Parkinson’s disease research has produced effective symptomatic therapies and, so far, no disease-modifying ones. In a summary of his talk from the 14th Alzheimer’s and Parkinson’s Drug Development Summit in Boston in February 2026, our Chief Innovation Officer, Patrick Howson, examines why the two problems have proven so different, and what the field can change to close the gap.
There is a key distinction between disease states and disease processes. Symptoms are driven by a state, such as the loss of dopaminergic neurons. If the severity of the symptom is the same from one individual to the next, then the underlying state is also likely to be similar. Current in vivo models reproduce that state well.[PH1]
Disease-modifying therapies act on an ongoing process, and those processes vary between individuals and shift within a single person over time. No single animal model captures that heterogeneity. Patrick sets out a route forward: define patient cohorts by their biology, using classifications such as SynNeurGe and the NSD-ISS, then match each cohort to a model that reproduces the process under investigation.
What You’ll Learn:
- States and processes. Why animal models have predicted symptomatic efficacy well and disease-modifying efficacy poorly, and what that means for choosing a model.
- Biological classification of Parkinson’s. How SynNeurGe and the NSD-ISS define cohorts around synucleinopathy, neurodegeneration, and genetics, and why that supports more homogeneous study populations.
- What today’s models capture. The processes current in vivo models reproduce well, and the gaps that remain, including the slow prodromal evolution of the disease, Lewy body formation, and non-cell-autonomous degeneration.
- Robustness and repeatability. Cross-laboratory variability in the α-synuclein PFF model, and the role of shared protocols and preclinical quality management (EQIPD) in improving reliability.
Download the Full Presentation
Talk to Us
If you are developing a Parkinson’s disease therapeutic and want to discuss model selection or endpoints, please reach out to schedule a call.