The Safety Gap: Why CAR-T Trials in MS Require the 2024 McDonald Criteria & R3 Governance 

December 22, 2025
The High-Stakes Era of Neuro-Immunology 

For decades, Multiple Sclerosis (MS) drug development was a game of “maintenance”—using Disease Modifying Therapies (DMTs) to slow progression. But as we look toward 2026, the paradigm is shifting toward “immune reset.” 

Therapies like CAR-T (Chimeric Antigen Receptor T-Cell) are offering the unprecedented potential to deplete deep-tissue B-cell reservoirs and halt disease activity entirely. 

 

The Foundation: Understanding the Potency 

We have previously documented the transformative potential of CD19-directed CAR-T cell therapy to cross the blood-brain barrier and deplete pathogenic B-cells in the CNS. (For a deep dive on the biological mechanisms and toxicity profiles like ICANS and CRS, read our comprehensive Accelsiors Spark White Paper here). 

However, establishing biological potency is only half the battle. High potency brings high risk. The operational challenge for 2026 is ensuring that this powerful tool is directed with absolute precision—a challenge that the 2017 diagnostic criteria were never designed to meet. 

Administering a toxicity-prone therapy like CAR-T to a patient who is actually a “vascular mimic” (misdiagnosed) is not just a protocol deviation; it is an ethical failure. To close this safety gap, clinical operations must align two new frameworks: the 2024 Revised McDonald Criteria and ICH-E6(R3). 

  1. The Diagnostic Safety Net: From Sensitivity to Specificity

The 2017 McDonald Criteria prioritized sensitivity—diagnosing MS quickly to start standard DMTs. But for CAR-T trials, we need specificity. We must be absolutely certain the pathology is MS before introducing engineered cells. 

The 2024 Revised McDonald Criteria provide the necessary guardrails. 

  • The “Rule-In” Biomarkers: The new criteria emphasize specific MRI markers: the Central Vein Sign (CVS) and Paramagnetic Rim Lesions (PRL). These features are highly specific to MS pathology. 
  • Ruling Out Mimics: In older populations or those with comorbidities, vascular lesions can look like MS lesions on a standard MRI. By mandating CVS/PRL assessment, we can confidently rule out these mimics. 

Operational Impact: Your site feasibility must now account for the imaging capabilities required to detect these biomarkers (e.g., Susceptibility-Weighted Imaging or SWI). 

  1. The Data Integrity Shield: ICH-E6(R3)

Identifying these complex biomarkers is difficult. Relying on local site radiologists to interpret a “rim lesion” creates a massive variable in your safety data. 

This is where ICH-E6(R3) moves from a guideline to a strategic tool. R3 mandates “Data Governance” and the identification of Critical to Quality (CtQ) factors. 

  • The Gatekeeper Model: In a CAR-T trial, diagnostic specificity is a CtQ factor. You cannot rely on retrospective monitoring. 
  • Governance in Action: Accelsiors implements centralized imaging gatekeepers. Raw MRI data is transferred securely to central experts who verify the presence of CVS or PRL before the patient is randomized. 

This ensures that every patient exposed to the risks of CAR-T truly has the pathology that CAR-T is designed to treat. 

  1. The Human Factor: Patient Centricity in Early Diagnosis

There is a human implication to these changes that often gets lost in the discussion of biomarkers. 

By utilizing the 2024 criteria (specifically Kappa Free Light Chains or kFLC), we can diagnose patients earlier and with less invasive burden than the traditional lumbar puncture for Oligoclonal Bands. However, these “early” patients are often younger, active, and working full-time. They will not tolerate the high site burden of legacy protocols. 

 

ICH-E6(R3) as the Enabler 

R3 supports “fit-for-purpose” data collection. This allows us to implement Decentralized Clinical Trial (DCT) elements—such as home health nursing for safety labs or digital endpoints for progression tracking—to keep these patients retained in long-term CAR-T follow-up studies. 

 

What is your experience? 

As we move toward 2026, we are actively tracking how these decentralized elements impact patient retention in early-stage MS cohorts. We invite you to share your perspective: Are you currently implementing kFLC or home health visits to reduce site burden? 
>> Join the discussion on LinkedIn

 

The Convergence: Designing the Protocols of 2026 

You cannot have the “Therapeutic Leap” (CAR-T) without the “Diagnostic Leap” (McDonald 2024). If your protocol includes high-potency therapeutics but relies on 2017 diagnostic standards or 1996 monitoring habits, you are exposing your study to preventable safety and data integrity risks. 

 

Looking ahead to 2026? 

We are preparing a comprehensive guide to these operational changes. You can get an early preview of our 2025 MS Trial Playbook to start auditing your protocol today.