A Vaccine for Alzheimer’s Disease? An Interview With AC Immune’s Prof Andrea Pfeifer

Swiss biopharma AC Immune has made targeting Alzheimer’s disease, the leading form of dementia, a top priority. The company’s numerous candidates and platforms target several different proteins and pathways thought to have a central role in Alzheimer’s pathology. 2020 has brought mixed news for AC Immune. In July, the company announced that their candidate tau vaccine, ACI-35.030, had shown “encouraging” safety data at a lower dose and was to be progressed to a higher-dose group. However, September brought the disappointing news that their anti-tau antibody, semorinemab, did not meet its primary efficacy endpoints in a Phase 2 trial. To discuss AC Immune’s work, Technology Networks spoke to CEO Prof. Andrea Pfeifer.

Ruairi Mackenzie (RM): Why should we be targeting tau in Alzheimer’s and why should we be trying to target it with a vaccine-based approach?

Andrea Pfeifer (AP): This is a question which goes very much to my heart, because we were one of the first companies in 2007 starting to work on tau and at that stage we were not sure how tau propagates from one cell to another and how it could propagate from one half of the brain to the other.

Since then, of course, the knowledge base has dramatically changed. Today, the two major, and I’m using my words very carefully, the two major targets linked to Alzheimer’s are definitely amyloid-β and tau. Despite the importance that is increasingly given to tau, amyloid-β still has a role in inducing the disease. There is very little or no disease if amyloid-β is absent, so that’s a very important thing.

The treatment windows when you can actually treat amyloid-β or tau are supposedly very different. In fact, the window for therapy of amyloid-β alone is probably the prevention stage in very early Alzheimer’s. The window where tau plays a role is probably from very early or even pre-clinical Alzheimer’s to a very late stage. Tau actually starts in the cells, in contrast to amyloid-β, which is an extracellular protein. The very first step in pathology happens in the neuron. Then you have this aggregated tau that can leave the neurons and migrate from one neuron to another neuron in an almost viral-like process.

This is really how this disease gets spread. Now the positive aspect is that if you have a molecule, which we have, to stop the intracellular, intraneural process, then that is the first, best step of intervention. For antibodies, there is still a large window, and this is when the tau aggregates become extracellular.

In this migration process, the antibodies really should stop spreading and subsequent seeding in the healthy neurons. Now the reason why we consider tau important is that in human studies, due to the availability of tau tracers, we can really look into the tau pathology and how it progresses from one part of the brain to the other part of the brain.

Linked to that, we can actually look at the cognitive decline. In fact, if there is one thing with Alzheimer’s disease progression where you have a correlation between biomarker and clinical outcome, it’s this tau pathology. What we recently learnt is that the tau pathology is not just starting when you see a signal in the brain by tau positron emission tomography (PET), which is normally linked to the symptomatic onset of the disease. The work of Oskar Hansson, which got a lot of recognition earlier this year, showed that soluble tau, in the plasma and cerebrospinal fluid, can be detected ten to twenty years before even symptoms start and before you have this tau PET signal inside the brain.

What this means is that intervention as we do today, meaning early-stage symptomatic Alzheimer’s treatment, might actually move to the prevention stage because it would allow you, and this is incredibly important, to use the biomarker to identify people at risk. You asked me about the tau vaccine, why I consider this project so important. These new plasma biomarkers would allow us to identify Alzheimer’s in a normal routine exam in the doctor’s practice. We could see if these tau biomarkers are there in a certain range and this would allow you to preventively vaccinate people, so they don’t get the disease.

This would not treat but prevent Alzheimer’s and honestly it would be more important than HIV treatment or oncology treatment in terms of impact on society.

RM: Tau has non-pathological forms too. how can we target the right kinds of tau with a vaccine in this way?

AP: To be honest, we do not know exactly which species are responsible for this spreading mechanism of the disease. Obviously, this is why our antibody is targeting all the tau species with a certain confirmational change that implies pathology. It is not yet fully defined, however there are at least three clusters in the molecule which are important. One is the N-terminal end, which is where most antibodies in late-stage development at the moment are targeted.

Then there’s the midterm domain and then there is the C-terminal. Our vaccine goes to the C-terminal end in one position (which is 396 for the experts), and the C-terminal end has been quite widely linked to the disease progression. Obviously when we developed the vaccine, we looked in every possible animal model and, alongside Janssen Pharmaceuticals, Inc., into human cellular models. What was extremely convincing for the vaccine was that it can really inhibit the spreading of the disease in vivo in animals and in all sorts of human cell models.

We have really tested both the qualitative and quantitative response of our vaccine and we are quite confident we have a very powerful vaccine here.

RM: Vaccination for Alzheimer’s disease isn’t new. Previous vaccines targeted amyloid.  Why do you think your compound will be more successful?

AP: I think one reason why so many studies failed was that they did not have the right diagnostic tools. It’s absolutely unimaginable today to start an Alzheimer’s study and not pre-screen the patients for the presence of amyloid-β. The first studies, for example the solanezumab studies, did not have this inclusion criteria. It was not because the researchers didn’t know they should do this, these tools were simply not available. The big advantage we have with tau is that the therapeutic tools and the diagnostic tools have become available, so we have already tau imaging to do a better clinical trial.

It’s very simple. The earlier you diagnose, the less neuronal loss you have and the easier it is to re-establish the function of the brain. This is why I’m so excited about this new data tracking phospho-tau-217 before symptoms arise because this is potentially when all the neurons are still there. It should allow us to really prevent the onset of the disease.

What we are learning now is that in later stage Alzheimer’s you have co-pathologies caused by alpha-synuclein, the protein in Parkinson’s disease, or by TDP-43, the protein in ALS which are actually co-present and do influence the progression of the disease. Again, it comes back to the fact that we didn’t have biomarkers available for alpha-synuclein and TDP-43, allowing us to look into the co-pathologies.

That brings me to my last point. I think realistically I do believe that if you target amyloid-β early enough, preventatively, if you target tau in the window of very early until late disease, and if you then have the possibility to add, according to the status of the brain, other treatments for amyloid-β, alpha-synuclein and TDP-43, I think this is the moment when you say you conquer the disease.

If you look at cardiovascular disease, the best example for me is cholesterol. Cholesterol is not a disease factor, it’s a predictor of a disease and when you have high cholesterol you go to statins. Statins prevent the disease rather than treating it. If you have cardiovascular disease then you are treated with multiple different treatment pathways. I believe in Alzheimer’s, we will move into that as well.

However, this is a very significant phase that we are in, in order to do this right, like we do in cardiovascular disease, you need to have the right diagnostic tools and in fact they’re really coming up now and I have great confidence that in the next two years you will see major breakthroughs in the treatment of this disease.

RM: What are your conclusions from the TAURIEL Phase 2 Trial? What has led to these disappointing results?

AP: The TAURIEL study met the primary safety endpoint but did not meet its primary efficacy endpoint of reducing the rate of cognitive and functional decline from baseline compared to placebo, as measured by the Clinical Dementia Rating-Sum of Boxes (CDR-SB) Score, in people with early (prodromal to mild) Alzheimer’s disease (AD).

The science shows that tau pathology progresses in a spatiotemporal pattern that is correlated with cognitive decline, so we are both surprised and disappointed by the top line results of the trial and look forward to the detailed data. This is the world’s first-ever Phase 2 anti-tau antibody data readout in AD and the full data, including biomarker results, must now be closely analyzed and will be crucial in bringing meaning to the field. The second Phase 2 (LAURIET) study of semorinemab in moderate AD is ongoing.

Our partner Genentech, a member of the Roche Group, is currently conducting additional analyses to further understand these topline data, including the biomarker data, which might provide additional insights to inform the path forward for the clinical development of semorinemab.

Broadening the discussion to how to treat neurodegenerative diseases more generally, one of AC Immune’s key strengths is our diversified approach and we plan to continue developing one of the industry’s broadest, most diversified therapeutic pipelines.  This includes potentially best-in-class small molecule, antibody and vaccine therapeutics. It addresses a number of different approaches, and leverages AC Immune’s two proprietary and proven discovery platforms, SupraAntigen^TM and Morphomer^TM, to develop breakthrough therapies and diagnostic candidates to enable precision medicine.

RM: What does this mean for your other approaches? Is a vaccine also targeted at tau now less likely to be successful? If not, why?

AP: AC Immune has a number of approaches to tackling neurodegenerative diseases. It is unlikely that a silver bullet for AD will exist in the form of a therapy that targets one protein. It’s more probable that a combination therapy with multiple points of intervention will be necessary to cure the disease. There is an increasing recognition that precision medicine and early diagnosis will be required to identify which proteins are causing the disease and to effectively deliver the appropriate combination therapies.

We are thus actively pursuing PET tracers against multiple relevant targets, including tau, alpha-synuclein and TDP-43. Availability of non-invasive diagnostic tools like PET imaging agents would allow accurate diagnosis and monitoring of disease progression and would potentially enable precision medicine approaches and longitudinal drug efficacy measurements in patients.

Looking at tau specifically, semorinemab is one of four AC Immune tau programs with clinical readouts this year.

Despite the TAURIEL top line results, the scientific rationale for targeting tau remains compelling and our anti-tau therapeutic candidates are each unique with key characteristics and a mechanism of action that differ from each other, allowing AC Immune to take a comprehensive anti-tau approach. Ongoing clinical studies are currently testing:

• ACI-35.030 (anti-tau vaccine) – this vaccine generates a robust polyclonal antibody response against epitopes that differ from that targeted by semorinemab. Importantly, these epitopes include an epitope specific to ptau, which is the pathological form of tau protein believed to be responsible for the formation of tangles in AD. Additionally, the antibodies generated by this active immunization approach have long-term pharmacokinetic properties that differ from the pharmacokinetic properties of periodic semorinemab injection. As such, we remain confident in the potential of ACI-35.030.
• ACI-3024 (anti-tau inhibitor) – this first-in-class small molecule enters the brain and passes more readily through the cell membrane, enabling inhibition of intracellular tau aggregates at the earliest point in tau pathology. Thus, there is reason to believe this approach may behave differently in the clinic compared to semorinemab, which is designed to act extracellularly to inhibit cell to cell spread of tau. 

The fourth clinical-stage tau program is a diagnostic candidate designed to improve assessment of tau-pathology in AD and diagnosis of other tau-related diseases such as progressive supranuclear palsy (PSP):

• PI-2620 our tau-PET tracer – this asset has the potential to work as a critical tool in the further development of anti-tau approaches by facilitating the design of clinical trials that treat earlier and target more homogeneous populations.

Andrea Pfeifer was speaking to Ruairi J Mackenzie, Science Writer for Technology Networks