The heterogeneous course of multiple sclerosis (MS) makes diagnosis and prognosis difficult. Even though the 2017 revision of the McDonald criteria for MS diagnosis places greater emphasis on radiographic evidence, the criteria don’t always accurately predict disease course, activity, progression, recurrence, or treatment response.

Researchers at Brigham and Women’s Hospital previously developed the Multiple Sclerosis Disease Activity (MSDA) Test, a multiprotein, serum-based biomarker assay. In Clinical Immunology, Tanuja Chitnis, MD, co-director of the Multiple Sclerosis Center at Brigham and Women’s Hospital, and colleagues have reported clinical validation of the assay.

Background

Earlier this year, in Proteomics Clinical Applications, the research team described measuring disease activity using concentrations of 18 proteins from the serum of patients with all types of MS. The final algorithm calculated four disease pathway scores (immunomodulation, neuroinflammation, myelin biology, and neuroaxonal integrity) and an overall disease activity score. The MSDA Test was accurate, sensitive, precise, and robust.

Methods

The new study evaluated the associations of the overall disease activity score and the disease pathway scores with three measures of MS disease status: gadolinium-positive (Gd+) lesions, new or enlarging (N/E) T2 lesions, and active versus stable disease. A blood sample was considered active if any Gd+ or N/E T2 lesion was present or if the patient showed evidence of clinical relapse within 30 days of the blood draw.

614 blood samples were included: 448 retrospective samples from the Serially Unified Multicenter MS Investigation (SUMMIT) and 166 prospective samples from the Rocky Mountain Multiple Sclerosis Clinic in Salt Lake City, UT.

The multiprotein model was trained on 426 samples (70%) based on the presence versus absence of Gd+ lesions, as determined by MRI performed within 60 days of the blood draw. The model was then tested on the remaining 30% of samples.

Principal Results

The MSDA Test:

• Achieved an area under the receiver operating characteristic curve of 0.781 relative to the Gd+ lesion presence/absence endpoint, 0.750 relative to the N/E T2 lesion presence/absence endpoint, and 0.768 relative to the active/stable disease endpoint

• Outperformed each of the highest-performing single-protein models (NfL, CD6, CXCL13, interleukin-12β, and TNFSF13), indicating a more accurate representation of the complex pathways, processes, and cell types involved in MS

• Performed well when differentiating samples from patients with ≥1 Gd+ lesions versus no lesions as well as when differentiating samples from patients with ≥2 Gd+ lesions versus 0 or 1 lesions: the odds of having ≥1 Gd+ lesions with a moderate/high MSDA score were 4.49 times that of a low score, and the odds of having ≥2 Gd+ lesions with a high MSDA score were 20.99 times that of a low/moderate score

Toward Individualized Disease Management

The MSDA Test was designed to favor sensitivity over specificity. It can help identify patients with subradiographic and subclinical MS, permitting more timely treatment. As a serum-based assay, it can detect disease activity regardless of the location of lesions in the central nervous system.

Now that the test has been validated, it could be used as a routine surveillance test to better monitor disease activity and progression (e.g., distinguish inflammation from silent disease progression), especially in patients with stable disease. It should also be valuable for evaluating treatment responses and helping patients consider alternative treatment options.