The Critical Importance Of Selectivity When Developing Antibody Therapies For Alzheimer’s Disease And Our Current Phase 1 Design Considerations.
James W. Kupiec, Md Chief Medical Officer
James W. Kupiec, MD
Chief Medical Officer, ProMIS Neurosciences, Inc.
Narrated Presentation, November 2018
“The critical importance of selectivity when developing antibody therapies for Alzheimer’s disease and our current phase 1 design considerations”
This is Dr. James Kupiec, Chief Medical Officer at ProMIS Neurosciences. Thank you for clicking on this presentation.
I recently joined ProMIS after 26 years in large pharma because it has, in my opinion, the most scientifically intriguing and innovative approach to developing therapies for Alzheimer’s disease and other neurodegenerative disorders.
In this presentation, I’ll discuss why the selective targeting of pathologic, misfolded proteins is critical when developing medicines for neurodegenerative disorders, particularly medicines that need to demonstrate both unequivocal safety and maximal therapeutic benefit.
At ProMIS, we’re in the midst of planning our first phase 1 clinical study in patients with Alzheimer’s disease. I will explain why PMN310, our next-generation antibody, has the potential to become the best in class therapy amongst antibody treatments for Alzheimer’s disease.
Finally, I would like to share some of the design elements under consideration for our upcoming phase 1 study and why biomarker advances in the field of Alzheimer’s research will allow us to maximize the value of the information coming out of this very first study.
Although the existence of plaque in the brains of patients with Alzheimer’s disease has been known for many years, it was only about 30 years ago that scientists determined that plaques are insoluble and composed of an aggregated protein called Amyloid Beta, or ABeta.
The figure on the right of your slide shows a section of brain from a patient with Alzheimer’s disease stained with a chemical to highlight the presence of numerous protein plaques.
The ABeta protein is produced at the surface of brain cells as a single monomer, a strand of protein, and this monomer is not toxic. This is the figure on the left.
Because scientists initially assumed that plaques were the most toxic type of ABeta in the brain, potential therapies targeting removal of plaque were devised and tested in small and large clinical studies.
Over the last dozen years, scientists across the Alzheimer’s disease research community have come to recognize that it is the non-plaque, non-monomer forms of the ABeta protein that cause neuronal toxicity, and produce both tau protein abnormalities and ultimately all the symptoms associated with Alzheimer’s disease.
The term oligomer comes from the Greek word meaning “a few parts”. Scientists use this term to designate a molecular assembly of a few monomer units. As indicated on the slide, oligomers composed of two, four or twelve monomers are called dimers, tetramers and dodecamers.
Larger assemblies of the basic monomer units are called polymers, and the Abeta protein can aggregate into large polymeric assemblies called protofibrils. Finally, plaques reflect an even larger aggregate.
It is important to understand that the ABeta protein, when it assembles into oligomeric units, has the tendency to change its shape, or what we call its conformation. When it does so, we characterize the protein as “misfolded”, and it this misfolded protein that is pathologic. This is why we refer to them as toxic oligomers.
It is also critical to note that toxic oligomers are the least abundant form of ABeta protein in the brain. Therefore, antibody therapies that target toxic oligomers need to be very selective for a particular form of ABeta, and not the other, non-toxic forms such as plaque.
A number of antibodies have failed in large phase 2 or phase 3 studies, including bapineuzumab, solanezumab and crenezumab. These antibodies all bind the ABeta monomer.
Another class of potential medicines called BACE inhibitors have failed as a class to demonstrate any benefit to patients with Alzheimer’s disease. These compounds are small molecules and they significantly, but not totally, inhibit the formation of monomer.
At the annual Clinical Trials in Alzheimer’s Disease scientific congress that convened in Barcelona in October, data from key studies showed that BACE inhibitors unexpectedly led to cognitive worsening and not cognitive improvement.
The two antibodies called aducanumab and BAN2401 also received a lot of attention from scientists at the recent CTAD meeting. Despite concerns articulated around the number of patients participating in these studies or various design components, it was the general consensus of the meeting participants that the phase 2 clinical studies with these antibodies showed a positive, dose-dependent signal of clinical benefit.
These two antibodies preferentially bind either plaque or protofibrils, the two types of aggregated ABeta protein. But these antibodies also bind – to a less significant degree – the smaller toxic oligomers. Thus, they are partially selective, and this partial selectivity provides the scientific basis for the clinical benefit shown in their phase 2 studies.
But because these antibodies also bind both aggregated protein in plaques and the ABeta protein aggregates found in the blood vessels of human brains, they can produce an adverse swelling or edema of brain tissue which scientists call ARIA-E. This stands for Amyloid Related Imaging Abnormality of the edema type. This is an unacceptable side effect, and its presence limits the administered dose of either aducanumab or BAN2401.
PMN310 is the first antibody we plan to take into the clinic.
Using its unique discovery platform, ProMIS applied state of the art, proprietary supercomputer modeling to predict the unique targets on toxic oligomers and then selected the best, most selective of over 300 antibody candidates by testing them against brain tissue from patients with Alzheimer’s disease. This is how we created PMN310.
There is extensive laboratory and animal data that encourages us to move this antibody into human testing. For example, it blocks the cognitive deficits in a mouse model that are produced by human toxic oligomers.
PMN310 does not bind to monomers. It does not bind to plaques or vascular deposits of amyloid. It only moderately binds high molecular weight protofibrils.
However, it strongly and preferentially binds the toxic oligomers from humans with Alzheimer’s disease. Therefore, it is highly selective.
We predict we should not observe a dose-limiting side effect such as ARIA-E. And because of its unique selectivity, we should also not waste therapeutic dose on the wrong targets. For that reason, we hope to eventually show a stronger clinical benefit than either aducanumab or BAN2401.
One key item I’ve not yet mentioned is that antibodies come in what we call different isotypes. An antibody of the IgG1 isotype, or immunoglobulin type 1, significantly stimulates the immune system. We call this “effector function” and it contributes to the incidence of ARIA-E.
PMN310 is an antibody of the IgG4 isotype which does not have effector function. As such, this physical characteristic should also contribute to the absence of ARIA-E amongst patients administered this antibody.
In summary, the strong, preferential binding of the PMN310 antibody to toxic oligomers and its IgG4 isotype both provide the ideal design characteristics that position this antibody as one that is potentially best in class and represents the next generation of therapeutic antibodies.
Before I end today, I’d like to discuss some key design elements under consideration for our first-in-human phase 1 study with PMN310.
As currently envisioned, we will conduct what I call a hybrid study evaluating the safety of PMN310 in both healthy volunteers and patients with Alzheimer’s disease, both single doses and multiple doses. Such an accelerated design represents a significant improvement in the conduct of phase 1 studies since I began testing antibodies over a decade ago.
When first evaluating the safety of a new drug candidate in a phase 1 study, low doses are initially evaluated, and then the doses are sequentially increased in an escalating, stepwise fashion only after safety is adequately demonstrated in the group – or what we call a cohort – of study participants at the lower dose.
As shown in this slide, we’ll begin clinical testing in healthy normal volunteers and administer PMN310 at two doses for the first two cohorts of volunteers. We’ll then transition to cohorts of patients with Alzheimer’s disease in whom we will also administer PMN310 in an escalating fashion.
It is a common technique to evaluate approximately 8 study subjects at each dose before escalating to the next dose – typically 6 on the active investigational drug and 2 on placebo. Once we reach the maximal safe and well-tolerated dose, we plan to evaluate that dose in more than 8 study subjects by adding additional subjects to that cohort.
Once the study subjects complete the basic double-blind assessment period, they will then be offered the opportunity to participate in an open-label extension phase of the study where all patients will receive the PMN310 antibody.
Serious side effects like ARIA-E typically occur within the first few months of testing. Our study design will allow us to collect an adequate amount of long-term safety data and confirm that PMN310 is not associated with dose-limiting side effects such as ARIA-E.
Including cutting-edge biomarker assessments in our phase 1 study is also a critical design component, and this is highlighted in the boxes on the right side of the slide. Biomarkers are typically a measurement of something in the blood or other body fluids that help us predict the potential effectiveness of a drug candidate under investigation.
There have been significant advances in the research field of biomarkers over the last few years. At ProMIS, we are actively engaged in working with colleagues at the National Institutes of Health and across the Alzheimer’s disease research community to both further biomarker knowledge and to capitalize on these learnings.
We will indeed benefit from the existing advances by incorporating a number of key biomarkers into our very first study. Biomarkers from both blood and cerebrospinal fluid will be collected, and we will have the opportunity to show the potential therapeutic benefit of PMN310 on both synaptic function and neuronal structure very early in clinical testing.
This advance in biomarker technology represents a real breakthrough in early drug development.
At ProMIS, we’re working very hard to ensure we have initial results from this phase 1 study in 2020, around the time the research community expects to see the results from the aducanumab phase 3 studies.
One final comment: as alluded to during my introduction, pathologic, misfolded proteins are also critical in the development of other neurodegenerative disorders such as Parkinson’s disease and Amyotrophic Lateral Sclerosis (more commonly known as ALS or Lou Gehrig’s disease).
ProMIS has also determined the pertinent targets for the tau protein, which is the second misfolded protein found later in Alzheimer’s disease.
As a physician that has worked in this area for many years, I’m quite thrilled that ProMIS has also generated selective antibodies targeting these misfolded proteins.
This is James Kupiec, CMO at ProMIS Neurosciences, and I look forward to sharing with you in the future our plans for developing these other therapeutic antibodies.
Thanks for joining me today. I sincerely appreciate your time and attention.