Human leukocyte antigens (HLA) class I molecules restrict the interaction between cytotoxic T cells and target cells. Abnormalities in HLA class I antigen expression and/or function may provide tumor cells with a mechanism for escaping immune surveillance and resisting T cell-based immunotherapies. The potential for applying T cell-based immunotherapy in the treatment of acute myeloid leukemia (AML) has stimulated interest in analyzing HLA class I antigen expression on leukemic blasts in this disease. Little information is available in the literature. We have analyzed HLA class I antigen expression on bone marrow samples from 25 newly diagnosed AML patients by indirect immunofluorescence staining with monoclonal antibodies. Five of these patients were also studied at relapse. Leukemic blasts were resolved from normal lymphocytes by staining with anti-CD45 antibody; CD45 expression is dim on leukemia cells, but bright on lymphocytes. HLA class I antigen expression was higher on leukemic blasts than on autologous lymphocytes in all but one case. Moreover, there was no significant change in HLA class I antigen expression at relapse. These results suggest that abnormalities in HLA class I antigens are infrequent in AML and should not represent a major obstacle to the application of T cell-based immunotherapies in this disease. Leukemia (2001) 15, 128-133.

Untreated de novo (n=421) and secondary (n=189) acute myeloid leukemia (AML) patients≥60 years received intensified chemotherapy, including daunorubicin 60mg/m2 and etoposide 100mg/m2 during days 1, 2, 3 with cytarabine 100mg/m2 during days 1–7, with a second induction if needed and one consolidation course with these drugs and doses for 2, 2 and 5 days, respectively. In all, 287 (47%) achieved complete remission (CR), 136 (22%) died and 187 (31%) were non-responders. CR rates were 27, 44 and 52% for complex karyo-types, rare aberrations and neither (P<0.001), 52 and 37% for de novo and secondary AML (P=0.003), and 53 and 42% for age 60–69 and ≥70 years (P=0.015). In multivariable analysis, CR predictors included non-complex/non-rare karyotypes (P<0.001), de novo AML (P<0.001), better performance status (PS) (P<0.001) and younger age (P=0.001). Disease-free (DFS) and overall (OS) survival medians were 6.8 (95% CI: 6.2, 7.8) and 7.2 (95% CI: 6.4, 8.6) months. In multivariable analysis, DFS was shorter for complex karyotypes (P<0.001) and increasing white blood count (WBC) (P<0.001) and age (P=0.038), and OS for complex karyotypes (P<0.001), increasing WBC (P=0.001) and age (P<0.001), poorer PS (P<0.001) and secondary AML (P=0.010). Outcomes and prognostic factors were similar to those in previous Cancer and Leukemia Group B studies.

As Dr Ribatti indicates, our recent brief report in Blood 1 described a patient with a myelodysplastic syndrome (MDS) who developed acute monoblastic leukemia (acute myeloid leukemia (AML), French-American-British type M5a) shortly after initiating erythropoietin (EPO) therapy, and in whom AML regressed rapidly following discontinuation of EPO. The patient's AML cells expressed the EPO receptor and proliferated in vitro in response to EPO. We concluded that leukemic transformation was EPO-dependent in this patient, and that the mechanism of transformation was proliferation of myeloid blasts in response to EPO. Dr Ribatti proposes a second mechanism by which EPO therapy in MDS may promote progression to AML. He has previously reported that EPO promotes angiogenesis in vitro and in vivo by binding to EPO receptors present on vascular endothelial cells and stimulating their proliferation. 2 There is ample evidence that angiogenesis is important in MDS and AML. Bone marrow microvessel density is increased in MDS and AML and correlates with the percentage of myeloblasts, suggesting a correlation between angiogenesis and progression to leukemia. 3 Additional evidence for the importance of angiogenesis in MDS is the documentation of therapeutic responses to the anti-angiogenic agent thalidomide. 4 Thus, EPO therapy in MDS has the potential to promote angiogenesis in the bone marrow, which could in turn stimulate blast proliferation and evolution to acute leukemia. It should certainly be emphasized that EPO therapy is generally safe in MDS and is beneficial in approximately 25% of MDS patients. 5,6 Leukemic transformation is rare. However, our data and those of Dr Ribatti establish two mechanisms by which EPO may promote disease progression in MDS patients. We conclude again that MDS patients treated with EPO need to be closely observed for disease progression, and that management of leukemic transformation in MDS patients treated with EPO should consist of discontinuation of EPO and observation for disease regression, if clinically feasible.