MN1 Expression Predicts Prognosis of Acute Myeloid Leukemia with Normal Cytogenetics.

Cytogenetic aberrations are important prognostic factors in acute myeloid leukemia (AML). However, approximately half of adult AML patients lack cytogenetic abnormalities and identification of predictive molecular markers might improve therapy. Fusion of meningioma-1 (MN1) to TEL (ETV6) has been found in AML and MDS with t(12;22)(p13;q11). However, expression levels of MN1 have not been reported previously in AML. We evaluated MN1 expression as a prognostic marker in 142 AML patients aged 18–60 years with normal cytogenetics, who were uniformely treated according to the AML-SHG 1/99 trial. Patients received intensive, cytarabine-based induction and consolidation treatment including allogeneic progenitor cell transplantation if an HLA-compatible sibling was available, or in case of relapse. Specimens were obtained at diagnosis, and routine cytogenetic, FLT3-mutation, and MLL-PTD analyses were performed. MN1 expression was quantified by real-time RT-PCR on a LightCycler using QuantiTect SYBR Green. AML samples were dichotomized at the median value resulting in two groups: a low MN1 group and a high MN1 group. Baseline characteristics and outcome parameters were compared between these two groups. In addition, CD34⁺ cells were immunomagnetically enriched from mobilized blood of a healthy donor using MACS CD34 isolation kit. Cells were cultured in IMDM medium with various cytokines including either G-CSF, M-CSF or EPO. At various time points, cells were harvested and analyzed for MN1 expression. There were no significant differences between low MN1 and high MN1 expressing patients with respect to age, gender, ECOG performance status, diagnosis of de novo or secondary AML, FAB morphology, white blood cell count, percentage of blasts in blood or bone marrow, FLT3 mutations, or MLL-PTD. Low MN1 expressing patients significantly more often achieved a good response to the first course of induction treatment defined as blasts in bone marrow below 5%, no blasts in peripheral blood, and no extramedullary manifestation at day 15 compared to high MN1 expressing patients (87.3% vs. 71.8%, p=.02). There was no significant difference for remission status between the two groups. High MN1 expression predicted significantly shorter event-free survival (19% vs. 45.8% at 3 years, log-rank p=.0009), shorter relapse-free survival (23% vs. 52.8% at 3 years, log-rank p=.001), and shorter overall survival (38.2% vs. 58.8% at 3-years, log-rank p=.03). The high MN1 group relapsed significantly more often compared to the low MN1 group (56.7% vs. 35%, p=.02), and thus received an allogeneic transplant significantly more often (50.7% vs. 33.8%, p=.04). In multivariate analysis including known risk factors only MN1 expression, age (above the median compared to below the median age), and ECOG performance status (0 or 1 compared to 2) remained significant (hazard ratio: 2 (p=.01), 2.1 (p=.005) and 2.8 (p=.005), respectively). MN1 expression in CD34⁺ cells was 37-fold higher compared to the CD34⁻ cell fraction. However, by in vitro differentiation of CD34⁺ cells using various cytokines including either G-CSF, M-CSF or EPO, MN1 expression dropped to levels found in the CD34⁻ fraction within 7 days of culture. In conclusion, high MN1 expression predicts adverse prognosis and may define an important risk factor in AML with normal cytogenetics. Its upregulation in hematopoietic progenitor cells hints at a functional role of MN1 in blocking differentiation.