Bringing Alzheimer’s in from the Cold
With big pharma seemingly pulling out of the field, what hope remains for novel treatments for Alzheimer’s disease? Here, we gain insight from a small but pioneering company and researchers in the Alzheimer’s research space to discover that – with the support of the wider community – there are significant reasons to be hopeful.
You’d be forgiven for thinking that the pharma industry has completely abandoned Alzheimer’s disease (AD) research. At the start of 2018, Pfizer made the decision to terminate its neuroscience discovery programs, leading to a significant backlash in the media with dramatic (and negative) headlines lamenting big pharma’s exit from the field. Over the past few years, numerous pharma companies have seen failures of once promising Alzheimer's drugs in their pipelines, and there hasn’t been a new drug approved for AD in over a decade.
Is there truly no hope left? Pfizer’s announcement in January wasn’t all doom and gloom – the company added that it would be creating a venture fund to invest in biotech companies conducting promising neuroscience research. And research in both company pipelines and academia continues. Here, I speak with some of the people refusing to give up on Alzheimer’s.
Untangling Alzheimer’s
There is no getting around the fact that developing treatments for AD is difficult. The disease is characterized by the development of two distinct pathologies in the brain:
- Senile plaques composed of β-amyloid located predominantly between neuronal cells that accumulate in normal aging without dementia.
- Tangles arising from tau aggregation, which forms small toxic pathological oligomers and then filaments within neuronal cells. These aggregates are highly correlated both with clinical dementia and with imaging abnormalities commonly used for diagnosis.
We still know surprisingly little about what triggers the aggregation of these proteins, but we do know that the process starts decades before any clinical symptoms of dementia appear. There is also little understanding of how abnormal processing of the proteins that give rise to β-amyloid plaques and tangles are linked, and which comes first. There are robust associations between clinical decline and the largely homogenous spread of the tangle pathology throughout the brain. It has been known for some time that the association between clinical decline and β-amyloid pathology is much weaker than for the tau aggregation pathology. This has been confirmed in recent years using specific ligands that permit the two forms of pathology to be visualized in living patients using PET imaging.
The only real success to date has been with the development of symptomatic treatments that modify cholinergic and glutamatergic neurotransmitter activity. But the last new approved treatment for AD was 15 years ago, so there has been a major failure in the field to develop fundamentally novel approaches, despite the pressing socio-economic need. Symptomatic treatments provide only modest and temporary relief from ongoing clinical decline in cognition and global functioning. Their widespread availability may also be creating problems for the development of novel approaches – because symptomatic treatments have become the “standard of care,” it is difficult to conduct clinical trials of drugs except as add-ons to these treatments. Moreover, the chronic brain stimulation produced by these drugs may actually interfere with the action of drugs that address the underlying pathology or other novel symptomatic treatments.
A further factor that makes the development of treatments difficult is the nature of the decline that patients undergo, which is slow and variable – whether treated or untreated with currently available drugs. Clinical trials need to be large and long, even if the treatment effect is small.
In my view, treatments that prevent the spread of tau aggregation pathology offer a very attractive avenue of attack. It is now known that the spread is mediated via prion-like processing that converts tau protein into an infectious particle – one that resists proteases and is able to seed further tau aggregation in previously healthy neurons.
Three Decades of Targeting Alzheimer’s
The story of TauRx began in 1984 when I embarked on my PhD studies at Cambridge University in the UK with Martin Roth and Aaron Klug at the Laboratory of Molecular Biology. I was set the task of identifying the compositional structure of the neurofibrillary tangles that form in the brain cells of people with AD. Martin was a psychiatrist who was the first to correlate these tangles with AD. Using the fiber structure analysis techniques that had earned Aaron the Nobel Prize in 1982, I found that the tangles were composed of a repeating subunit arising from de novo assembly, and were not the product of a simple collapse of the existing fibrous backbone of the nerve cell.
In 1988, I discovered that a certain class of synthetic small molecules could dissolve these tangle fibers in a test tube. One of these small molecules, methylthioninium chloride or methylene blue, was already in clinical use for other indications, which opened up the possibility of developing a novel disease-modifying treatment for AD targeting tangles. It became the focus of my research for the next 30 years, initially in Cambridge and then at the University of Aberdeen, where I moved with my team in 1997.
Our work progressed at a scientific level with the wonderful help of colleagues in Aberdeen with strong synthetic chemistry skills, medical imaging expertise, brain neuropharmacology and mathematics. These collaborations, which still continue, represent the fundamental scientific bedrock of TauRx. The first clinical trial that we undertook used a novel version of the molecule to treat AD based on inhibition of abnormal aggregation of Tau protein. It showed a strong clinical efficacy signal as a monotherapy, supported by brain imaging. However, this form of the molecule was found to have limitations in terms of absorption and tolerability and we have since gone on to develop an improved form of this molecule, known as LMTX, which TauRx has taken into Phase III trials.
The study involved over 1,900 patients with mild to moderate AD and with behavioral variant frontotemporal dementia, using over 250 study centers around the world. The two studies involving patients with AD produced remarkably consistent results. Those patients taking LMTX as monotherapy (which is to say, without other drugs approved for the treatment of AD) showed statistically significant and clinically meaningful benefit in terms of cognitive and functional clinical outcomes, accompanied by MRI evidence of reduction in the rate of progression of brain atrophy. However, no such benefit was seen in patients taking LMTX as an add-on to standard symptomatic AD treatments. Studies are under way to better understand the mechanism, and to determine whether it is a class or a molecule effect. We have also launched the LUCIDITY study, which uses FDG-PET imaging to confirm the potential of LMTX in delaying progression of disease pathology in the brain in patients with mild AD.
A disease of the future
As human lifespans increase and the world’s population ages, the incidence of AD is predicted to reach epidemic proportions. It is estimated that there are 47 million people currently living with AD, and that this figure will increase to 75 million by 2030 and to 132 million by 2050. If left unchecked, this disease has the capability not only to devastate the lives of patients, families and caregivers but also to have a major financial impact on public healthcare systems across the world. As both a doctor and the co-founder of a company, I believe that any research that contributes to the understanding of AD is vital and should be continued.
Though news headlines tend to be dominated by the failures of large-scale clinical trials to achieve their primary endpoints, even failed studies generate useful data that offer new insights into disease pathology, biomarkers and clinical progression. The AD research community has more useful data – and more sophisticated data analysis tools – than ever before. And from the point of view of market opportunity, the scale of the problem to be solved only increases.
Nevertheless, the field remains challenging, for large and small companies alike. Larger companies benefit from access to greater resources and a more focused effort on research programs and clinical trials. On the down side, the corporate decision-making that deploys large clinical trial resources is understandably cautious and must find its bearings with reference to prevailing scientific opinion. If prevailing opinion happens to be on the wrong track then large corporations can lose their way. The discovery in the 1990s of mutations in the amyloid precursor protein (APP) gene led to an almost universal adoption of the amyloid cascade hypothesis as being the prime driver of AD. New product development in multiple companies has, therefore, been focused on agents able to clear β-amyloid plaques and prevent ongoing formation. There have been promising results in the laboratory and in animal models, but the positive effect on pathology has not been reflected in clinical symptoms in humans. Though I believe that the detection and tracking of β-amyloid plaque build-up may be useful for diagnostic purposes, this pathway seems to represent an inherently inefficient approach to the treatment of the disease.
Fortunately, other therapeutic approaches are under investigation. Smaller companies are more nimble than larger entities and many are focusing on impressive science. Such companies are less reliant on prevailing opinion and able to take scientific risk. Small companies (including my own), however, are faced with the problem of the cost and resource effort required to undertake large global clinical trials. The key underlying challenge lies in Phase III clinical trial design and the unavoidable heterogeneity of the AD patient population. In around 99 percent of cases, AD is diagnosed in septuagenarians, at which time a number of other co-morbidities are likely to be present. The costs and risks associated with undertaking large-scale clinical studies in this population are significant, particularly if the drugs involved have modest treatment effects.
Hope remains
The number of products in clinical development that target the tau tangle pathway has increased markedly in the past two years, reflecting the fact that the research community is embracing this potentially more effective approach to treating the disease.
With greater collaboration and open-mindedness to novel research – and with greater investor focus on this area of medicine – we will go further, faster. If the research community had supported the tau-tangle approach and other avenues for fighting AD in the 1990s rather than simply focusing on the amyloid hypothesis, would we be looking at a very different world today?
Claude Wischik is Co-Founder and Executive Chairman at TauRx Pharmaceuticals.
The Research Conundrums
The big question facing AD research today: what explains the failure of numerous clinical trials? In my view, the greatest challenge for both researchers and companies working on AD is testing subjects with experimental therapeutics prior to or near the time of onset of their amnestic and cognitive symptoms. What the numerous failed Phase II and III AD trials have in common is that they attempted to treat subjects with mild-to-moderate symptomatic AD. In some cases, the agents themselves were weak, toxic or otherwise flawed (e.g., R-flurbiprofen, tramiposate, solanezumab and semagacestat), but these and other agents were also administered too late in the degenerative process. Like atherosclerotic cardiovascular disease, AD is a chronic, slowly progressive disorder in which potential disease-modifying agents largely need to be administered prior to onset of symptoms. The other great challenge now is the need to have the first drug achieve FDA/EMA approval so that a second agent can be tested with it to move toward combination therapy, as is usually required for other chronic diseases.
Another frustrating setback has been the failure of certain agents because their trials included some (or many) patients who did not actually have AD – amyloid PET imaging was available, but not used in several of the trials to screen and select the right patients.
These challenges and disappointments have led some companies to decide to withdraw entirely from clinical research on AD and other neurodegenerative diseases. Fortunately, other companies are pushing ahead. We owe it to the world’s AD patients and society as a whole not to give up, but to instead put more resources and good ideas into AD preclinical and clinical research. This added support is now coming from both biopharmaceutical companies and from the US National Institutes of Health, AD-directed foundations and major philanthropists. The costs are high, but once a single successful agent is found, it will pry open the floodgates for much greater investment in AD translational research.
I’m excited about agents coming through pipelines that efficiently target soluble Aβ oligomers (oAβ) in the brain, because a great deal of preclinical evidence suggests they are the principal pathogenic moiety that initiates the neurodegenerative process. Some of the monoclonal antibodies now in trials can bind and clear these diffusible oligomeric species (e.g., Biogen’s aducanumab and Roche’s crenezumab, among others), and that is encouraging. I am also excited about the advent of tau immunotherapy (vaccines and monoclonals). As tau accumulates prior to symptoms but after oAβ, combining an Aβ -lowering agent with an anti-tau agent is an attractive idea.
Delving deeper into the mechanisms of microglial alteration in late-onset AD is also important, although I think specific compounds that safely modify this ubiquitous feature of the AD cascade are not yet near.
Learning from failure
For companies that feel AD research is not viable, at present there are still contributions to be made – particularly with the vigorous sharing of natural history data and any other archived data, such as those from failed trials. Placing all clinical trial results that are not intended for an NDA or BLA into a shared public database will allow pooling of data-rich archives and improve statistical power in natural history studies and help us understand the details of progression from placebo cohorts. Companies not currently working on specific AD therapeutics should also be encouraged to contribute financially and scientifically to the global effort to advance translational research on AD by supporting public-private consortiums and AD-oriented philanthropy.
Despite what the headlines might say about the current state of AD research, I don’t think AD patients have been abandoned by any means – there are a lot of good preclinical and clinical studies underway or being contemplated. And one success will encourage those who’ve left the field because of the challenges to reassess and possibly return. It is unfortunate but understandable when companies choose to focus on other therapeutic areas after pipeline failures. But we must remember that there are specific reasons why AD clinical trials have not seen success so far. To cite some examples, semagacestat (a putative γ-secretase inhibitor from Eli Lilly) had a half maximal inhibitory concentration (IC50) for Notch processing that was about the same as its IC50 for amyloid precursor protein cleavage (i.e., a low therapeutic index), thus leading to significant adverse events from inhibiting signaling by Notch – and probably that of other normal substrates. Solanezumab showed a small (~15 percent) slowing of cognitive decline, but its predilection for binding Aβ monomers (not implicated in AD cytotoxicity) over Aβ oligomers made it unlikely to yield a significant clinical signal. Pfizer’s bapineuzumab was a potent anti-oAβ antibody, but its induction of amyloid-related imaging abnormalities with edema (ARIA-E) was a first for the field and forced the antibody to be dosed too low and ultimately abandoned. Interestingly, aducanumab has moved forward despite some occurrence of ARIA-E because the field has come to realize that the latter is a sign of moving amyloid out of the brain.
In short, more rigorous preclinical testing in iPSC-derived human neurons and in two or more rodent models of AD is needed to obtain a robust initial dossier before advancing an agent into clinical trials. The “doom and gloom” about the failure of AD trials needs to be assessed scientifically, not emotionally – there are specific reasons for the failures, and we all need to learn from those lessons.
Dennis Selkoe is Coates Professor of Neurology at Brigham and Women's Hospital and Harvard Medical School, and former chair of the external neuroscience advisory board for Pfizer.
A Charitable View
With John Davis, Chief Scientific Officer of the Alzheimer’s Research UK Oxford Drug Discovery Institute.
On the institute
The Alzheimer’s Research UK Oxford Drug Discovery Institute (ARUK-ODDI) was created with funding from Alzheimer’s Research UK, and is one of three drug discovery institutes sponsored by the charity. We are located at the University of Oxford and focus on novel targets for treating dementia. We are lucky to not only have staff who are dedicated to finding new ways to treat neurodegenerative diseases; we are also surrounded by academic scientists passionate about their research and applying it to the fight.
On the challenge of AD
Developing drugs in such a disease area is complex and painstaking. Originally, familial genetics and histopathology identified some key targets, and these avenues have finally come to fruition in the recent trials of secretase inhibitors and antibodies targeting species of beta-amyloid. A particularly exciting recent development has been the advent of the omics era, which has enabled the profiling of genetic risk factors and transcriptional, proteomic and metabolomic changes. The linking of changes in the immune system with AD will certainly spawn a series of approaches to be tested as treatments. It’s important to remember that every experiment we perform, no matter how successful or unsuccessful the result, both educates us and raises further questions. The trials using anti-beta-amyloid antibodies have produced results that have been less efficacious than hoped for, but we have learned a lot about the heterogeneity of AD patients – as well as the importance of considering at what point in the course of a disease to treat with a particular type of drug.
On support from pharma
The whole landscape for drug discovery is changing – and not just in the central nervous system space. Pharma companies engaged in drug development for many different diseases are realizing that it may be better to concentrate on clinical development and allow smaller, nimbler and more focused biotech teams to concentrate on the discovery phase. Hence, some pharma companies that may appear to have withdrawn from discovery for AD have not abandoned the field, but instead have simply refocused, and are still active in clinical development.
Projects require different types of support at different phases and pharma companies are already incredibly supportive. Of course, there is never a bottomless pit of resources, and pharma has chosen priorities that may not align with any given academic research group’s specific interest. However, significant progress has been made and there are now many different public and private funding models available to an academic with a well-defined plan. One area where pharma could further help groups like the ARUK-ODDI is in the sharing of assay methodology that is not necessarily of critical proprietary value, or making compound collections available via open access, or at least with as much freedom to operate as possible.
Another current issue is that research groups and pharma generate large amounts of valuable data that reaches a limited audience, or is leveraged minimally. Publication via journals is a system that provides valuable quality control and organized publishing, but is otherwise an outmoded route of communication. More “open access” sharing of resources and data would both minimize time delays due to publication procedures or contract negotiations, and also reach many more researchers. The coordinated placing of appropriately validated methods and data onto public portals would help accelerate research.
On the future path of AD research
Despite disappointing recent trial results, increased awareness, increased funding and new technologies have produced an encouraging landscape for future drug discovery for AD. We must keep up the momentum and continually remind ourselves that the treatment of dementia represents one of the world’s largest medical needs – and deserves very considerable investment in time, resources, and sweat.
I have an extensive academic background in the life sciences, having studied forensic biology and human medical genetics in my time at Strathclyde and Glasgow Universities. My research, data presentation and bioinformatics skills plus my ‘wet lab’ experience have been a superb grounding for my role as a Deputy Editor at Texere Publishing. The job allows me to utilize my hard-learned academic skills and experience in my current position within an exciting and contemporary publishing company.
