Studies proposing blood biomarkers for MS prognosis.

Eighteen studies nominated 110 mRNA or proteins that, when measured in the blood at a single time point, may allow the prediction of an individual's course of the disease. The table provides study details, e.g., number of patients and technology used to quantify gene expression, as well as information on how long-term disease progression was evaluated. In the column “Number of Genes”, the entry “Suggested” gives for each study the number of genes that were proposed as predictive markers. “Affymetrix custom” gives the number of custom probe sets that detect the suggested genes. Custom probe sets uniquely relate to GeneCard genes and are defined by a custom CDF, which we used to preprocess our Affymetrix HG-U133 A and B microarray data. Genes for which no specific custom probe set exists are given in brackets in the rightmost column. “Affymetrix original” gives the number of probe sets according to the original CDF, which was used by Gurevich et al. to preprocess their Affymetrix HG-U133 A and A 2.0 data. Some genes are assayed by more than one probe set in the original annotation. Genes described as being predictive concordantly by more than one study are written in bold in column “List of Genes”. In addition to reassessing the prognostic value of the listed genes, we included the cytokines IL17A-E and the receptors IL17RA-E in the analysis. In total, we examined the expression signals of 112 different genes that are represented by 112 custom probe sets in our data and 204 original probe sets in the data by Gurevich et al., which we used for further validation of the results.