The MDS-UPDRS Part III has an intraclass correlation coefficient for postural tremor that falls below 0.14 before calibration in controlled in-person settings. Add a camera angle, variable lighting, and video latency, and the number gets worse. That single data point explains why the FDA’s September 2024 final guidance on conducting clinical trials with decentralized elements asks neurology sponsors to resolve a binary choice that cannot be deferred: the in-person or remote designation for each assessment must be specified in the protocol before the first patient enrolls, and cannot be handled case-by-case during the study.
The practical split is sharper than many teams expect when they first read the guidance. Daily ePRO symptom diaries, caregiver-reported functional scales like the ADCS-ADL, and eConsent workflows all carry high remote feasibility. The C-SSRS can be administered remotely by a certified rater, subject to escalation protocols for threshold responses. The MoCA has telephone validation, though it is not considered a substitute for a primary endpoint assessment. Those elements belong in a hybrid design. The ADAS-Cog 13, the MDS-UPDRS Part III motor examination, and the EDSS do not: each requires physical materials, examiner proximity, or direct observation of ambulation that video cannot replicate under validated conditions. Using any of them remotely as a primary endpoint without protocol-level accommodation and SAP modifications would put the sponsor out of compliance with the data quality standard the guidance requires.
The consent architecture for cognitively impaired populations adds another layer that cannot be patched mid-study. A hybrid Alzheimer’s trial needs three distinct consent pathways configured from the start: participant self-consent, legally authorized representative co-consent when cognitive status or IRB requirements demand it, and re-consent when protocol amendments require updated consent documentation. Programs in Alzheimer’s or Parkinson’s face enrollment challenges where patient burden is high,, so enrollment efficiency from decentralized elements matters. Losing those gains because an eConsent platform was not built for cognitively impaired navigation, or because amyloid PET collection was treated as potentially remote when it requires a qualified imaging site, is an avoidable cost.
The design decision that determines whether a hybrid CNS program holds its primary endpoint at a regulatory review is not the telehealth platform or the ePRO vendor. It is whether the team sorted each assessment into the right column before the protocol was locked, and built the statistical analysis plan around that sort from day one. The first patient enrolled under a misclassified primary endpoint is the point of no return.
Moe Alsumidaie, MBA, MSF, is founder and Chief Editor of Vanguard Publications, which publishes Clinical Trial Vanguard, Pharma Vanguard and BullScope, and Head of Research at CliniBiz. He has two decades in clinical trial operations and data science, with earlier roles at Genentech, Abbott Vascular and Stanford University Medical Center, and is a guest lecturer in clinical trial sciences at Rutgers University.
