Background: In adult acute myelogeneous leukaemia (AML) no firm conclusion is still achieved regarding the relationship between blast cell kinetics and prognosis. Aim: We performed the analysis of the nucleolar organiser regions (AgNORs) on bone marrow biopsies of patients with AML, to verify their association with patient and/or leukaemia cell characteristics, and prognosis. Materials and Methods: Fifty-seven previously untreated patients (27 males and 30 females, mean age: 41.9 years) with “de novo” AML were retrospectively studied. According to the French-American-British (FAB) classification, there were 11 M1, 9 M2, 19 M3, 16 M4 and 2 M5 leukaemia, which were grouped into three rather homogeneous groups: myeloblastic (M1-M2, 20 cases), myelomonocytic and monoblastic (M4-M5, 18 cases), and promyelocytic (M3, 19 cases) leukaemia. M1-M2, and M4-M5 patients were treated for remission induction and consolidation with the EORTC/GIMEMA protocols 8A /8B. M3 patients received idarubicin and all-trans retinoic acid (AIDA protocol). The mean follow-up for censored patients was 64 months. Jamshidi needle biopsies, taken before therapy, were fixed in formalin, decalcified and embedded in paraffin. Sections were stained with the standardised AgNOR method, and the mean AgNOR count in blast cells was calculated. Results: The mean AgNOR count for the whole series was 5.79 (median: 5.55, SD: 1.68, range: 2.91-10.39); it was lower in M3 (4.21) than in M1-M2 (6.38, p <0.0001) or M4-M5 leukaemia (6.80, p <0.0001). Five year survival rates were 84% for M3, 48% for M4-M5 and 20% for M1-M2 patients (p =0.0001). In the whole series, no association between AgNOR count and survival was found. However, when the phenotypically homogeneous groups were separately considered, and their respective AgNOR medians were used as a cut-off, the five year survival rates were 30% for M1-M2 patients with AgNOR count >6.25, but only 10% for those with lower count (p=0.0009). On the contrary, all the M3 patients with AgNOR count <4.28 were alive after 5 years, but only 66% of those with higher count survived (p=0.05). No difference in survival was found in M4-M5 patients between those having high (>6.43) or low AgNOR count (p=0.7) Conclusions: AgNOR analysis indicates that the proliferative activity of AML is rather heterogeneous, and appears to be associated to the FAB classification. In particular, it is lower in M3 than in all other categories. Furthermore, in M1-M2 leukaemia, the higher the AgNOR count, the longer the survival; on the contrary, in M3 leukaemia, the lower the AgNOR count, the longer the survival. This suggests that in M1-M2 leukaemia the rapidly proliferating blasts better respond to the induction and consolidation chemotherapy. It is likely that in M3 leukaemia the association of chemotherapy and differentiating agents, like ATRA, needs a slow proliferative activity to induce long lasting remission.
Nucleolar organiser regions in bone marrow biopsies of patients with acute myelogeneous leukaemia
PICH, Achille;
2000-01-01
Abstract
Background: In adult acute myelogeneous leukaemia (AML) no firm conclusion is still achieved regarding the relationship between blast cell kinetics and prognosis. Aim: We performed the analysis of the nucleolar organiser regions (AgNORs) on bone marrow biopsies of patients with AML, to verify their association with patient and/or leukaemia cell characteristics, and prognosis. Materials and Methods: Fifty-seven previously untreated patients (27 males and 30 females, mean age: 41.9 years) with “de novo” AML were retrospectively studied. According to the French-American-British (FAB) classification, there were 11 M1, 9 M2, 19 M3, 16 M4 and 2 M5 leukaemia, which were grouped into three rather homogeneous groups: myeloblastic (M1-M2, 20 cases), myelomonocytic and monoblastic (M4-M5, 18 cases), and promyelocytic (M3, 19 cases) leukaemia. M1-M2, and M4-M5 patients were treated for remission induction and consolidation with the EORTC/GIMEMA protocols 8A /8B. M3 patients received idarubicin and all-trans retinoic acid (AIDA protocol). The mean follow-up for censored patients was 64 months. Jamshidi needle biopsies, taken before therapy, were fixed in formalin, decalcified and embedded in paraffin. Sections were stained with the standardised AgNOR method, and the mean AgNOR count in blast cells was calculated. Results: The mean AgNOR count for the whole series was 5.79 (median: 5.55, SD: 1.68, range: 2.91-10.39); it was lower in M3 (4.21) than in M1-M2 (6.38, p <0.0001) or M4-M5 leukaemia (6.80, p <0.0001). Five year survival rates were 84% for M3, 48% for M4-M5 and 20% for M1-M2 patients (p =0.0001). In the whole series, no association between AgNOR count and survival was found. However, when the phenotypically homogeneous groups were separately considered, and their respective AgNOR medians were used as a cut-off, the five year survival rates were 30% for M1-M2 patients with AgNOR count >6.25, but only 10% for those with lower count (p=0.0009). On the contrary, all the M3 patients with AgNOR count <4.28 were alive after 5 years, but only 66% of those with higher count survived (p=0.05). No difference in survival was found in M4-M5 patients between those having high (>6.43) or low AgNOR count (p=0.7) Conclusions: AgNOR analysis indicates that the proliferative activity of AML is rather heterogeneous, and appears to be associated to the FAB classification. In particular, it is lower in M3 than in all other categories. Furthermore, in M1-M2 leukaemia, the higher the AgNOR count, the longer the survival; on the contrary, in M3 leukaemia, the lower the AgNOR count, the longer the survival. This suggests that in M1-M2 leukaemia the rapidly proliferating blasts better respond to the induction and consolidation chemotherapy. It is likely that in M3 leukaemia the association of chemotherapy and differentiating agents, like ATRA, needs a slow proliferative activity to induce long lasting remission.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
