Prognostic value of immunophenotypic detection of minimal residual disease in acute lymphoblastic leukemia.

PURPOSE The identification of immunophenotypic aberrancies through multiparametric flow cytometry makes the differentiation between normal and leukemic cells relatively simple and quick, and is therefore an attractive method for the investigation of minimal residual disease (MRD). In this report, we have analyzed the impact on relapse and relapse-free survival (RFS) of detecting immunophenotypical aberrant cells in acute lymphoblastic leukemia (ALL) patients in cytomorphologic complete remission (CR). MATERIALS AND METHODS Two hundred eleven bone marrow (BM) samples from 53 consecutive ALL (37 precursor B-ALL and 16 T-ALL) patients were analyzed. The only selection criteria were to have at least one aberrant immunophenotypic feature at diagosis and to have achieved cytomorphologic CR after induction therapy. For MRD detection, all follow-up samples were analyzed with triple labelings using a two-step acquisition procedure, in which 106 cells were screened for the possible persistence of residual leukemic cells with the same phenotypic aberrancy as that identified diagnosis. RESULTS Patients who displayed a gradual increase in MRD levels showed a higher relapse rate (90% v22%; P < .00001) and shorter median RFS (12 months v not reached; P < .0001) than those with stable or decreasing MRD levels. This adverse prognostic influence also was observed when children and adults, as well as B-ALL and T-ALL patients, were analyzed separately. An MRD level > or = or greater than 10(-3) discriminated two risk groups of ALL patients with significantly different relapse rates and RFS at all treatment phases (end of induction, consolidation, maintenance, and out of treatment). CONCLUSION Multiparametric flow cytometry of MRD in ALL patients is a valuable tool for relapse prediction and for the identification of a cohort of patients with very poor prognosis.