Marking time: does lack of change raise more questions than answers?

Clinical trials of potentially neuroprotective agents in Parkinson’s Disease face several major problems, Brit Mollenhauer (University of Göttingen, Germany) told the PD Therapeutics conference.  

• Diagnosis comes comparatively late in the disease course: probably more than half our dopaminergic neurons have degenerated before the classical symptoms of PD become apparent.
• There is lack of accuracy in diagnosis. It is thought that half our clinical diagnoses have to be revised in the light of information that emerges over the subsequent five years.
• With the possible exception of change on the Unified PD Rating Scale (UPDRS), there are no reliable endpoints in early disease.

The beauty of the Initiative is serial collection of fluid biomarkers from many centers

This background provides a clear rationale for the longitudinal Parkinson’s Progression Markers Initiative which enrolled newly diagnosed and untreated people with likely PD. The essence of the study was that it would be long-term, and comprehensive – including assessment of motor function, olfaction, neurophysiology, and the serial collection of biofluid samples from blood and CSF for analysis of alpha-synuclein, total and phosphorylated tau, beta-amyloid 1-42 (Aβ42), DNA and RNA.  As well as being bold in its scope, the initiative is wide in its reach, involving 21 centers across the United States and Europe.

One-year markers showed lack of change and discrimination

Over the period 2010 to 2013, 196 healthy controls and 413 people thought to have PD were recruited. Some who were thought to have PD turned out to have no dopaminergic deficits on DaT scanning. Subsequently, Brit Mollenhauer and co-investigators added a prodromal cohort consisting of 35 subjects with idiopathic REM behavior disorder, 20 with hyposmia, and a small number with known PD-associated mutations.

The first longitudinal analysis, at twelve months, included data from 173 people with PD and 112 matched healthy controls. Perhaps surprisingly, the paper published last year in Neurology¹ revealed no changes in biomarkers that were specific to PD progression.

Both in people with PD and in controls, there was a small but significant increase in CSF Aβ42 from baseline to 12 months. Total alpha-synuclein levels in CSF remained relatively stable. From baseline to 12 months, p-tau significantly increased in patients with PD but did not rise in HCs.

An apparently significant correlation between UPDRS and CSF markers became non-significant after adjustment for multiple comparisons. Patients taking PD medication, which was allowed after six months, had a greater decline in CSF alpha-synuclein than those not taking such medication. One gene polymorphism (in the SNCA gene) predicted change in Aβ42.

Overall, the stability of the measured markers over a year suggests we need longer follow-up, better markers, or both

The decision to extend follow-up to 36 months was accompanied by the decision to re-analyze all samples from baseline, 6 and 12 months. Discrepancies were identified and we need to be aware of this, Dr Mollenhauer said. Reasons for the variability in results are being looked into but may have to do with the analytic techniques themselves or the handling of samples before analysis.

Overall, there was stability over a year in the core markers of interest measured in people with de novo PD. Importantly change in markers did not parallel the progression that was seen in motor symptoms and on Dat scans.

These core markers may show change over a longer period, or in more advanced disease, or perhaps with disruption of the blood-brain barrier. It may also be that we need greater refinement in our markers, such as use of specific forms of alpha-synuclein that might be more closely related to pathology.