The revised El Escorial criteria "clinically probable laboratory supported ALS"—once a promising now a superfluous category?. Issue 1 (2nd January 2020)

• Record Type: Journal Article
• Authors: Braun, Nathalie; Macklin, Eric A.; Sinani, Ervin; Sherman, Alexander; Weber, Markus
• Is Part Of: Amyotrophic lateral sclerosis and frontotemporal degeneration. Volume 21:Issue 1/2(2020)
• Page Start: 24
• Page End: 28
• Publication Date: 2020-01-02
• Subjects: Amyotrophic lateral sclerosis -- El Escorial criteria -- Edaravone -- PRO-ACT dataset -- slow progressors; 616.839
• DOI: 10.1080/21678421.2019.1666875
• Languages: English
• ISSNs: 2167-8421

Abstract

Abstract: Over the past two decades, the El Escorial criteria (EEC) have been used as eligibility criteria in major randomized controlled trials. One of the goals of the revised EEC was to allow earlier diagnosis and, thus earlier trial inclusion by introducing a new category, namely "clinically probable laboratory supported" ALS. This category allowed EMG findings to be taken into account assuming that EMG is more sensitive than the clinical examination in detecting lower motor neuron signs. Recently, Edaravone has been licensed in several countries for the treatment of ALS based on a randomized controlled trial in a selected group of ALS patients excluding the EEC category "clinically probable laboratory supported". The major reason was that in a post hoc analysis of the first Edaravone trial this group comprised many slow progressors. As it is unclear whether this bias towardslow progressors was a study-specific problem or related to the category itself, we performed an analysis in the PRO-ACT dataset. In the PRO-ACT dataset, progression in ALS patients included at baseline into the "clinically probable laboratory supported" category was significantly slower (–0.53 in ALSFRS/month) compared to the other EEC categories (–0.68 in ALSFRS/month; p < 0.001) and exhibited a significantly longer diagnostic delay (13.5 months vs. 11.7 months, p < 0.001). This suggests that the bias toward slow progressors in the "clinically probable laboratory supported" category is an inherent problem of the category and thus does not fulfill the previous goal of earlier diagnosis, raising several questions concerning the application of this category.