A clinical operations lead gets the news from her biostatistics team: the new digitally enhanced composite endpoint just cut the required sample size by nearly 25%. She pulls up the enrollment projection, recalculates the per-patient budget, and starts thinking about which sites she can release early. Then her regulatory affairs counterpart sends a follow-up email: “We’ll need a protocol amendment before we can switch endpoint definitions. IRB submissions at all sites. eCOA validation package. DHT bridging data. Plan for six to nine months.” The projection goes back in the drawer.
That gap between the statistical win and the operational reality is exactly what the CONSONANCE Phase 3b framework published in medRxiv forces sponsors and site teams to think clearly about. Evaluated in Roche’s 629-patient ocrelizumab study, a digitally enhanced composite confirmed disability progression endpoint (cCDP) reduced required trial sample size by 24.9% at Week 48 compared to the standard cCDP definition. That is a real number with real budget consequences. It is also the beginning of an operational to-do list that most trial teams are not ready for.
What the 24.9% Actually Buys You
In a large Phase 3 MS trial, a 25% sample size reduction translates to meaningful cost relief. Per-patient costs in progressive MS trials commonly run into six figures when you account for EDSS assessments, MRI reads, neuro-cognitive testing, and coordinator time across two to four years. Shaving 150 to 170 patients off an enrollment target at that cost basis changes the economics of the program, not just the Gantt chart. It also shortens the enrollment window at each site, which matters for PI retention and coordinator morale more than most sponsors acknowledge at the outset.
The CONSONANCE framework earns that reduction by converting high-frequency smartphone-based gait data into progression events and folding them into the composite alongside traditional in-clinic measures. Instead of waiting for a patient to return for a timed 25-foot walk or a 9-hole peg test, the model captures ambulatory deterioration between visits and counts qualifying remote events toward the endpoint. More events, detected sooner, with greater precision. The statistical logic is sound, and the EMA’s acceptance of stride velocity as a primary endpoint in ambulatory Duchenne Muscular Dystrophy studies confirms that regulators can qualify digital gait measures when the validation package is rigorous.
But that validation package is where site operations teams need to pay attention, because the work does not happen in the background.
The Amendment Queue Nobody Budgeted For
Integrating a remote digital endpoint into an existing composite requires a protocol amendment. Not a clarification, not an administrative update: a substantive amendment that touches the endpoint definition, the statistical analysis plan, the informed consent document, and the eCOA specifications. Industry data puts the average amendment implementation timeline at 260 days from identification to final ethics board approval. That figure has nearly tripled over the prior decade. For a sponsor planning to introduce a digitally enhanced endpoint mid-study, 260 days is not a planning assumption, it is a warning.
Sites feel this differently than sponsors do. A protocol amendment means re-consenting enrolled patients, which requires a coordinator to schedule and document individual conversations with every active participant. In a 629-patient trial spread across 40 or 50 sites, that is a coordinator burden that arrives without additional budget unless the contract was written to include it, and it rarely is. The re-consent process also restarts the clock on certain data collection requirements, which creates TMF entries, deviation risk, and monitoring visit findings if the rollout is uneven across sites.
Then there is the eCOA and digital health technology validation stack. The FDA’s December 2023 final guidance on digital health technologies for remote data acquisition in clinical investigations sets out what sponsors must demonstrate before relying on DHT-derived data in a regulatory submission: fitness-for-purpose evidence, data flow verification, user verification, and a plan for handling missing or intermittent remote data. Every one of those requirements generates documentation that flows into the TMF. Sites that have never received a smartphone-based gait platform before need a site-level technology qualification, an SIV addendum, at minimum, and in many cases a separate site technology initiation visit before patients can even be enrolled on the digital arm of the composite.
Across the sites I work with, the technology initiation step alone adds two to four weeks to startup for sites that are new to the ePRO or remote monitoring vendor. That delay compounds when the vendor’s site support team is spread thin across multiple concurrent studies, which it almost always is during a broad rollout.
Where Sponsors Underestimate the Site Cost
The sample size reduction math is clean on paper. The operational math at the site level has more variables. When a sponsor introduces a digital gait endpoint, the coordinator’s job expands: device provisioning, patient training on the smartphone application, troubleshooting connectivity issues, chasing missing data flags from the central monitoring dashboard, and documenting every instance where a patient’s remote data falls below the completeness threshold. None of that is captured in the original per-visit budget because it does not map to a visit. It maps to coordinator hours that accumulate between monitoring windows.
The CONSONANCE framework ran in a trial where the digital infrastructure was built in from the start. For sponsors considering this approach on an ongoing study, or designing a new study around a blended endpoint, the operational question is not whether the endpoint is statistically valid. It is whether the site-level budget reflects the true coordinator burden, whether the amendment timeline is factored into the enrollment projection, and whether the eCOA vendor has the site support capacity to onboard 40 sites in the window the sponsor has planned.
The 25% sample size reduction in CONSONANCE is a genuine signal that blended digital-clinical endpoints can do real statistical work in progressive MS. Roche’s MUSETTE trial, which did not meet its primary cCDP endpoint in relapsing MS, shows how much rides on endpoint sensitivity in this therapeutic area. Getting the endpoint architecture right matters. So does getting the site operations plan right before the amendment queue opens.
For sponsor clinical operations teams designing the next progressive MS study: build the DHT validation package before you lock the protocol, model the amendment timeline into your enrollment projection as a hard constraint rather than a buffer, and price coordinator re-consent and technology initiation work into the site budget explicitly. For site directors evaluating feasibility for a blended endpoint trial: ask the sponsor directly how many active studies are already on the remote monitoring vendor’s platform, and whether site technology initiation is budgeted as a separate line or absorbed into the SIV. The answer will tell you whether the 25% reduction flows to your enrollment timeline or disappears into an amendment queue that nobody planned for.
The next sponsor to file a blended digital-clinical endpoint package that survives regulatory scrutiny will set the evidentiary bar for progressive MS trials. Every site operations team that has worked through one of these amendments will know exactly what that filing cost in coordinator hours.
References
- medRxiv Neurology, “Remote ambulation monitoring enhances composite disability measurement in MS to deliver reduced trial sample size”
- CM Health Law, “FDA Releases Guidance on Digital Health Technologies for Clinical Investigations” (December 2023)
- Roche, CONSONANCE Phase 3b Trial (ocrelizumab/Ocrevus), Hoffmann-La Roche
- Roche, MUSETTE Trial Primary Endpoint Results (April 2025)
- Applied Clinical Trials, “Shining a Light on the Inefficiencies in Amendment Implementation” (260-day average amendment timeline)
- PMC, EMA qualification of Stride Velocity 95th centile (SV95C) as a primary digital endpoint in ambulatory DMD studies
