Conflicting results exist regarding the prognostic importance of CD7 expression in acute myelogenous leukemia (AML). Differences in the method of determining CD7 positivity, the antibody used, the therapy administered, and the CD7 level used as a cutoff point to reduce it to a binary variable have all been postulated to account for the discordant findings. We determined the level of CD7 expression by flow cytometric analysis using the Leu9 monoclonal antibody in 331 patients with newly diagnosed AML and attempted to determine the impact of CD7 on AML prognosis. This study used the same methodology and antibody as three of the four studies that reported a positive association between CD7 expression and prognosis in AML. Optimal cutpoint analysis was used to divide the population into CD7-positive (CD7+) (>10.5% expression) and CD7-negative (CD7-) (< 10.5% expression) groups with the largest survival difference. At the optimal cutpoint, the difference in survival was not statistically significantly different (P = 0.068 uncorrected, P = 0.244 corrected for optimal cutpoint search). There was a marked imbalance in the distribution of favorable cytogenetic abnormalities [t(8;21), inversion 16, t(15;17)], with 95% segregating to the CD7- group. Analysis excluding patients with favorable cytogenetic abnormalities revealed no prognostic importance for CD7 expression (P = 0.24 uncorrected). The response rate (CR) and survival experiences of CD7+ and CD7- patients were similar with six different regimens. CD7 expression was not a significant independent prognostic factor in a Cox regression model that included cytogenetics as a predictive variable, but it was marginally significant when cytogenetics was excluded. We conclude that regardless of the antibody used, the therapy received, or the cutoff point selected to determine CD7 expression, CD7 is not associated with response rate, prognosis, or survival in AML. The 'optimal cutoff point analysis' utilized in this study has applicability to other biologic parameters as well.