular biomarkers can be achieved.
The hope is to move on to larger scale, longer-term clinical trials, but Bateman admits that even one such trial will require at least 400 people, twice the number currently enrolled in DIAN.
Familial cases account for only 1% of Alzheimer’s disease, and differ in some ways from the majority sporadic cases. So some scientists wonder if therapies developed for familial cases will work for the others. Familial cases, for example, are thought to be caused be an overproduction of amyloid-β, whereas sporadic Alzheimer’s disease seems rather to result from a failure to clear amyloid. The familial cases also show some symptomatic differences, including a spastic gait, not seen in the sporadic disease.
Still, most biomarkers are found in both types of disease, and progression of the dominant symptom — memory loss — is very similar. “At the phenotypic level [the familial cases] mimic sporadic AD. We should not be afraid of the heterogeneity,” says Martin Rossor, who heads University College London’s DIAN centre.
A trial with an even larger number of familial cases is also in the works. The Alzheimer Prevention Initiative, based at the Banner Alzheimer’s Institute, in Phoenix, Arizona, plans to have 1,500 members of the largest familial Alzheimer’s cohort — Colombians who have a mutation in PSEN1 — enrolled by the end of this year with the aim of starting a trial by the end of 2012. Unlike DIAN, they have already decided on the trial treatment, but are not making that public.
Some researchers are even starting to target pre-symptomatic patients in the sporadic population. Sperling is putting together a project called the Anti-Amyloid trial in Asymptomatic AD, which will treat some 600 older, amyloid-positive [in PET scans] but clinically normal patients. “They might have a headful of amyloid, but we want to get them before they’ve lost their neurons,” she says.
Some people are concerned that giving drugs to clinically normal patients could mean subjecting people who would never have got the disease to risky drugs. But Sperling argues that a focus on the pre-symptomatic phase in Alzheimer’s disease is no different from what happens for other diseases. “This is how we try to treat cancer, cardiovascular disease, HIV and osteoporosis,” she says. “The only exception is Viagra. That’s a good drug after symptoms appear.”