Analytical validation of a multi-protein, serum-based assay for disease activity assessments in multiple sclerosis

Abstract

Purpose: To characterize and analytically validate the MSDA Test, a serum-based multiplex protein biomarker assay developed using Olink® PEA methodology. Experimental design: Two lots of the MSDA Test panel were manufactured and subjected to a comprehensive analytical characterization and validation protocol to detect biomarkers present in the serum of patients with MS. Biomarker concentrations were incorporated into a final algorithm used for calculating four Disease Pathway scores (Immunomodulation, Neuroinflammation, Myelin Biology and Neuroaxonal Integrity) and an overall Disease Activity score. Results: Analytical characterization demonstrated that the multiplex panel satisfied the criteria necessary for a fit-for-purpose validation considering the assay's intended clinical use. This panel met acceptability criteria for 18 biomarkers included in the final algorithm out of 21 biomarkers evaluated. VCAN was omitted based on factors outside of analytical validation; COL4A1 and GH were excluded based on imprecision and diurnal variability, respectively. Performance of the four Disease Pathway and overall Disease Activity scores met the established acceptability criteria. Conclusions and clinical relevance: Analytical validation of this serum-based proteomic multiplex assay is the first step in establishing its potential utility as a quantitative, minimally invasive, and scalable biomarker panel to enhance the standard of care for patients with MS.

Association between Serum Biomarker Profile and Real-World Evidence of Disability in Multiple Sclerosis

“…Leveraging the Proximity Extension Assay (PEA) methodology on the Olink™ platform, 28 a prior study reported the analytical characterization of a custom serum-based proteomic multiplex immunoassay (PMI) panel (including sNEFL and sGFAP) associated with inflammatory MS disease activity and pertaining to key biological pathways in MS pathogenesis. 29 Here, we deployed statistical learning methods to examine the association between serum biomarker profiles using the custom PMI panel and patient-reported disability in pwMS. Specifically, we hypothesized that serum protein profiles informative of inflammatory MS disease activity would also improve the prediction of real-world neurological disability when compared to clinical profiles or single serum protein.…”

Clinical Validation of a Multi-protein, Serum-based Assay for Disease Activity Assessments in Multiple Sclerosis

“…Full details on the algorithm and model parameters are presented elsewhere. 23 Single-protein models were fit using L2-penalized logistic regression with presence or absence of Gd+ lesions as the dependent variable and an intercept and the protein biomarker as independent variables. Protein concentrations were limit of quantitation (LOQ)-imputed, log 10transformed, and demographically adjusted (with age and sex, based on Ordinary Least Squares [OLS] modeling) prior to being used in the Disease Pathway, Disease Activity, and single-protein models.…”

“…Previous research and development studies, samples from a cohort of healthy controls, and those from the Train subset were used to establish the biomarker-specific demographic adjustment strategy, which included removing protein concentration outliers, accounting for OLS coefficient sign consistency across the three studies and establishing statistical significance related to both age and sex. 23 Metrics for model performance including the area under the receiver operating characteristic (AUROC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and odds ratios were used to evaluate model performance. The prevalence of Gd+ lesions was enriched in this dataset, and it is important to note that PPV, NPV, and accuracy all depend on the prevalence.…”

“…There are currently no validated clinical tests that leverage multiple serum biomarkers to monitor disease activity or progression in patients with MS. We have previously established that the MSDA Test is accurate, sensitive, precise, and robust, which serves as a critical first step in the validation of this assay. 23 The MSDA Test uses 18 protein biomarkers, which reflect various biological pathways associated with MS pathophysiology.…”

“…Previous research and development studies, samples from a cohort of healthy controls, and those from the Train subset were used to establish the biomarker-specific demographic adjustment strategy, which included removing protein concentration outliers, accounting for OLS coefficient sign consistency across the three studies and establishing statistical significance related to both age and sex. 23…”