Letter to the Linked Clinical Trials Committee

The following was sent to the leadership of the Linked Clinical Trials Committee in August of 2019. We are republishing it today because we just heard that sadly the phase 3 trial of Exenatide did not meet its primary endpoints. We ask again the research community to read what we wrote in the hopes that we learn from mistakes of the past and begin working together towards better therapies for Parkinson’s Diseases.

The Linked Clinical Trials initiative is one of the most exciting and promising developments in all of neurology. It continues to serve as a living testament of Tom Isaacs’ brilliance by giving hope to individuals and families living with Parkinson’s disease around the world. We are grateful for the bold vision of Dr. Richard Wyse and the steadfast commitment of all those involved that make the LCT possible. 

However, as all platforms periodically do, we believe it is need of a reassessment of some of its founding principles as well as critical review of the path it follows. Below are recommendations for changes that we believe will enable the LCT to one day deliver on improved therapies for people diagnosed with Parkinson’s disease.

These recommendations can be broadly lumped into the following categories:

1. Defining the Problem
2. Care vs. Cure: A False Dichotomy
3. Identify Responders
4. Raising the Bar: Exenatide

Defining the Problem

What is the problem that the LCT is trying to solve? Its stated goal is “To identify readily available treatments that have the potential to slow, stop or reverse Parkinson’s disease.” Before proceeding we first ask, does Parkinson’s disease exist as a single biological entity that can be targeted pharmacologically? 

The pathophysiology of Parkinson’s disease is murky at best. For decades experts have acknowledged that PD has a multi-causal heterogenous etiology that results in a wide spectrum of disorder. However, we continue to put forward therapies that target single pathways, acting as if all that heterogeneity converges on a unifying pathology. It is our recommendation that we dispel the latter and put our understanding of the complexity of PD into practice. 

Thus we advise that if an intervention is proposed to target a mechanism presumed to be etiologic in PD, every attempt should be made to specify what “flavor” of PD is presumed, and not target the entire spectrum, to be considered viable. The concept of “subtyping” PD may depend upon the nature of the target(s) to be potentially modified.

Furthermore, due to the aforementioned complexity, any ‘curative’ trial would likely have to be able to target multiple pathways simultaneously. Trials that only act on one molecular pathway, particularly if the selection of the molecularly suitable subtype of PD is missing, are best to be avoided. 

(A note on the most favored convergent pathway of today – Parkinson’s disease began as a clinical definition used to triage individuals who exhibited a common range of symptoms. Over time, evidence emerged that PD is associated with the aggregation of toxic strands of the protein alpha-synuclein. As it stands, we believe we do not have enough evidence to definitively conclude that the relationship is causal (except for those with a mutation in the alpha-synuclein gene, a rare and very aggressive form of Parkinson’s); rather, it may be the net effect of multiple causal processes. In its proceedings, the committee acts under the assumption that the relationship is causal, often referring to PD as a synucleinopathy, and that agents promoting disaggregation will be therapeutic. Unless the multiple prior causes are addressed, this may well be a misplaced effort. Until we have proof-of-concept in humans that aggregation of alpha synuclein is itself a toxic end-product, we recommend a more agnostic definition of the problem we are trying to solve.)

Care vs. Cure: A False Dichotomy

The committee seems to feel a debt to live up to the name of its parent organization by only seeking treatments that have curative properties. We believe this places a false hurdle on possible therapies. The distinction between symptom-modifying and disease-modifying has not been clearly elucidated. Case-in-point: Levodopa is not considered a disease modifying therapy, yet its introduction in the 1960’s doubled life expectancy of people diagnosed with Parkinson’s disease. (Another more recent example is  the KARMET constipation study – NCT03781791)

This shift towards consideration of therapies that appear to predominantly have symptomatic benefit would also give us clearer targets as outcome measures in trials. Measuring constipation, time spent On vs. Off, or rate of falls, give much more tangible outcomes than the amorphous ‘disease progression’. In addition, the history of modern medicine indicates that treatments for complex chronic diseases are most effective when they aim to replace lost function rather than intercede in biological processes, particularly when we don’t adequately understand the underlying biology.

Finally a reminder that for patients, our symptoms are the disease. We can’t feel genetic mutations, or mitochondrial dysfunction; we can only feel their macro effects. Any functional cure would have to ameliorate symptoms to earn the name.

(Also note that frequently used words like ‘disease-modifying’ and ‘neuroprotective’ are not suitable goals for pharmacological intervention in PD as they lack clear physiological definitions and are not something that can be put on a drug label.)

Identifying Responders

It is our belief that every effort should be made to put forward compounds that target properly segmented biological cohorts or have methods in place for teasing out responders. If a compound has only shown efficacy in so-called ‘PD models’, or has epidemiologic data from broad cohorts of ‘people with PD’, further effort should be made to specify cohort or trial participants should be characterized well enough to increase the chance of subtype specific biomarker discovery and better post-hoc analysis, to be then carried forward to verification in future trials.

Raising the Bar: Exenatide

If we are to learn any lessons from past failure, it is that we should not take Phase 2 results and simply extrapolate them along longer timelines and larger cohorts, hoping that the trend lines from Phase 2 remain steady. This is especially true when we know that there were many problems in the Phase 2 trial, such as failure of random assignment attempts to equate groups on pretreatment characteristics. 

For these reasons it is our recommendation that the LCT committee consider redoing a Phase 2 trial for exenatide before moving on to Phase 3.

—

Tackling complex brain disorders is a humbling experience for all involved. In almost all of neurology, psychiatry, and pain syndromes involving the CNS, we are still grasping at elusive hints from animal models that inadequately translate to benefits in humans suffering from these syndromes. Our past attempts are littered with a succession of failures that have yet to lead to the critical insights needed to properly address these crippling disorders. We are grateful to the LCT for the bold step taken to include people with the lived experience, such as ourselves, during critical discussions when decisions regarding the path forward are being made. We hope that the feedback we have provided in this letter based on our observations of those discussions will help hasten our advance towards our shared goal.

We once again thank you for the dedication you have shown to helping people diagnosed with Parkinson’s disease and look forward to working with you to bring improved therapies to patients everywhere.

Sincerely,

Benjamin Stecher and Dr. Karen Raphael

(With valuable input from Kevin McFarthing and Sue Buff, as well as endorsement from Martin Taylor and Gary Rafaloff)