Background: Venetoclax (ven) in combination with azacitidine (aza) has become the standard of care for the first line treatment of acute myeloid leukemia (AML) for unfit patients (pts). In the clinical trial setting, inpatient initiation of treatment was mandatory by protocol. Inpatient treatment can strain medical resources and impair the quality of life, but may be unnecessary for a subset of pts.
Aims: To analyze characteristics and outcomes of AML pts treated with ven-aza and compare between pts selected to initiate ven-aza treatment as inpatients (inpts) or as outpatients (outpts) in a medical center with inpatient and outpatient facilities
Methods: All pts diagnosed with AML and treated with first line ven-aza at Tel Aviv Sourasky Medical Center between Jan 2019 and Dec 2022 were included in this analysis. We compared categorical and continuous variables as appropriate, and examined survival outcomes by a Kaplan Meier and Log-Rank test. IBM SPSS version 29 was applied for the statistical analyses.
Results: 100 newly diagnosed AML pts treated with ven-aza were included. Treatment was initiated as inpts for 53 pts and as outpts for 47 pts. Pt characteristics are described in figure 1. Overall, the median age was 77 (range 40-95), 53% had secondary AML, and 55% had ELN adverse risk AML. Reasons for initiating treatment as inpts were frailty (32%), infection at presentation (24.5%), social issues (13.2%), cytopenias (15%), oncologic emergencies (7.5%), or other (7.5%).
Inpts were more frequently unmarried, and had a significantly worse performance status (PS), higher peripheral and bone marrow blast counts, and a greater incidence of fever at presentation than outpts. Outpts had a greater frequency of prior hypomethylating agent treatment. A trend towards higher WBC counts and LDH levels at presentation was seen in inpts. Inpts were treated earlier than outpts (median 4 vs. 8 days p<0.001)
Of 43 outpts, 27 (57.4%) were subsequently hospitalized (out2inpts) within a median time of 6 days from initiation of treatment (range 1-57 days). The main reason for hospitalization was fever (96%). The median length of hospitalization for out2inpts was shorter compared to inpts (12 vs 22.5 days, p=0.003).
Twenty pts completed induction treatment without hospitalization (alloutpts). No significant differences in baseline characteristics between out2inpts and alloutpts were found, most likely due to small numbers. Blast counts of out2inpts were numerically higher, and their Hb was lower and PS worse than of alloutpts
30 day mortality was higher in inpts vs outpts (17% vs 2% p=0.015) as was the incidence of clinically significant TLS (15% vs. 2% p=0.024). There were no differences in rates of febrile neutropenia, neutropenia or thrombocytopenia.
Overall, 66% of patients achieved CR/CRi. The median length of follow up was 6.4 months. Median overall survival (OS) and relapse free survival (RFS) were 7.9 months (CI 3.6-12.1) and 23.4 months (CI 6.8-39.9), respectively. Causes of death were progressive disease (57%) infections (23%) and other causes (20%). No differences in CR/CRi rate, OS (figure 1) or RFS, were detected between inpts and outpts. Alloutpts had a significantly longer OS than inpts or out2inpts (med OS 30.36 vs 6.3 vs 3.3 months respectively, p=0.014).
Summary/Conclusion: Initiation of ven-aza treatment as outpatients is feasible and safe for a subset of AML pts with good social support and performance status, and a lower burden of disease and enables less days in hospitalization without compromising treatment outcomes.
Keywords: Real world data, Acute myeloid leukemia, Venetoclax