Prognostic Value of Immunophenotyping in Elderly Patients with Acute Myeloid Leukemia: A Single Institution Experience.

The poor prognosis of elderly patients with acute myeloid leukemia (AML) raises the question about the benefit of intensive chemotherapy in these patients. The impact of initial characteristics on prognosis has previously been addressed in elderly paitents, but very few data are available regarding the prognostic value of immunophenotypic characteristics in this setting. We investigated the expression of the membrane antigens CD13, CD15, CD33, and CD34 by flow cytometry in elderly patients with newly diagnosed AML, and analyzed whether these parameters possessed clinical or prognostic relevance, in order to help physicians in their choice of therapy. Immunophenotyping was performed in 273 patients aged 60 years or more (median 69 years). CD13 was expressed in 73%, CD15 in 43%, CD33 in 64%, and CD34 in 66% of cases. Complete remission was obtained in 157 cases (58%). The median overall survival was 8.1 months with a 3-year survival rate of 14%. Three risk groups were defined based on CD34 and CD33 antigen expression: Poor risk in patients with CD34⁺ CD33⁺ or CD34⁻ CD33⁻ disease, intermediate risk in patients with CD34⁺ CD33⁻ disease, and favorable risk in patients CD34⁻ CD33⁺ disease. Immunophenotype was, after cytogenetics, the most significant prognostic factor in terms of survival in a multivariate analysis (p = 0.03 and p < 0.0001 respectively). When combining immunophenotypic and cytogenetic parameters, patients were classified into four prognostic groups: Group A (3-year survival: 33%) including favorable and normal karyotypes with a favorable immunophenotype; Group B (3-year survival: 28%) including normal karyotypes with an intermediate immunophenotype; Group C (3-year survival: 8%) including intermediate or normal karyotypes with an unfavorable immunophenotype; and Group D (3-year survival: 2%) including all unfavorable cytogenetics. Immunophenotypic characteristics appeared to be a major prognostic factor in this patient population. Using two simple parameters assessed at time of diagnosis, we devised a prognostic system of immediate clinical utility for prognostic stratification and risk-adapted therapeutic choices.