Abstract Details

Accurate prediction of long-term disease outcome remains a challenge for patients with primary progressive multiple sclerosis (PPMS).

To develop blood transcriptome-based prognostic model to predict severe disease progression in untreated PPMS patients. 

Peripheral blood samples of PPMS patients included in the placebo-arm of the ORATORIO clinical trial (NCT01194570), were subjected to RNA-sequence analysis by Illumina NovaSeq S2. We applied 2-levels cross-validation algorithm (www.partek.com) to predict 12-weeks confirmed disability progression (12W CDP) during 120 weeks follow up, and the percent of brain volume change (PBVC) at 24, 48 and 120 weeks of follow-up.

RNA samples from 65 PPMS patients, age 43.9±1.4 years, female/male ratio 21/44, baseline EDSS 4.5±0.2 were analyzed. Correct classification rate of the 37% of patients with 12W CDP was 90.8% (95% CI 80.6 – 100.0%) using a 10-gene classifier. Correct classification rates of 63%, 57% and 64% of patients with 24, 48 and 120 weeks of brain volume change respectively, was 74.0%, (95% CI 65.1% – 83.0%), 84.0%, (95% CI 75.1%-93.1%) and 82.9% (95% CI 73.8%-95.6%), respectively.

Transcriptome data correctly predicts disability progression and brain volume change in untreated PPMS patients that could benefit from early treatment.