Outpatient Vs Inpatient Hidac Consolidation for AML: The MSKCC Experience

Background: Standard management for non-adverse risk acute myeloid leukemia (AML) in fit patients (pts) is intensive induction chemotherapy followed by post-remission high-dose cytarabine (HiDAC). Historically, HiDAC has been administered in the inpatient (IP) setting due to associated prolonged cytopenias, infection risk, and transfusion and supportive care needs. However, improvements in antimicrobial prophylaxis as well as nosocomial infection risk and increasing financial burden associated with prolonged IP chemotherapy hospitalization have highlighted the potential role for outpatient (OP) chemotherapy administration. To date, there is limited data supporting the use of IP vs OP HiDAC administration. Here, we compared outcomes in pts receiving IP vs OP HiDAC consolidation at a single large academic cancer center.

Methods: We performed a retrospective medical record review of adult pts with AML who received HiDAC consolidation at Memorial Sloan Kettering Cancer Center (MSKCC) between 1/1/2014 and 6/1/2023. Toxicity-related hospitalization was defined as an independent hospitalization for any adverse event associated with HiDAC administration. Kaplan-Meier method was used to estimate overall survival (OS) and relapse-free survival (RFS). Wald test was used to compare OS and RFS between groups, adjusting for age and European LeukemiaNet (ELN) risk. Poisson regression was used to assess toxicity-related hospitalization rates.

Results: 198 pts received HiDAC consolidation at MSKCC during the specified interval. 30% (59/198) received IP HiDAC and 70% (139/198) received OP HiDAC. Median age of the overall cohort was 59 years (yrs) (interquartile range (IQR) 46, 67), with IP recipients being significantly older [64 yrs (IQR 55, 70) vs 57 yrs (IQR 44, 66), p=0.002]. Rates of de novo vs secondary vs therapy-related AML were comparable across the two groups. Eastern Cooperative Oncology Group (ECOG) scores were significantly higher in the IP (0=41%, 1=56%, 2=2.9%) than the OP group (0=62%, 1=38%), p=0.034, while Karnofsky Performance Scale scores were comparable. There was no difference in disease risk categories by ELN between groups. Patients received between 1-4 cycles of HiDAC consolidation, with course determined by their primary oncologist; no difference was observed between groups in number of consolidation cycles received.

Median follow-up for the entire cohort was 45 months (mos) (95% CI 36-50), with 2-year OS and RFS of 79% (95% CI 73-85%) and 67% (95% CI 61-74%), respectively. There was no difference in OS between the IP vs OP groups [2-year OS 76% (95% CI 65-88) vs 80% (95% CI 73-87), p=0.7], adjusting for age and ELN risk. There was also no difference in RFS following IP vs OP HiDAC [2-year RFS 69% (95% CI 57-82%) vs 67% (95% CI 59-76%), p=0.8]. Univariable analysis demonstrated that toxicity-related hospitalization rates were significantly higher in the IP than OP group [incidence rate ratio (IRR) 1.86 (95% CI 1.39-2.48), p<0.001]. Multivariable regression adjusting for AML type, age, gender, ELN, and ECOG revealed that toxicity-related hospitalization rates were comparable in the IP vs OP groups, with a trend towards higher rates in the IP group [IRR 1.45 (95% CI 0.92-2.28), p=0.11]. Hospitalizations for febrile neutropenia were similar across IP and OP groups by both univariable [IRR 1.07 (95% CI 0.70-1.59), p=0.8] and multivariable regression analysis [IRR 0.80 (95% CI 0.42-1.47), p=0.5]. For those hospitalized in association with HiDAC for reasons other than chemotherapy administration, there was a trend toward longer median length of stay (LOS) in the IP cohort [6 days (5, 11) vs 5 days (3, 7)].

Conclusions: In this single center cohort, there was no difference in 2-year OS or RFS for patients receiving IP vs OP HiDAC consolidation. There was also no difference in rates of hospitalization for neutropenic fever in the IP vs OP groups, though there was a trend towards higher rates of toxicity-related hospitalization and longer non-chemotherapy admission LOS for the IP group. Our data demonstrate that OP HiDAC does not compromise safety or efficacy and can reduce hospitalizations for chemotherapy administration. OP HiDAC has the potential to reduce nosocomial infections and healthcare costs and improve quality of life. Further investigation is warranted to better characterize the impact of OP HiDAC and establish the safety and efficacy of this approach across multiple centers.