The 2024 McDonald diagnostic criteria for multiple sclerosis, developed by the International Advisory Committee on Clinical Trials in Multiple Sclerosis and published in The Lancet Neurology in 2025, introduce structural changes to how MS is confirmed at the point of initial diagnostic workup. The revision is not incremental. A comparative analysis published in Neurology found that applying the 2024 criteria raised the MS diagnosis rate by 28 percent at initial workup: 220 patients diagnosed versus 172 under the 2017 McDonald criteria (p < 0.01). Median time to diagnosis dropped from 84 days to 40 days (p < 0.01). Any Phase 2 or Phase 3 trial whose inclusion/exclusion criteria reference the 2017 McDonald criteria as the diagnostic standard is now operating on a definition that the clinical community has formally superseded.
That population-level shift in diagnosis speed and volume has direct consequences for trial enrollment, baseline patient characterization, and the validity of historical comparator arms.
What the 2024 Criteria Actually Changed
The most operationally significant revision in the 2024 McDonald criteria is the addition of the optic nerve as a fifth topographic site for assessing dissemination in space (DIS), alongside the four sites recognized under the 2017 version: periventricular, juxtacortical or cortical, infratentorial, and spinal cord. According to the National MS Society’s summary of the revised criteria, this addition expands the anatomical footprint over which DIS can be demonstrated, meaning patients who previously lacked a qualifying lesion count across the four 2017 sites may now satisfy DIS criteria through optic nerve involvement confirmed on MRI or by clinical episode with appropriate paraclinical support.
The practical consequence: a patient population that was previously ineligible for trial enrollment under 2017-anchored inclusion criteria may now carry a confirmed MS diagnosis under 2024 criteria. Sponsors who enrolled early-cohort participants using 2017 McDonald language and are now expanding enrollment under the same protocol are accumulating a heterogeneous baseline population without a corresponding protocol amendment to acknowledge it.
The criteria also formalize a role for biomarkers, including cerebrospinal fluid-specific oligoclonal bands and kappa free light chains, in establishing dissemination in time (DIT). This matters for how sponsors define “confirmed MS” in their eligibility criteria and how baseline disease burden is characterized in stratified randomization schemes. A trial powered on the assumption that all enrolled patients have MRI-confirmed DIT under 2017 definitions may be enrolling a meaningfully different patient when 2024 biomarker-supported DIT is accepted at the site level.
The IND and Protocol Exposure Sponsors Must Address
FDA has not issued MS-specific guidance tying IND eligibility criteria to the McDonald revision cycle. The agency’s framework for real-world data and evidence integration, articulated in its overarching Real-World Evidence Program framework, requires that sponsors prospectively specify the data standards and diagnostic definitions used to characterize enrolled populations. A trial that references “McDonald criteria” without specifying the version year creates an ambiguity the agency will surface at review. Sponsors filing INDs or protocol amendments in the next two quarters should designate the 2024 McDonald criteria explicitly by publication year and confirm whether sites have been instructed to apply 2017 or 2024 standards during the enrollment window.
The 28 percent increase in diagnosis rate documented in the Neurology comparative study also complicates the construction of external control arms. A sponsor building a synthetic control from registry data or electronic health records collected before 2024 is drawing on a patient population characterized by criteria that are now formally outdated. FDA’s real-world data guidance requires that data elements used in external control construction reflect definitions consistent with the trial population’s diagnostic standard. A 2017-characterized registry cohort used as a comparator against a 2024-enrolled trial arm introduces a diagnostic definition mismatch that sponsors should address in their statistical analysis plan before a Type B meeting, not at the advisory committee stage.
The biomarker dimension adds a separate layer. The FDA granted Breakthrough Device designation to Quanterix in April 2022 for a serum neurofilament light chain (sNfL) assay intended to predict disease activity risk in MS. As of July 14, 2026, no blood-based biomarker for MS diagnosis has received FDA approval. The 2024 McDonald criteria’s formalization of CSF biomarkers in the DIT pathway does not create a validated regulatory standard for blood-based tools, but it does signal the committee’s direction. Sponsors developing companion diagnostics or stratification algorithms tied to sNfL or kappa free light chains should not assume the McDonald revision constitutes regulatory pre-clearance for those approaches; a separate analytical validation pathway under 21 CFR Part 820 or the relevant IVD framework applies.
Strategic Context and the Quarter-Level Directive
The 2024 McDonald revision fits inside a broader pattern in which MS diagnostic standards have accelerated ahead of trial infrastructure. The 2017 McDonald criteria themselves represented a significant compression of the diagnostic timeline relative to the 2010 version, and the field did not complete the protocol realignment cycle before the 2024 update superseded them. Sponsors with active INDs referencing 2017 McDonald criteria represent the largest near-term operational exposure: their enrollment data, stratification logic, and any planned external comparator arms are all anchored to a definition the clinical community has formally moved past.
This quarter, MS program sponsors should complete three concrete actions. First, audit all active IND protocols to determine whether McDonald criteria are cited with a version year; where version year is absent, file a protocol amendment to specify 2024 McDonald criteria with an effective enrollment date. Second, review statistical analysis plans for any external control or synthetic control arm to confirm that registry or EHR source data was characterized under a compatible diagnostic standard, and document the bridging rationale explicitly if a mismatch exists. Third, confirm with site principal investigators whether local clinical practice has migrated to 2024 McDonald criteria ahead of any formal protocol amendment, creating a de facto version split within the enrolled cohort.
The next relevant regulatory marker is FDA’s anticipated updated guidance on patient-enrichment strategies in CNS trials, which has been signaled in the agency’s 2025 guidance agenda. When that document publishes, its interaction with the 2024 McDonald criteria’s expanded diagnostic population will determine whether the 28 percent enrollment uplift identified in the Neurology study translates into a meaningful change in trial powering assumptions or a requirement to respecify subgroup analyses by diagnostic pathway.
References
- Nature Medicine — “Challenges and future directions for multiple sclerosis after the 2024 McDonald diagnostic criteria”
- National MS Society — “2024 McDonald Diagnostic Criteria”
- Neurology — Comparative analysis of 2024 vs. 2017 McDonald criteria: 28% higher diagnosis rate, median time to diagnosis reduced from 84 to 40 days
- DIA Global Forum — “FDA’s Real-World Evidence Program: Moving Forward”
- Healthy Blue NC / Quanterix — FDA Breakthrough Device designation for serum neurofilament light chain (sNfL) assay, April 2022
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.
