Last week was a tough one for Parkinson’s clinical trials. First came news from Biogen that their antibody targeting alpha-synuclein had failed to meet its endpoints. Then the French pharma company Sanofi released news that their drug for people with a GBA mutation had also failed a phase 2 trial.
The first was not a huge shock, researchers on both sides of the protein aggregation debate were questioning whether this antibody would have any effect in slowing disease progression. But the latter bit of news was a little harder to take. It was believed that this more precise sort of therapy should give us a better chance of slowing disease progression in people that carry that mutation.
These failures have raised more questions about where we go from here in disease modifying strategies to treat Parkinson’s. Maybe disease modification is too ambiguous a term for us to target? Perhaps we should wait until we have a clearer understanding of what the disease is that we are trying to modify and until then stick to symptom modification or replacement therapies where we have a track record of success.
As many GBA-PD experts have pointed out, we may need to be more precise than simply targeting people with an associated genetic variant. So, what might it take for us to get the therapies we need? Will we have to go beyond just genetics? I wrote about that in the following repost of a blog post first written for the Journal of Parkinson’s Disease.
The first generation of precision medicine therapies for Parkinson’s disease (PD) should enter phase 3 clinical trial in 2019 (update for 2021: none of those trials have begun, man do things move slow in pharma.) If even one of them succeeds in modifying disease progression it will open the doors for more targeted therapies and officially mark the beginning of a move towards precision medicine for PD.
But, what if they are not precise enough? After all, most are based on correcting faulty genetic pathways, but in the vast majority of cases, even in those that have a variant associated with PD, genetics can only explain a part of the reason why a person develops Parkinson’s disease. As Ziv Gan-Or, assistant professor at the Department of Human Genetics at McGill University, stated: “We have measured heritability and determined the overall contribution of all genetic risk factors, that number is around 30%, which means most factors are not genetic.” Those remaining factors likely include a wide variety of environmental exposures as well as changes that happen to our biology as we age. However we don’t have good ways of even identifying, let alone treating, all those other factors that may contribute to one’s disease.
The problem is further complicated by the fact that even among therapies in development that target specific genetic cohorts, there are a variety of different strategies being clinically tested and some might be best for certain sub-subsets of these populations. Here is University of Dundee Professor Dario Alessi from a September interview discussing the development of LRRK2 inhibitors, “…some will be more potent than others, some might penetrate the brain better than others, some might last longer in the body, and each will have different off-target effects that can be quite unpredictable. We will probably end up with 7 or 8 different clinical trials and some might work better for certain LRRK2 mutations. Then the work will shift to finding out which patient might be best suited for which inhibitor.”
This challenge of identifying the right sub-subset of patients to fit each trial is one that might need to be solved before we can realize the potential benefits of these therapies. As stated by Anthony E. Lang, Professor of Neurology at the University of Toronto, “How do you screen people for one drug or the other when you have a limited number of people that fit the criteria for entry into these trials? This is an ethical concern that our field needs to consider.”
So, what could we use to identify these sub-subsets and are there any new biological targets that the field should move towards identifying? I put that question to Clemens Scherzer, Associate Professor of Neurology at Harvard Medical School, this was his response:
“Disease progression, not susceptibility to disease, is the key determinant of patients’ wellbeing.
How can we stop or prevent the disease from progressing instead of waiting until the disease progresses and playing catch up? Identifying the genetic drivers of disease progression – not susceptibility – in patients with PD will allow (for) slowing the progression and possibly prevent disabling complications such as dementia from ever occurring.
This will require a paradigm shift in genetics from susceptibility genetics to prognosis genetics. Almost all of what we know about the genetics of PD today relates exclusively to susceptibility for developing the disease, not progression. For example, we know much about genetic loci that increase the risk of healthy people to develop PD in the future. This might lead to primary preventive therapies but this is a very long term goal in the absence of biomarkers. But patients with PD already have PD. The question is no longer about primary prevention, but preventing progression. The key question for slowing there progression is what are the progression loci, not susceptibility loci, and how can we target them.
The reason (no one has done this yet) is that one needs deeply phenotyped longitudinal patient cohorts to do this fundamental experiment. That would require a lot of hard work over a long period of time and substantial amounts of funding as we would need to follow thousands of patients for many years. We are currently looking at 5000 patients over the course of 12 years in the International Genetics of Parkinson’s Disease Progression (IGPP) Consortium, but that’s just the beginning, tens of thousands would be needed.
All susceptibility genetics/GWAS studies just look at shallow data (diagnosis, age, sex) from poorly characterized patients with no longitudinal follow up. It’s the low hanging fruit, but not necessarily most pertinent to patients.”
If Prof. Scherzer is correct, then the genetic therapies in clinical development for genetic cohorts of PD are more likely to be effective as preventative therapies in those yet to be diagnosed. Those already diagnosed may require therapies that target pathways associated with disease progression rather than onset.
This idea was also discussed by Dr. Michaela Johnson, myself and colleagues in a recent paper entitled “Triggers, Facilitators and Aggravators” (1), which hypothesized that some genetic factors, which we called, ‘facilitators’, act during the second phase of the disease process. If the pathology has already developed beyond this to the ‘aggravator’ phase, then just targeting the facilitator may have a minimal effect as it is no longer the dominant driver of pathology.
So, expanding the definition a little further, ‘super’ precision medicine would not only mean identifying which targeted therapy is best suited for which variant within each genetic cohort, but also developing therapies targeting pathways playing a role in the progression of each subset of the disease. While this ‘super’ precision medicine approach certainly adds a degree of complexity to an already incredibly complex problem, it may be a necessary step towards modifying the progression of Parkinson’s disease.
It seems precision therapies of the future might look a little more like this…
Note: One point of clarification. Thanks to the work of groups like the International Parkinson’s Disease Genetics Consortium we do now have genetic studies with much more depth of analysis than we had at the original time of writing this post. Click here for more.