Comparison of multidimensional flow cytometry with standard morphology for evaluation of early marrow response in pediatric acute lymphoblastic leukemia.

Abstract

We compared multidimensional flow cytometry (MDF) with morphology in evaluating early marrow response to induction chemotherapy in pediatric ALL. Chemotherapy response was determined by standard morphology or by MDF assessed by residual leukemic cell percentage remaining in the marrow on days 7, 14, and 28 of induction. Bone marrow response was classified as M3 (>25% leukemic blasts) or M1/M2 (< or = 25% leukemic blasts). Multidimensional flow cytometry evaluation was compared with that of standard morphology. Available day-7 and day-14 marrow slides were also reevaluated by a single pathologist without patients' clinical information. Of 46 day-7 specimens, eight (17%) had discordant MDF and morphologic results (P < 0.001), including six classified as M3 by morphology but were M1/M2 by MDF, and two were classified as M3 by MDF but were M1/M2 by morphology. Of 24 day-14 bone marrow specimens, five (20.5%) were discordant (P < 0.001), including two classified as M3 by morphology but were M1/M2 by MDF, and three were classified as M3 by MDF but were M1/M2 by morphology. Reevaluation of the blinded day-7 and day-14 marrow slides yielded discordance between repeated pathology readings of 11% (P < 0.001) and 6% (P = 0.04), respectively. Our data show significant discordance between the morphologic and MDF evaluation of early marrow response. Early response to therapy is a significant prognostic indicator in pediatric acute lymphoblastic leukemia and is used to alter subsequent treatment; thus, precise assessment of response is important. A larger comparison of MDF with morphology for the evaluation of early response, including correlation with clinical outcome, is warranted.