Therapeutic lag: Is treatment effect delayed in progressive MS?

Area of Science:

• Neurology
• Clinical Trials
• Biostatistics

Background:

• Randomized clinical trials (RCTs) for progressive multiple sclerosis (MS) often fail to show significant treatment effects on disability progression.
• This lack of significance may stem from a delayed onset of treatment efficacy.
• Baseline characteristics might influence this delay.

Purpose of the Study:

• To investigate if a delayed treatment effect explains the non-significant findings in progressive MS RCTs.
• To analyze the impact of baseline characteristics on the timing of treatment benefits.

Main Methods:

• Re-analysis of data from two RCTs (SPECTRIMS and PROMISE) involving interferon-beta and glatiramer-acetate in progressive MS.
• Application of a time-dependent Cox model allowing for a delayed treatment effect (t0).
• Modeling the treatment delay as a function of baseline Expanded Disability Status Scale (EDSS) scores.

Main Results:

• A time-dependent Cox model revealed significant treatment benefits in both studies when a delay was incorporated (SPECTRIMS: t0=2.5 years, HR=0.65; PROMISE: t0=2 years, HR=0.65).
• In the merged dataset, an EDSS-dependent delayed effect showed a significant benefit (HR=0.68, p<0.001).
• The assumption of a delayed treatment effect improved model fit, indicating a significant, albeit shifted, treatment benefit.

Conclusions:

• Incorporating a delayed treatment effect assumption significantly improved the analysis of progressive MS RCT data.
• This approach uncovered significant treatment benefits that were not apparent with standard analysis, suggesting a need to consider delayed efficacy in trial design and interpretation.