Exploring miRNAs' Based Modeling Approach for Predicting PIRA in Multiple Sclerosis: A Comprehensive Analysis

Abstract

The current hypothesis on the pathophysiology of multiple sclerosis (MS) suggests the involvement of both inflammatory and neurodegenerative mechanisms. Disease Modifying Therapies (DMTs) effectively decrease relapse rates, thus reducing relapse-associated disability in people with MS. In some patients, disability progression, however, is not solely linked to new lesions and clinical relapses but can manifest independently. Progression Independent of Relapse Activity (PIRA) significantly contributes to long-term disability, stressing the urge to unveil biomarkers to forecast disease progression. Twenty-five adult patients with relapsing-remitting multiple sclerosis (RRMS) were enrolled in a cohort study, according to the latest McDonald criteria, and tested before and after high-efficacy Disease Modifying Therapies (DMTs) (6-24 months). Through Agilent microarrays, we analyzed miRNA profiles from peripheral blood mononuclear cells. Multivariate logistic and linear models with interactions were generated. Robustness was assessed by randomization tests in R. A subset of miRNAs, correlated with PIRA, and the Expanded Disability Status Scale (EDSS), was selected. To refine the patient stratification connected to the disease trajectory, we computed a robust logistic classification model derived from baseline miRNA expression to predict PIRA status (AUC = 0.971). We built an optimal multilinear model by selecting four other miRNA predictors to describe EDSS changes compared to baseline. Multivariate modeling offers a promising avenue to uncover potential biomarkers essential for accurate prediction of disability progression in early MS stages. These models can provide valuable insights into developing personalized and effective treatment strategies.

Keywords: miRNA; modeling approach; multiple sclerosis; progression independent of relapse activity (PIRA).

Figure 1

Sample heatmap visualization. The heatmap plot is based on Log2 normalized and standardized expression data (zero-centered, SD = 1.0) of the 9 miRNA genes significantly correlated with PIRA status (0/1). Samples are labeled according to PIRA.

Figure 2

Principal component analysis of samples. The PCA plot is based on Log2 normalized expression data of the 9 miRNA genes significantly correlated with PIRA status and EDSS score change. Samples are labeled according to PIRA.

Figure 3

EDSS score variation and trajectory based on PIRA. (A) EDSS trajectory based on PIRA status. Subjects are divided according to the PIRA status and compared using the Mann–Whitney two-sided test. (*) p < 0.05. (B) EDSS score variation between T4 and T0 time points. Subjects are divided according to the PIRA status. The two groups are compared using the Mann–Whitney two-sided test. (****) p < 0.0001.

Figure 4

(A) Predicted scores obtained by the logistic model. The horizontal dashed line corresponds to the cut-off, estimated as the maximum Youden index in the corresponding ROC curve, between the two levels 0/1 = negative/positive of PIRA status based on actual clinical data. Negative PIRA subjects in clinical data are plotted in blue and similarly positive ones in red. All model coefficients, including intercept, are significant (Wald test, p < 0.05); see Equation (1). (B) ROC curve of the binary classifier logistic model. The AUC is significantly large, according to reference null distributions obtained by randomizing PIRA status (p < 0.01) or miRNA predictors (p < 0.05); see Section 4.

Figure 5

Correlation plots between selected miRNAs and EDSS (T4-T0). The four miRNAs used for the multilinear regression model in Equation (2) are significantly correlated to EDSS change at T4. The linear regression line in red is surrounded by the confidence interval in dark gray. MiRNA and EDSS data are normally distributed (Shapiro–Wilk test and Kolmogorov–Smirnov test, p < 0.05).

Figure 6

Multilinear model to predict EDSS (T4-T0). The multilinear model is based on the same four miRNA predictors used for the logistic model. The gray band represents the confidence interval around the linear regression line in red. The Pearson correlation between actual data and prediction is significant (p < 0.00001); see Equation (2). Prediction values and EDSS data are normally distributed (Shapiro–Wilk test and Kolmogorov–Smirnov test, p < 0.05).