Melanie Titze (1), Annemijn Jonkman (1), Magnus Munch (1), H. Maxime Lagraauw (1,2), Piet H. van der Graaf (2), Bernard Muller (1), Inez de Greef (1)
(1) Treeway B.V., Tilburg, The Netherlands; (2) Systems Pharmacology, Leiden Academic Centre for Drug Research, Leiden University, Leiden, The Netherlands.
Background: Amyotrophic lateral sclerosis (ALS) is a progressive and deadly motor neurone disease, in which the motor neurons innervating skeletal muscle degenerate. World-wide it affects around 400.000 patients and their families. The average life expectancy upon diagnosis with ALS is only 3-5 years, during which muscle function gradually declines and eventually leads to respiratory dysfunction and death. Progression of the disease is monitored by means of the ALS Functional Rating Scale (ALSFRS), evaluating the ability to perform several physical tasks. However, disease progression, as assessed by this symptomatic scale, is highly variable between patients, thus complicating accurate estimation of patient’s prognosis and clinical trial design.
Objectives: During the 10-week Treeway Summer Challenge 2015, a group of 9 MSc. and PhD students aimed to gain more insight in the complex pathobiology of ALS and to develop a disease progression model to better understand patient heterogeneity and predict disease progression on population and individual level.
Methods: Longitudinal ALSFRS scoring data was obtained from the Pooled Resource Open-Access ALS Clinical Trials (PRO-ACT) Database. Patients from both active and placebo treatment arms, but not those on riluzole, were included. Using these criteria data from 1065 ALS patients and a total of 8174 ALSFRS evaluations were available for the fitting of the longitudinal disease progression model.
Results: Similar to the only available ALS disease progression model in literature [1], disease progression was best described by the Weibull function. Subsequent stepwise covariate model building resulted in the inclusion of baseline ALSFRS (BAS) on all three function parameters and site of onset as a covariate on the time-dependent parameter (td). With this model the occurrence of two distinct subpopulations could be explained by the site of onset: bulbar- vs. limb-onset.
Conclusions: The current model provided an adequate model to describe the longitudinal disease progression as measured by the ALSFRS score and could distinguish two subpopulations with distinct disease progression rates based on site of onset. Still, there remains a high degree of interindividual variability which should be explored in further research.
References:
[1] Gomeni R, Fava M, PRO-ACT Consortium. Amyotrophic lateral sclerosis disease progression model. Amyotroph Lateral Scler Front Degener. 2014 Mar 1;15(1-2):119–29.
Reference: PAGE 25 (2016) Abstr 5984 [www.page-meeting.org/?abstract=5984]
Poster: Drug/Disease modeling - CNS