From the Ectrims site:

http://www.akm.ch/ectrims2006/

Friday, September 29, 2006, 15:30 - 17:00
Effectiveness of interferon-beta in delaying the age at secondary progression and at assignment of irreversible disability milestones in multiple sclerosis

M. Trojano, M.P. Amato, F. Pellegrini, D. Paolicelli, A. Fuiani, V. Zipoli, E. Di Monte, E. Portaccio, G.B. Zimatore, P. Livrea (Bari, Florence, Santa Maria Imbaro, Chieti, I)

-------------------------------------------------------------
Background: Recent studies suggest that the reaching of disability landmarks in Multiple Sclerosis (MS) is age-related, at least in part. Survival techniques which take account of the assessment of ages at assignment of Secondary Progression(SP)and irreversible disability milestones may provide accurate outcome for therapeutic trials.

Objective: To evaluate the effectiveness of IFN-beta in a cohort of Relapsing Remitting (RR) MS patients on two milestones: age at SP and age at EDSS 4.

Methods: A sample of 1,452 RR MS patients (1046 IFN-beta treated and 406 untreated) was prospectively followed,at the MS Centers of Bari and Florence, for at least 1 year from first visit up to 7 years(median follow-up = 4.7 years).

Ages at milestones were calculated as time from birth to each milestone and considered as survival time. To account for baseline imbalances between treated and untreated groups, a Cox regression model adjusted for propensity score (PS) quintiles was used. Estimated ages at milestones were obtained using PS adjusted survival curves. We also performed a sensitivity analysis to account for potential residual confounding deriving from the effect of an unmeasured binary covariate.

Results: PS adjusted Cox models for age at milestone SP showed that IFN-beta group vs.untreated group had a highly significant 70% reduction of incidence of SP during 7 years (HR=0.30, 95% CI 0.17-0.52, p<0.0001).

Milestone EDSS 4 also reported a significant 36% difference in favour of IFN-beta group (HR=0.64, 95% CI 0.43-0.97, p<0.05).PS adjusted survival curves translated HRs into the estimated delay of age at the two milestones. Milestones SP and EDSS 4 were reached by 30% of patients with 11 years(48.6 for untreated vs.59.8 for treated) and 4.2 years (45.7 for untreated vs. 49.9 for treated) of difference respectively between the two groups.

Sensitivity analysis for milestone SP showed that significant effect of IFN-beta might be altered by an unmeasured confounder with a HR of at least 4 with a prevalence imbalance between treated and untreated groups of at least 40%. As to milestone EDSS 4, an unmeasured confounder with HR > = 2 and 10% of imbalance would be sufficient to alter the significant effect of treatment.

Conclusions: The results of this observational study in a large MS population demonstrate that IFN-beta treatment delays the age at assignment of SP and irreversible EDSS 4. However appropriate statistical analyses need to properly estimate the real treatment effectiveness.