This year is pivotal for the Alzheimer's research community, with two late-stage trials due to generate clinical data that will provide a clear indication of whether targeting amyloid is an effective strategy in patients with symptomatic disease.
The two drugs, Eli Lilly's solanezumab and Pfizer/Elan's bapineuzumab, are both antibodies that bind to amyloid beta and are being studied for their potential to delay the progression of mild-to-moderate Alzheimer's disease.
The deposition of amyloid beta in the brain in the form of plaques is one the hallmarks of Alzheimer's and the 'amyloid hypothesis' suggests that these plaques are the primary influence driving the neurodegeneration seen in the disease.
There has been a string of failed phase II and phase III studies among amyloid-targeting drugs, however, and this has generated a fair degree of pessimism about the prospects for solanezumab and bapineuzumab, as well as the validity of the amyloid hypothesis.
“The amyloid hypothesis has not yet been tested in the clinic adequately, but it is still the most complete description of the disease,” said Dr Eric Karran of medical charity Alzheimer's Research UK in an interview.
Earlier phase III studies are generally acknowledged to have been 'long-shots', helping to test the amyloid hypothesis but with only modest expectations of success.
The number of disappointments for amyloid-targeting drugs is fairly substantial. Two small-molecule candidates, Neurochem's Alzhemed (tramiprosate) and Lundbeck/Myriad Genetics' Flurizan (r-flurbiprofen), failed a few years ago, although the preclinical information on them was not robust and their lack of efficacy was not unexpected.
Karran, who was formerly with Eli Lilly, worked on another much-anticipated drug, the gamma secretase inhibitor semagacestat, that also gave disappointing results in phase III trials reported in 2010. Gamma secretase is involved in the processing of soluble amyloid precursor protein (APP) into amyloid beta, so works upstream of the formation of amyloid plaques.
Semagacestat was found to cause side effects, including an increased risk of skin cancer, but more importantly was associated with worsening of clinical measures of cognition and the ability to perform activities of daily living compared to placebo. Most disconcertingly, the negative effects on cognition with semagacestat were not reversed after drug therapy was discontinued.
Meanwhile, other gamma secretase inhibitors such as Bristol-Myers Squibb's (BMS) BMS-708163 generated similarly negative results, though it has not been discontinued, while Elan decided to drop its own candidate ELND007 last year. BMS-708163 and some other gamma secretase inhibitors remain in development.
“What the semagacestat project revealed was that far more selective gamma secretase will be needed to block amyloid production without an effect on other physiological systems,” Karran continued.
The failures mean that all attention is focused on solenuzumab and bapineuzumab, which have slightly different mechanisms of action in that they attempt to bind to amyloid beta and clear it from the brain.
“What we are increasingly understanding is that the deposition of amyloid is a very early event in the disease process, perhaps as much as 15 years before symptoms develop,” explained Karran.
The delay in symptom development is thought to be a result of 'cognitive reserve' with the brain adapting to compensate for the neurological damage caused by amyloid, in the same way that function can be restored after a stroke, for example.
“One of the issues with current trials is that they are taking place in patients who already exhibit symptoms of Alzheimer's,” he added.
Karran believes there are three possible ways in which amyloid plays a role in the pathology of Alzheimer's. One possibility is that the deposition of amyloid triggers other events, and once these are set in motion then any treatment which affects amyloid is doomed to failure as amyloid itself has become irrelevant.
The second possibility is that there is a threshold level of amyloid in the brain to drive the disease forward, while thirdly amyloid beta could be the primary driver of the disease, as originally proposed. In the latter two cases, lowering amyloid beta in the brain could have a material impact on the progression of the disease.
“Most of the evidence that I have seen to date, and especially the genetic evidence, would point to amyloid beta being a trigger or threshold, rather than a driver of the disease,” Karran continued.
The current crop of phase III trials is predicated on the idea that amyloid beta is driving the disease forward and, if Karran's interpretation is correct, have a slim chance of showing an impact on cognition and activities of daily living, which are the two key measures that Alzheimer's drugs must improve in order to stand a realistic chance of being licensed.
Whether successful or not, the pivotal trials of solanezumab and bapineuzumab are hugely anticipated by Alzheimer's researchers because of the invaluable information that will be gleaned from them.
For example, if the drugs succeed in reducing amyloid burden in the brain but have no impact on cognition, something which was also seen in an earlier phase II of Elan's AN-1972, an active immunisation approach to reducing amyloid, it will provide further evidence for the trigger hypothesis of Alzheimer's.
“If solanezumab and bapineuzumab work, everyone will clearly be delighted,” stated Karran. “But if they fail, the argument will be that they were not given early enough in the course of the disease.”
The implications of that are profound, as it suggests that future studies will have to be designed in which anti-amyloid therapies are given not to patients already exhibiting symptoms, but those who are healthy but whose genetic profile renders them at risk of going on to develop the disease.
“This will mean giving a drug to healthy patients and, given that the Alzheimer's can take 15 years to develop, placing them on that therapy for a very long time,” explained Karran. “For a commercial enterprise to do that is challenging as by the time you've discovered whether your drug is effective you've probably exceeded your patent life.”
Crenezumab prevention study
A trial which will try to gauge the impact of early intervention in healthy but high-risk individuals has just been announced by Roche's Genentech unit, the US National Institutes of Health and the Banner Alzheimer's Institute.
The five-year study will examine whether Genentech's crenezumab, licensed from AC Immune and also known as MABT5102A, can actually prevent Alzheimer's disease from developing among the members of an extended family group in Colombia who carry a gene called presenilin-1 (PSEN1) that predisposes them to develop Alzheimer's in their 40s.
In the trial, crenezumab will be given in an attempt to prevent the build-up of plaques in the brains of the test subjects, thereby hopefully preventing the cognitive decline that accompanies the disease.
“This is the ultimate test of whether amyloid beta is the toxic substance that causes the disease; up to this point this is still a hypothesis that has not been proven,” according to Richard Scheller, executive vice president of research and early development at Genentech.
Even if crenezumab can have an impact on this population, however, there will still be a lot of work to be done to see if it can provide similar preventive efficacy in the more common, sporadic forms of Alzheimer's.
At the moment, there are about 30-35 clinical projects in the pipeline targeting one aspect of amyloid processing or another, so the prospects for a disease-modifying Alzheimer's therapy in the near- to mid-term could face a major setback if the lead candidates fail.
That said, quite a lot of companies are starting to focus their attention on another hallmark of Alzheimer's pathology, the production of abnormal versions of a microtubule-associated protein called tau that aggregate into neurofibrillary tangles.
Arguably the furthest ahead in development at present is BMS's microtubule stabilizer BMS-241027, which is in phase I testing and could be 10 years away from generating phase III results.
“Tau actually correlates better with neuronal loss and dementia than amyloid in Alzheimer's,” pointed out Karran, although he noted that the target was very challenging to develop and there were already deficiencies in the hypothesis analogous to those already elucidated for amyloid.
One of the main obstacles to tau R&D is that, unlike amyloid beta, there is currently no imaging agent for the protein, so researchers are heavily reliant on post-mortem studies.
Still other companies are working on serotonin receptors, based on investigative data that modulating these (notably the 5-HT6 and 5-HT4 receptors) may provide symptomatic improvement in Alzheimer's but also reduce the production of beta amyloid.
Examples of the latter group include GlaxoSmithKline's SB-742457 (in phase III) and Lundbeck's Lu AE58054 (phase II), which block the 5-HT6 receptor, and Nanotherapeutics' PRX-3140 (phase II) which acts as an agonist of the 5-HT4 receptor.
There is considerable debate about whether these drugs are in fact disease-modifying, however, or simply delay the onset of symptoms like the current generation of Alzheimer's drugs.
“If one of the current amyloid agents works we're in a much better place, but if they fail we're in a challenging place and will have to redouble our efforts in research,” said Karran.
That scenario also prompts a challenging discussion beyond scientific considerations, given that pharmaceutical companies do not have a moral obligation to develop drugs for Alzheimer's over other high-burden diseases, such as diabetes or cancer, if there is little chance of recouping their investment.
The agents currently targeting amyloid may not fit the current model in the pharmaceutical industry for funding drug discovery and development, according to Karran: “That's a huge societal issue that I don't think anyone knows how to deal with yet.”
Phil Taylor is a freelance journalist specialising in the pharmaceutical industry