Background:Various demographic, clinical and cytogenetic factors have been shown to affect the outcome of patients with acute myeloid leukemia (AML). Some of these risk factors are well-described with performance status and patient age being principal predictors of early death while cytogenetic and molecular factors allow stratification into prognostic categories. However, several critical factors have not been adequately assessed. Typically AML patients treated with standard induction chemotherapy will have cytopenias requiring intervention. Often the neutropenia is prolonged and results in neutropenic fevers and a high risk of infections. We hereby describe a single institution retrospective analysis at the University of Oklahoma Health Sciences Center (OUHSC) in which we evaluate the effect of duration of days with at risk absolute neutrophil (ANC) and absolute lymphocyte counts (ALC) and the occurrence of neutropenic fever on survival and response rates of AML patients.Methods:This is a retrospective chart review of AML adult patients who were diagnosed and treated at OUHSC between 2000 and 2014 and were undergoing initial induction chemotherapy. Outcomes of interest were overall survival (OS), event free survival (EFS: in case of relapse or death) and complete response (CR). Variables of interest wereneutropenic fever, and duration of decreased ANC and ALC counts (ANC<500, ANC<100, ALC<500, ALC<100). Institution policy did not allow granulocytes colony stimulating factors use because of the interference with bone marrow aspirate results. Nonetheless it was rarely used in few cases of septic shock, so this variable was not included in this analysis. Descriptive and bivariate analyses were conducted to evaluate the variables of interest and the outcomes.Models were used to assess the relationship between the variables of neutropenic fever, duration of decreased counts (ANC<500, ANC<100, ALC<500, ALC<100) and the outcome of interest after adjusting for age, race, gender, risk group andwhite blood count (WBC). Multivariable Cox proportional hazard models were used for OS and EFS and multivariable logistic regression models were used for CR. SAS 9.4 was used for all analyses. An alpha of 0.05 was used.Results:A total of 153 patients were analyzed. Mean age was 50 years, 35.7 % female, 64.3% male. Based on cytogenetics 16.9% were in favorable category, 28.6% in intermediate, 25.3% in unfavorable risk group, and 29.2% unknown. Mean number of weeks with ANC <500 was 3.58 (25 days), ANC <100 was 2.81 (20 days), ALC < 500 was 4.47 (31 days), ANC <100 was 3.24 (23 days). Incidence of neutropenic fever was 86.9%. Total number of positive cultures was 55.2%, of which 78.3% were bacterial, 25.3% fungal and 7.7% viral.After adjustment for age, race, gender, WBC and risk group, both neutropenic fever and duration during which ANC < 100 were associated with worse OS (p value <0.05) while duration of ANC< 500 was not significant (p value < 0.0576). When we considered EFS, the duration where ANC<500, ANC <100, and neutropenic fever occurrence were significant (p value <0.05).The hazard ratio of death is 3.15 for those with neutropenic fever compared to those without.With regard to CR, duration of ANC < 100 was significant (p value 0.0272). The duration of ALC <500 or <100 was not significant for OS nor EFS. Examining duration of neutropenia, for every week with ANC <500, there was a 9% higher hazard ratio of death after adjusting for other covariates.Conclusion:In this single-institution, retrospective study, we identified that the duration of neutropenia along with the presence of neutropenic fever during induction therapy adversely affected OS and EFS. Shorter duration of neutropenia significantly correlated with CR. Duration of days with ALC count <500 was not predictive of OS nor EFS. Larger studies are needed to examine the prognostic significance of neutropenia duration and its relationship to OS and CR. Neutropenia duration during induction chemotherapy may be an important risk factor in making decisions regarding future treatments including allogeneic transplant tolerability and aggressiveness of consolidation chemotherapy.It is important to look at this variable in more intensive induction regimens (using higher dose of cytarabine or adding a nucleoside analogue) as prolonged neutropenia may be a marker of poor general health rather than the induction regimen. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.