Background: Flow cytometric analysis of minimal residual disease (MRD) depends on detecting phenotypically abnormal populations. However, little is known about how phenotypic shifts between diagnosis and relapse affect MRD detection in childhood acute lymphoid leukemia (ALL).
Methods: We compared diagnostic and relapse bone marrow specimens in 42 children with precursor B-ALL studied with the two-tube panel CD19-APC/CD45-PerCP/CD10-PE/CD20-FITC and CD19-APC/CD45-PerCP/CD9-PE/CD34-FITC.
Results: At least 29 cases had phenotypic shifts of intensity or coefficient of variation of distribution of one or more markers. Shifts were complex and could not be explained by change in maturation stage. In the majority of cases MRD populations more closely resembled the diagnostic than the relapse specimen. In 6 of 7 MRD negative cases we did not identify an abnormal population that resembled diagnosis or relapse. In the remaining case, in which CD34 and CD10 were lost between diagnosis and relapse, it is possible that we could have missed an MRD population resembling relapse.
Conclusions: Phenotypic shifts are common, but do not affect MRD recognition. At most 1 of 42 cases might have harbored an abnormal population undetected because of shift. However, MRD analysis with rigid gating (looking strictly for abnormal phenotypes at diagnosis) might have missed many positive cases, 8 of 22 (36%) in this series.
(c) 2005 Wiley-Liss, Inc.