Sanofi/Ad Scientiam Presents Interim MS-DETECT Data on MS Copilot for Disease Monitoring

Sanofi/Ad Scientiam's MS Copilot Shows Promise in Monitoring MS Progression

Sanofi and Ad Scientiam presented interim analysis from the ongoing MS-DETECT study (NCT05816122) at the 12th Congress of the European Academy of Neurology (EAN) 2026. The study demonstrated that Microsoft (MS) Copilot Detect, a smartphone app, can effectively monitor disease progression in 336 patients with multiple sclerosis across the US, Canada, and Europe. The tool's scores correlated with the established MSFC-4 framework over a 12-month period, indicating its potential for sensitive and scalable remote assessment of MS disability progression.

• The MS-DETECT study's interim analysis revealed a strong correlation between MS Copilot Detect scores and the revised four-component Multiple Sclerosis Functional Composite (MSFC-4) framework. This correlation was observed for assessments of walking capacity, cognitive processing speed, upper-limb fine dexterity, and low-contrast visual acuity, both in clinic and when performed independently by patients at home.
• MS Copilot Detect addresses a significant unmet need for sensitive and scalable clinical measures of disability progression in MS patients, as traditional assessments like EDSS and MSFC-4 can be time-consuming. Its ability to track digital biomarkers via smartphones positions it as a promising tool for decentralized data collection in clinical trials and real-world settings.
• While promising, the full results from the MS-DETECT study are crucial to evaluate long-term adherence and patient attrition rates, which are critical for real-world feasibility. Further validation against established blood, CSF, and imaging biomarkers (e.g., GFAP, NfL, PRLs, SELs, brain atrophy) is also required to strengthen its role as a surrogate or complementary endpoint in future clinical trials.

Addressing Unmet Needs in Multiple Sclerosis Progression Monitoring

Despite meaningful advances in disease-modifying therapy (DMT) development, significant gaps remain in the equitable diagnosis, monitoring, and management of multiple sclerosis (MS) across diverse patient populations. The unmet need landscape spans geographic inequities, therapeutic limitations in specific disease subtypes, prognostic uncertainty, and the complexity of managing special patient populations.

• Global Access Disparities: Large segments of populations in Africa, Latin America, the Middle East, and Southeast Asia lack access to adequate MS diagnostic procedures, compounded by a critical shortage of neurologists. Healthcare disparities exist across virtually every WHO member state, with Social Determinants of Health (SDOH) contributing to elevated morbidity and higher rates of disability. Active conflicts and large refugee populations further strain healthcare systems, constituting substantial barriers to adequate MS management.

• Underserved Domestic Populations: Within high-prevalence, economically developed countries, minorities, immigrants, and other marginalized groups face suboptimal access to care. MS is increasingly identified among populations historically considered uncommonly affected, underscoring the need for broader epidemiological vigilance and targeted outreach.

• Treatment Access and Complexity: A coordinated international effort is required to facilitate access to highly effective but costly treatments, including monoclonal antibodies and B-lymphocyte–depleting agents. The treatment landscape has grown increasingly complex with the expanding number of DMTs and their associated safety profiles, necessitating personalized strategies that balance efficacy, tolerability, PML risk, and patient-specific factors such as pregnancy planning.

• Primary Progressive MS: Primary progressive MS remains severely undertreated, with B cell depletion representing the only FDA-approved therapeutic class. Novel treatment approaches are needed, with emerging research into CC chemokine signaling pathways and biomaterial-based models offering potential new targets.

• Prognostic Framework Gaps: MS is characterized by considerable prognostic uncertainty, and no unified framework currently exists to integrate diverse prognostic factors — including lesion burden, neuroaxonal injury, structural and cognitive reserve, and digital biomarkers. Current prognostic approaches are largely focused on disease burden and fail to adequately account for compensatory mechanisms that complicate long-term outcome evaluation.

• Special and Transitioning Populations: Tailored management strategies remain insufficient for patients in special situations including pregnancy, postpartum and breastfeeding periods, active infection (including COVID-19), malignancy, and perioperative care. Additionally, real-world evidence on outcomes following switching or discontinuation of second-line therapies remains limited — with safety concerns (32.0%), inadequate efficacy (29.9%), tolerability issues (13.6%), and pregnancy planning (8.9%) identified as primary drivers of therapy changes.

• Disease Progression Monitoring: There is a recognized need for non-invasive complements to MRI for assessing disease severity and progression, with retinal imaging emerging as a promising modality. The absence of validated, integrative monitoring tools limits the ability of clinicians to track disease trajectory and adapt therapeutic strategies in a timely manner.

MS-DETECT: Unpacking the Design and Interim Findings for MS Copilot

Phase III pivotal trials remain the regulatory standard for disease-modifying therapy (DMT) approval in multiple sclerosis, and their architecture has grown substantially more sophisticated over time. Core design parameters consistently evaluated include trial duration, sample size, choice of comparator (placebo-controlled or active-controlled), and a tiered endpoint structure encompassing primary clinical outcomes, MRI-derived measures, and patient-reported outcomes. Modern trials demand large, multinational patient cohorts to achieve adequate statistical power, with contract research organisations playing a central logistical role. Randomised direct comparative trials are recognised as the only rigorous method for evaluating relative efficacy between therapies, given that cross-trial comparisons are confounded by differences in study methodology, patient population characteristics, placebo group behaviour, and trial timing. Standardised, consensually validated protocols for both Phase II and Phase III settings have been developed, incorporating standardised definitions for short-term clinical and radiological endpoints. Key design innovations include complex adaptive frameworks, Bayesian approaches to reduce sample size requirements — particularly for paediatric populations — and real-world evidence integration via Bayesian network meta-analysis, which modelling has shown can reduce per-arm patient requirements by at least 40% and compress timelines by six or more months.

Several landmark trials illustrate these principles in practice. The ULTIMATE I and II studies — two-year, multicenter, randomised, active-controlled, double-blind trials conducted from September 2017 to November 2020 — enrolled 985 adults with relapsing MS, of whom 851 (>85%) entered the ULTIMATE open-label extension (OLE). Key efficacy endpoints were annualized relapse rate (ARR), 24-week confirmed disability progression (CDP24), and 24-week confirmed disability improvement (CDI24). Participants switching from teriflunomide to ublituximab (TER-UBL) experienced a 58.4% ARR reduction at year one post-switch (0.182 vs. 0.076; rate ratio 0.42; 95% CI 0.29–0.60; P < .001), with ARR continuing to decline to 0.048 and 0.045 at years four and five, respectively. By year five, CDP24 was 8.0% in the continuous ublituximab arm (UBL-UBL) versus 14.3% in TER-UBL (P = .01), while CDI24 favoured UBL-UBL at 17.0% versus 12.2% (P = .02), with 92% of UBL-UBL participants remaining free from CDP24. The tolebrutinib Phase 2b trial employed a 16-week randomised, double-blind, placebo-controlled crossover design across 40 centres in ten countries, enrolling 130 participants across four dose cohorts (5 mg, 15 mg, 30 mg, 60 mg). Its primary endpoint — new gadolinium-enhancing lesions after 12 weeks of active treatment — demonstrated a statistically significant dose-dependent reduction (P = .03), with the 60 mg arm achieving a mean of 0.13 new lesions versus 1.03 on placebo. The subsequent GEMINI Phase 3 programme, using teriflunomide as an active comparator, further revealed a hazard ratio for 24-week confirmed disability worsening of 0.71 (95% CI 0.53–0.95), despite a neutral relapse rate ratio of 1.03 — distinguishing tolebrutinib as the only agent in its comparison set to demonstrate a disability benefit independent of a measurable relapse effect.

Notwithstanding these advances, persistent methodological gaps constrain the field. Safety reporting quality across 30 Phase III trials submitted for FDA approval between 1995 and 2022 was rated only fair, with an average score of 10.2 out of 15; laboratory-defined toxicity criteria were present in only 53% of publications and adverse event definitions in only 40%. Trial durations of two to three years remain insufficient to characterise long-term benefit-risk profiles, and high-quality efficacy and safety data extending beyond the pivotal trial window are almost entirely absent. Future design evolution is expected to prioritise patient-reported outcomes, surrogate biomarkers capable of predicting individual treatment response, and enriched progressive MS populations through targeted biomarker strategies. Platform trials and sustained registry infrastructure are recommended to support more pragmatic trial designs, while comorbidity profiling of enrolled populations — and stratified analyses by comorbidity status — are identified as critical, currently underimplemented components of both trial design and pharmacovigilance planning.

The Emerging Role of Digital Biomarkers in MS Clinical Trials

Over the past three years, the MS clinical trial landscape has seen a meaningful shift away from reliance on annualized relapse rate (ARR) and MRI lesion burden as primary endpoints, toward measures that better capture the full spectrum of neurological deterioration. A central development is the emergence of Progression Independent of Relapse Activity (PIRA) as a distinct and clinically meaningful outcome measure. In large real-world cohorts — including the Italian MS Register of over 30,000 patients — PIRA ranged from 38.8% to 74.1% over a mean follow-up exceeding 11 years, underscoring its prevalence and prognostic importance. Alongside PIRA, Confirmed Disability Accumulation (CDA) and Relapse-Associated Worsening (RAW) — including RAW at first demyelinating event and recurrent RAW — are being evaluated as discrete endpoints. The phase 3 tolebrutinib trial exemplifies this evolution: confirmed disability progression sustained for at least six months served as the primary endpoint, with 22.6% of treated patients versus 30.7% of placebo patients meeting this threshold (HR 0.69; 95% CI 0.55–0.88; P=0.003). These progression-oriented endpoints reflect a growing recognition that silent neurological worsening — not captured by relapse or MRI monitoring — represents a critical unmet measurement gap.

Composite endpoints and patient-centered measures are also gaining traction as the field moves toward more holistic assessments of disease burden. NEDA-3 (No Evidence of Disease Activity), defined as the simultaneous absence of relapses, confirmed disability progression, and MRI activity, is now widely adopted as a therapeutic benchmark, though it was achieved by only 29.2% of patients after one year of first-line DMT in a Brazilian real-world cohort. Its extension, NEDA-4, incorporates cognitive stability — most commonly assessed via the Symbol Digit Modalities Test (SDMT) — and in real-world ofatumumab studies, 68.5% of patients with longitudinal cognitive data achieved NEDA-4, while 80.4% of NEDA-3 achievers also met NEDA-4 criteria. Cognitive endpoints more broadly are attracting attention, with SDMT and the Paced Auditory Serial Addition Test (PASAT) now incorporated into longitudinal trial designs, though both face methodological challenges related to practice and learning effects. Patient-reported outcomes — encompassing fatigue via the Neurological Fatigue Index MS, quality of life, sleep quality, self-efficacy, depression, and functional well-being — are increasingly being analyzed alongside clinical measures, particularly in age-stratified subgroup analyses.

On the biomarker front, several emerging endpoints are transitioning from research tools toward trial-relevant measures. Serum neurofilament light chain (sNfL) has the most established evidence base, with growing assay availability enabling its use alongside MRI and clinical endpoints to monitor neuronal injury and treatment response — as demonstrated in the ASCLEPIOS I/II trials comparing ofatumumab with teriflunomide. Brain volume loss (BVL), particularly whole-brain, thalamic, deep grey matter, and cortical atrophy measures, is being tracked as a short-term risk marker for disability progression even in the absence of relapses or new lesions; a decline in BVL/year from −0.5% to −1.0% was associated with a relative increase in EDSS progression risk of 26%. Optical coherence tomography (OCT)-derived ganglion cell layer (GCL) thinning is emerging as a sensitive structural biomarker of neurodegeneration, with ofatumumab-treated patients showing stable GCL over 24 months (+0.28 µm/year; p=0.06) versus significant thinning in controls (−0.99 µm/year; p<0.001). CSF biomarkers — including neurofilament heavy chain (NFH), sTREM2, CXCL13, GFAP, and myelin basic protein — and novel measures such as the ALPS index for secondary progressive MS further reflect the field's ambition to develop objective, sensitive endpoints that capture neurodegeneration and neuroinflammation beyond what conventional clinical and radiological tools can detect.