Prediction of risk of disease recurrence by genome-wide cDNA microarray analysis in patients with Philadelphia chromosome-positive acute lymphoblastic leukemia treated with imatinib-combined chemotherapy

Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ALL) reveals very poor prognosis due to high incidence of relapse when treated with standard chemotherapy. Although >96% of patients with Ph+ALL achieved complete remission (CR) with imatinib-combined chemotherapy in a phase II clinical trial conducted by the Japan Adult Leukemia Study Group (JALSG), 26% of them experienced hematological relapse in a short time after achievement of CR. In this study, to establish a prediction system for risk of relapse, we analyzed gene expression profiles of 23 bone marrow samples from patients with Ph+ALL using cDNA microarray consisting of 27,648 cDNA sequences. Using the 19 randomly-selected test cases, we identified 16 genes that were expressed significantly differently between patients with (n=8) and without (n=11) continuous response; from the list of 16 genes, we selected the 6 'predictive' genes and constructed a numerical scoring system by which the two groups were clearly separated, with positive scores for the former and the negative scores for the latter. Scoring of 4 cases that were reserved from the original 23 cases predicted correctly their clinical responses. In addition, three cases whose BCR-Abl transcript levels failed to reduce sufficiently after induction therapy, also revealed negative scores. We also developed a quantitative reverse transcription-PCR-based prediction system that could be feasible for routine clinical use. Our results suggest that achievement of continuous response with imatinib-combined chemotherapy can be predicted by expression patterns in this set of genes, leading to achievement of 'personalized therapy' for treatment of this disease.