Cup-Like AML - New Disease or Morphological Phenomenon?.

Abstract Recently Kussick et al. (Leukemia2004;18:1591–98) reported cases of AML containing a distinctive, cup-like nuclear indentation in 10% or more of the blasts, which was associated with a loss of HLA-DR expression and FLT3 mutation. To investigate this feature in a larger cohort, we re-evaluated peripheral blood (PB) and bone marrow (BM) smears from 266 patients (male, n=148; female n=118; median age 57 years) from the German AML96 trial for blast cell morphology. Two-hundred blasts in PB and BM each were differentiated to determine the percentage of cells with cup-like nuclear invagination by two consecutive investigators. The slides were seen by two further investigators, who assigned the AML to a cup-like positive or cup-like negative cohort. Both cohorts were analysed for FAB-type, number of PB and BM blasts, CD34 expression, karyotype abnormalities, FLT3 and NPM mutations, as well as survival data. In addition, electron microscopy was performed in cases with high numbers of cup-like blasts. The median number of blasts with cup-like morphology was higher in PB compared to BM (2% vs. 0.5%). With a cut-off value of >=5% of blasts with nuclear invagination in PB and/or BM, 91% (n=48/53) and 94% (n=15/16) of cases were assigned to the cup-like positive cohort, respectively. Therefore we used this cut-off value to define cup-like-positive AML in the further analyses. Age and gender were not different between cup-like-positive (n=55) and -negative cases (n=211). FAB types M1 and M2 represented 62% of patients with cup-like AML, M5A was diagnosed in 20%, and in 11% and 4% the morphology was AML M4 and M4Eo. The median number of CD34+ blasts was 7% in cup-like positive AML compared to 27% in cases without this morphology (p=0.001). FLT3 and NPM aberrations were significantly more frequent in the cup-like positive cohort (FLT3, p=0.001; NPM, p<0.0001; FLT3+NPM, p<0.001) compared to the cup-like negative patients. In line with this, cup-like AML was significantly associated with normal karyotype (p=0.004). The levels of significance were even more pronounced if monocytic leukemias (M4, M4Eo, M5A and M5B) were excluded from the analyses (due to the characteristic cleaved or folded nuclear morphology in these FAB types). Cup-like morphology did not influence response rates (CR after second cycle of induction chemotherapy) and the survival parameters (DFS, OS), neither for the total cohort, nor for subgroups (non-monocytic leukemias, patients <60 years, patients with FLT3 or NPM mutation). Furthermore, multivariate analysis failed to demonstrate cup-like nuclear morphology as an independent risk factor. Electron microscopy revealed that the phenomenon is caused by an accumulation of cytoplasmatic components rather than by primary nuclear changes. The study confirms the hypothesis that cup-like nuclear morphology in AML blasts cells represents a specific morphological phenomenon. Because of the strong association to aberrations in FLT3 and/or NPM genes the feature should be used as a surrogate marker, making a subsequent molecular analysis meaningful. The biological background needs further investigation.