Evaluation of Early Marrow Response in Childhood Aneuploid Acute Lymphoblastic Leukemia: Flow Cytometric DNA Analysis versus Standard Morphology

Abstract

Despite improved survival rates for childhood acute lymphoblastic leukemia (ALL), the relapse rate remains at 20–30%. Early peripheral blood and bone marrow (BM) responses have been associated with more favorable outcomes; all current children's cancer group (CCG) protocols for ALL require BM evaluation at days 7 and 14 with subsequent therapy based on the results. Morphologic interpretation of aspirate smears during induction chemotherapy is challenging, as the samples are often hypocellular, excessively friable, and cytologically altered by drugs. We have shown discordancy of day 7 and 14 BM lymphoblast counts using morphologic and flow cytometric immunophenotypic analyses (FC). The aim of our study was to determine the utility, reliability, and cost effectiveness of lymphoblast enumeration using DNA content analysis by flow cytometry (DNA-FC) and to further demonstrate the subjectivity of morphologic review. All new cases of ALL had DNA-FC and FC analyses. The percentage of lymphoblasts as determined by both methods was compared for 82 aneuploid cases. Three pathologists independently reviewed aspirate smears from 39 bone marrow samples of aneuploid ALL that were obtained during early induction. These results were compared among themselves and with the results obtained by DNA-FC. We found excellent correlation between the percentage of lymphoblasts as determined by DNA-FC and FC (R ² = 0.97) over a range of 0 to 99%. Pathologic review agreed with the DNA-FC, on average, 68%. The sensitivity, specificity, and positive and negative predictive values of morphologic review, averaged 53, 84, 78, and 63%, respectively, when using DNA-FC as the “gold standard.” All three pathologists achieved agreement of lymphoblast percentage by morphology in 72%. In our laboratory, the use of DNA-FC equates to one-sixth the time and one-half the price of FC per exam. We have shown a strong correlation between blast counts determined by DNA-FC and FC. DNA-FC is an objective, economical, and reliable method to assess early response in induction marrows from aneuploid ALL where morphology is often uninterpretable. This test is highly reproducible and available at most pediatric institutions. Prospective studies need to be employed to evaluate the effect of more definitive methods (DNA-FC and FC) of assessing the early response in bone marrows on prognosis.