Figure 4

Lipids differentially abundant in the plasma of PPMS patients compared with controls. (A-D,F) Graphs show normalized levels of differentially abundant lipids in the plasma of healthy controls, PPMS-NP and PPMS-P patients. The levels of (A) Di-hexosylceramide(d18:1/18:2) [DiHexCer(d18:1/18:2)], (B) DiHexCer(d18:1/18:3), (C) phosphatidylethanolamine-O-34:3 (PE-O-34:3), and (F) sphingomyelin SM(d18:1/14:0) were significantly lower in the plasma of PPMS patients compared with controls, while the levels of (C) mono-hexosylceramide(d18:1/20:0) [MonoHexCer(d18:1/20:0)] were significantly higher. Boxes indicate the interquartile range, horizontal lines indicate group medians, whiskers connect the lowest and the highest observations. The statistical analysis was performed using the One-way ANOVA with Dunnet’s multiple comparison test (*p < 0.05, **p < 0.01, ***p < 0.001). Control, n = 8; PPMS-NP, n = 10; PPMS-P, n = 9. (G and F) Graphs show correlations between normalized plasma lipid levels of (E) MonoHexCer(d18:1/20:0) and (G) SM(d18:1/14:0) with changes in MRI metrics at the 1-year follow-up. Changes in MRI volumes were computed as follow-up values minus baseline values. (E) Higher abundance of MonoHexCer(d18:1/20:0) in the plasma of PPMS patients correlated with more severe brain volume loss (calculated as percentage brain volume change, PBVC). (G) Higher levels of SM(d18:1/14:0) correlated with less severe loss of CblWMV. Lines represent the linear regressions of all patients (black), PPMS-NP (green), and PPMS-P (blue) subgroups. Correlations were assessed using both the Pearson and Spearman methods. R-coefficient (r) and p-values (p) are indicated in black for the entire PPMS population, in green and blue for the PPMS-NP and PPMS-P group respectively. P < 0.05 (in bold) was considered significant. PPMS-NP, n = 10; PPMS-P, n = 9.