Background: Although cognitive evaluation has been incorporated in recent MS clinical trials, the definition of cognitive progression is not clear and recent data are questioning the 4-point cutoff using the SDMT at the individual level. We aimed to evaluate the behavior of cognitive performance over time using different cutoffs.
Methods: Cognitive performance over six years was analyzed in a cohort of 42 relapsing-remitting MS patients and 30 controls using verbal/visual memory and information processing speed tests. Fixed cutoffs were: 10% and 20% change (all tests) and a 4- and 8-point change (SDMT). The relative cutoff established by regression-based models was a 1SD change.
Results: The distributions of "worsening", "stability", and "improvement" showed low concordance rates across the cutoffs (p < 0.001 for most comparisons). Most patients classified with worsening initially using fixed cutoffs had subsequent improvement in all cognitive tests, yielding a low sensitivity to predict later cognitive worsening. Using the relative cutoff, the proportion of patients with subsequent improvement was noticeably smaller.
Conclusions: Fixed cutoffs classify a high proportion of patients with cognitive improvement. Most patients categorized with worsening initially presented subsequent improvement. Instead, the relative cutoff generally had a better performance. These data raise concerns about how we are defining cognitive worsening so far, especially at the individual level.
Keywords: Cognition; Disease progression; Multiple sclerosis; Outcome assessment; Prognosis.
Copyright © 2022 Elsevier B.V. All rights reserved.