Disease phenotype prediction in multiple sclerosis
Stephanie Herman,
Staffan Arvidsson McShane,
Christina Zjukovskaja,
Payam Emami Khoonsari,
Anders Svenningsson,
Joachim Burman,
Ola Spjuth,
Kim Kultima
Affiliations
Stephanie Herman
Department of Medical Sciences, Clinical Chemistry, Uppsala University, Uppsala, Sweden; Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Staffan Arvidsson McShane
Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Christina Zjukovskaja
Department of Neuroscience, Neurology, Uppsala University, Uppsala, Sweden
Payam Emami Khoonsari
Department of Medical Sciences, Clinical Chemistry, Uppsala University, Uppsala, Sweden; Department of Biochemistry and Biophysics, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Stockholm University, Box 1031, 17121 Solna, Sweden
Anders Svenningsson
Department of Clinical Sciences, Danderyd Hospital, Karolinska Institutet, Stockholm, Sweden
Joachim Burman
Department of Neuroscience, Neurology, Uppsala University, Uppsala, Sweden
Ola Spjuth
Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Kim Kultima
Department of Medical Sciences, Clinical Chemistry, Uppsala University, Uppsala, Sweden; Corresponding author
Summary: Progressive multiple sclerosis (PMS) is currently diagnosed retrospectively. Here, we work toward a set of biomarkers that could assist in early diagnosis of PMS. A selection of cerebrospinal fluid metabolites (n = 15) was shown to differentiate between PMS and its preceding phenotype in an independent cohort (AUC = 0.93). Complementing the classifier with conformal prediction showed that highly confident predictions could be made, and that three out of eight patients developing PMS within three years of sample collection were predicted as PMS at that time point. Finally, this methodology was applied to PMS patients as part of a clinical trial for intrathecal treatment with rituximab. The methodology showed that 68% of the patients decreased their similarity to the PMS phenotype one year after treatment. In conclusion, the inclusion of confidence predictors contributes with more information compared to traditional machine learning, and this information is relevant for disease monitoring.
