Long Term Safety of Teriflunomide When Added to Interferon-Beta or Glatiramer Acetate in Patients With Multiple Sclerosis

AE are any unfavorable and unintended sign, symptom, syndrome, or illness observed by the investigator or reported by the participant during the study.

AE with potential risk of occurrence were defined as follows:

• Hepatic disorders;
• Immune effects, mainly effects on bone marrow and infection;
• Pancreatic disorders;
• Malignancy;
• Skin disorders, mainly hair loss and hair thinning;
• Pulmonary disorders;
• Hypertension;
• Peripheral neuropathy;
• Psychiatric disorders;
• Hypersensitivity.

PCSA values are abnormal values considered medically important by the Sponsor according to predefined criteria based on literature review.

Hepatic parameters thresholds were defined as follows:

• Alanine Aminotransferase [ALT] >3, 5, 10 or 20 Upper Normal Limit [ULN];
• Aspartate aminotransferase [AST] >3, 5, 10 or 20 ULN;
• Alkaline Phosphatase >1.5 ULN;
• Total Bilirubin [TB] >1.5 or 2 ULN;
• ALT >3 ULN and TB >2 ULN;

ARR is obtained from the total number of confirmed relapses that occured during the treatment period divided by the sum of the treatment durations.

Each episode of relapse - appearance, or worsening of a clinical symptom that was stable for at least 30 days, that persisted for a minimum of 24 hours in the absence of fever - was to be confirmed by an increase in Expanded Disability Status Scale [EDSS] score or Functional System scores.

To account for the different treatment durations among participants, two Poisson regression models with robust error variance were used (total number of confirmed relapses as response variable, log-transformed treatment duration as "offset" variable and:

• Model 1 (IFN-β groups): treatment group, region of enrollment and IFN-β dose level as covariates
• Model 2 (GA groups): treatment group and region of enrollment as covariates)

12-week sustained disability progression was defined as an increase from baseline of at least 1-point in EDSS score (at least 0.5-point for participants with baseline EDSS score >5.5) that persisted for at least 12 weeks.

If no disability progression was observed on or before last EDSS evaluation before study drug discontinuation, then the participant was considered as free of disability progression.

Probability of disability progression at 24 and 48 weeks was estimated using Kaplan-Meier method on the time to disability progression defined as the time from randomization to first EDSS increase. Participants free of disability progression were censored at the date of the last on-treatment EDSS evaluation.

Kaplan-Meier method consists in computing probabilities of non occurrence of event at any observed time of event and multiplying successive probabilities for time ≤t by any earlier computed probabilities to estimate the probability of being event-free for the amount of time t. Probability of event at time t is 1 minus the probability of being event-free for the amount of time t.

Total lesion volume is the sum of the total volume of all T2-lesions and the total volume all T1-hypointense post-gadolinium lesions measured through T2/proton density scan analysis and gadolinium-enhanced T1 scan analysis.

Least-square means were estimated using two Mixed-effect models with repeated measures [MMRM] on cubic root transformed volume data:

• Model 1 (IFN-β groups): treatment group, region of enrollment, IFN-β dose level, visit, treatment-by-visit interaction, baseline value (cubic root transformed), and baseline-by-visit interaction as factors;
• Model 2 (GA groups): treatment group, region of enrollment, visit, treatment-by-visit interaction, baseline value (cubic root transformed), and baseline-by-visit interaction as factors.

Number of Gd-enhancing T1-lesions per scan is obtained from the total number of Gd-enhancing T1-lesions observed during the study divided by the total number of scans performed during the study.

To account for the different number of scans among participants, two Poisson regression models with robust error variance were used (total number of Gd-enhancing T1-lesions as response variable, log-transformed number of scans as "offset" variable and:

• Model 1 (IFN-β groups): Treatment group, region of enrollment, IFN-β dose level and baseline number of Gd-enhancing T1-lesions as covariates
• Model 2 (GA groups): Treatment group, region of enrollment and baseline number of Gd-enhancing T1-lesions as covariates)

Total volume of Gd-enhancing T1-lesions per scan is obtained from the sum of the volumes of Gd-enhancing T1-lesions observed during the study divided by the total number of scans performed during the study.