Background: Acute myeloid leukemia (AML) is the most common form of acute leukemia affecting adults, with incidence increasing with patient age. Previous studies have found that older AML patients, constituting the majority of the AML population, generally have poor outcomes, high healthcare expenditures, and median survival of <3 months. Because up-to-date information on treatment patterns, survival trends, and costs of care for elderly AML patients are lacking in the literature, we examined Medicare fee-for-service enrollees with primary AML to update these estimates and report on changes in treatment for this population.
Objective: The primary objective of this study was to examine real-world data on treatment patterns, survival, and costs in elderly patients with primary AML. Factors associated with receipt of chemotherapy and with mortality also were assessed.
Methods: This is a retrospective database analysis using the Surveillance, Epidemiology, and End Results cancer registry and linked Medicare claims. Patients aged 65 years and older, who were newly diagnosed with AML between 1 January 1997 and 31 December 2007 were selected if they had no previous neoplasm or hematological disease. Patients were followed until death or to the end of the observation period (31 December 2007). Study measures included chemotherapy and supportive care (SC) received, survival time, and all-cause healthcare utilization and costs accrued from AML diagnosis until death or observation period end. Regression analyses assessed factors associated with receipt of chemotherapy (logistic) and mortality among chemotherapy and SC users (Cox).
Results: Of the 4,058 patients meeting the inclusion criteria, 43 % received chemotherapy; 57 % received SC only. Among patients receiving chemotherapy, 69.1 % died within 1 year; median survival was 7.0 months. Among patients receiving only SC, 95.0 % died within 1 year; median survival was 1.5 months. The most significant factors associated with receipt of chemotherapy were patient age [odds ratio (OR) = 0.420 among patients 75-84 years and 0.099 among patients 85+ years, compared with patients aged 65-74 years) and Charlson Comorbidity Index (CCI) score (OR = 0.614 for patients with a CCI = 2 or 3 and 0.707 for patients with a CCI >3, compared with patients with a CCI = 0) (all P < 0.001). The most significant factors associated with mortality among patients receiving chemotherapy were patient age [hazard ratio (HR) = 1.321 among patients 75-84 years and 1.832 among patients 85+ years, compared with patients aged 65-74 years] and CCI score (OR = 1.287 for patients with a CCI = 2 or 3 and 1.220 for patients with a CCI >3, compared with patients with a CCI = 0) (all P < 0.01). Mean (standard deviation) all-cause healthcare costs were $96,078 ($109,072); the largest component was inpatient utilization (76.3 %).
Conclusions: Younger patients with fewer comorbidities were more likely to receive chemotherapy and had longer survival. AML is associated with a substantial economic burden, and treatment outcomes appear to be suboptimal, with limited therapy options currently available.