Clinical impact of leukemic blast heterogeneity at diagnosis in cytogenetic intermediate-risk acute myeloid leukemia.

BACKGROUND: Individual cellular heterogeneity within the acute myeloid leukemia (AML) bone marrow samples can be observed by multi parametric flow cytometry analysis (MFC) indicating that immunophenotypic screening for leukemic blast subsets may have prognostic impact. MATERIAL AND METHODS: Samples from de novo AML patients of all cytogenetic risk groups were collected at diagnosis and subjected to MFC based on a four-color antibody panels against 33 CD membrane markers and retrospectively analyzed for the leukemia blast expression pattern and mean fluorescence intensity. Identification of the leukemic blast cells was based on right angle light scatter (SSC) and expression of CD45 and the cellular heterogeneity identified by the presence of at least two distinct subsets by any CD marker. RESULTS: Analysis of marrow samples from 86 patients with cytogenetic intermediate risk identified recurrent heterogeneous blast phenotypes for selected CD markers, three of which had prognostic impact with loss or gain of CD58, CD117, or CD14 expression. Multivariate Cox regression analysis of diagnostic variables identified poor prognostic factors: Age >55 years, presence of extramedullary disease, WHO performance score >2, a heterogeneous CD58, CD117, or CD14 expression on blast cells. Each variable added to a simple and clinical useful and MFC based prognostic score system associated to inferior survival in the intermediate risk group of AML patients. CONCLUSIONS: These observations support that leukemic blast heterogeneity detected by MFC has additional prognostic significance in de novo AML; however, the score system needs to be prospectively validated in future clinical trials before implementation.

Prediction of haemorrhage in the early stage of acute myeloid leukaemia by flow cytometric analysis of platelet function.

Haemorrhage is often responsible for the lethal course of acute myeloid leukaemia (AML). Previously, multiple platelet function defects were identified by flow cytometric analysis of platelet activation markers in AML. The role of flow cytometric analysis of platelet function in characterization of prognostic markers of haemorrhage in AML patients has not been well elucidated. The objective of this prospective study was to analyse platelet function in 50 AML patients at diagnosis and to compare results with clinical bleeding score, graded by common toxicity criteria. Platelet activation markers CD62P, CD42b, CD63 and PAC-1 were analysed following in vitro activation by thrombin receptor activating peptide. The following plasma haemostasis parameters were measured: soluble P-selectin, activated partial thromboplastin time, thrombin time, prothrombin time, D-dimer, fibrinogen, and von Willebrand factor antigen. In a multivariate analysis, P-selectin (CD62P) <36 molecules of equivalent soluble fluorochrome x 10(3) (P < 0.0015) and platelet count <40 x 10(9)/l (P = 0.01) were significant predictors of haemorrhage at diagnosis. Haemorrhage at diagnosis predicted grade 3-4 haemorrhage in the first 28 d following diagnosis (P = 0.018). The presented results indicate that low P-selectin is a prognostic marker of haemorrhage in AML.