Joint Analysis of Two Distinct but Interrelated Disease Processes in Multiple Sclerosis

• Activity Number: 549
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Tristan Massie*
• Companies: FDA
• Keywords: joint distribution modeling ; interval censoring ; conditional ; multiple sclerosis

Abstract

This paper investigates a joint model of two disease processes: a random temporary flare of disease and a true progression of disease. For example, analysis of progression of disability in Relapsing Remitting Multiple Sclerosis is complicated by the effects of transient relapses on the efficacy measure. The most common analysis is a time to event analysis in which the event is a confirmed progression which is defined as having at least a 1 point increase in the EDSS disability measure above baseline at two consecutive visits separated by 3 months. This analysis requires that confirmation of a suspected progression must occur at a time when a relapse of disease is not ongoing. The relapses occur at random. A joint model of the disease progression and the random transient relapse process is developed. Progression is not evaluated continuously in practice, so the effects of interval censoring on the traditional analysis is also investigated. Comparisons are made between the models which suggest the new joint model may be preferable.