TL;DR
This study reveals that non-clinical markers, specifically new MRI activity and serum neurofilament light chain (sNfL) levels, are highly effective at distinguishing between patients receiving natalizumab and placebo for multiple sclerosis (MS). These markers, when combined, offer a more practical and potentially superior approach to monitoring treatment response compared to traditional clinical measures, paving the way for more personalized MS management.
What Problem Does This Paper Address?
Multiple sclerosis (MS) is a complex autoimmune disease affecting the central nervous system, characterized by periods of inflammation and neurodegeneration. Effectively managing MS requires diligent monitoring of disease activity and treatment response to preserve neurological function. While clinical measures like relapses and disability progression (assessed by EDSS) have been standard, they have limitations. Composite measures like "No Evidence of Disease Activity" (NEDA) have emerged, but concerns remain about their practicality and optimal composition for individual patient monitoring. The field is moving towards precision medicine, necessitating outcome measures that are quantitative, reproducible, and applicable across diverse clinical settings. This paper addresses the need to identify the most effective measures for assessing treatment response in MS, particularly in the context of potent disease-modifying therapies (DMTs) like natalizumab, and to explore whether non-clinical markers can outperform or complement traditional clinical assessments.
Methods
The researchers conducted post-hoc analyses on data from the AFFIRM trial, a large, 2-year, randomized, placebo-controlled study of natalizumab for relapsing MS. They employed multivariable logistic regression to assess the independent contributions of various clinical and non-clinical measures in distinguishing between the natalizumab and placebo groups. The key predictors included:
- Clinical Measures: Relapses, Expanded Disability Status Scale (EDSS) progression, and neuroperformance measures (PASAT, 9HPT, T25FW).
- Non-Clinical Measures: New or enlarging T2 or gadolinium-enhancing (Gd+) lesions on MRI, brain atrophy (annualized whole brain volume loss - BVL), and serum neurofilament light chain (sNfL) levels.
sNfL levels were measured from stored serum samples and compared against age-normative 97.5th percentiles. The analysis focused on identifying measures that independently discriminated between treatment groups and then built models to find the best combination of predictors. Tenfold cross-validation was used to ensure model robustness. They also compared these findings to established NEDA composite measures.
Datasets
The primary dataset used was from the AFFIRM trial, a randomized, placebo-controlled study of natalizumab in adults with relapsing MS. This trial provided data on clinical relapses, EDSS scores, MRI scans, and neuroperformance measures collected over two years. Additionally, stored serum samples from AFFIRM participants were used for sNfL measurements. A separate dataset of 130 healthy controls from Johns Hopkins University was used to establish age-normative sNfL values.
Key Insights and Ideas
- Non-Clinical Dominance: The study strongly suggests that non-clinical markers—specifically MRI activity and sNfL levels—are powerful discriminators of treatment response, potentially more so than traditional clinical measures alone. This shifts the paradigm towards objective, measurable biological and imaging markers.
- sNfL as a Key Biomarker: Serum neurofilament light chain (sNfL) emerges as a crucial indicator of neuroaxonal injury and treatment response. Its ability to distinguish between treatment groups, even when used alone, highlights its potential as a readily accessible blood-based biomarker.
- MRI Activity's Continued Importance: New or enlarging T2 or Gd+ lesions on MRI remain a critical indicator of inflammatory disease activity and a strong predictor of treatment response.
- Limited Value of Traditional Progression Measures: Measures of disease progression like EDSS worsening, brain atrophy, and neuroperformance decline showed less independent discriminatory power in the early (2-year) treatment period. This implies that potent anti-inflammatory DMTs may suppress inflammatory activity so effectively that downstream effects like atrophy are not immediately apparent or are secondary to inflammation.
- Practicality of New Composites: The proposed two- and three-component composites (combining MRI activity and sNfL, with relapses added for the latter) demonstrated comparable or superior discrimination to NEDA-3 and NEDA-4, suggesting they could be more practical for routine clinical monitoring.
Key Results
- The best-fitting multivariable model for distinguishing natalizumab from placebo included no new or enlarging T2 or Gd+ MRI activity (OR: 7.2), sNfL levels below the 97.5th percentile at year 2 (OR: 4.1), and no relapses in years 0-2 (OR: 2.1).
- A two-component model consisting of no MRI activity and sNfL <97.5th percentile at year 2 performed nearly as well as the three-component model, with an AUC of 71.3 for the composite.
- The addition of relapses to the MRI activity and sNfL model only marginally increased the AUC from ~78.9 to ~80.8, indicating that MRI activity and sNfL alone captured most of the discriminatory power.
- Measures like EDSS progression, T25FW worsening, and brain atrophy had significantly lower odds ratios and AUCs, suggesting they were less effective discriminators in this 2-year trial context.
Limitations and Future Work
- Short-Term Data: The study is based on 2-year data from a clinical trial. Longer-term follow-up is needed to confirm the association of these measures with long-term progression and to assess their utility beyond the initial treatment phase.
- Clinical Trial vs. Real-World: The findings are from a post-hoc analysis of a clinical trial. Confirmation in real-world settings across a broader range of MS subtypes is necessary.
- Natalizumab Specificity: While the study focuses on natalizumab, the generalizability of these findings to other DMTs with different mechanisms of action needs further investigation.
- Antibody Development: The study did not exclude patients who developed antibodies to natalizumab, which could influence treatment response.
- Future Directions: The authors suggest that new and enlarging T2 lesions and sNfL levels could become standardized metrics for monitoring individual patient responses to anti-inflammatory therapy in the early stages of MS. Further research should focus on the technical and clinical validation of sNfL, understanding confounding variables, and establishing robust age-related reference values to integrate these biomarkers into clinical practice.
Summary updated