Description
We compared the gene expression profile from a group of children with T-cell acute lymphoblastic leukamia who remained in continuous complete remission (CCR) (n = 7) with that from a group who relapsed (n = 5), using Affymetrix HG-U133A arrays. Using the decision-tree based supervised learning algorithm Random Forest (RF), genes were ranked with respect to their ability to discriminate between patients who remained in CCR and those who relapsed. From the 300 top-ranked probe sets 9 genes were selected for further investigation and validation in an independent cohort of 25 T-ALL patients using quantitative real time polymerase chain reaction.