Medicare Fee-for-Service Enrollees with Acute Myelogenous Leukemia: An Analysis of Treatment Patterns and Patient Survival

Acute myelogenous leukemia (AML) is the most common type of leukemia among adults in the US. The incidence of AML increases with age. Older AML patients, constituting the majority of the AML population, generally have poor outcomes with median survival less than 3 months. Published information on treatment patterns and survival trends in elderly patients with AML is outdated and newer information on treatment patterns and survival is lacking.

The goal of this retrospective database analysis is to examine treatment patterns, overall patient survival, predictors of which patients are likely to receive chemotherapy, and predictors of mortality among Medicare fee-for-service enrollees diagnosed with AML in the most recent available database.

Medicare patients aged 65+ years in the SEER (Survey, Epidemiology, and End Results) cancer registry with a new AML diagnosis between 1/1/1997 and 12/31/2007 were selected for study inclusion. Patients were required to have at least 6 months of pre-AML Medicare Part A and B benefits and no evidence of managed care (Medicare Part C) enrollment post-AML diagnosis. Patients were excluded from the analysis if they had evidence of another tumor (either solid or hematological) in the SEER registry before the first AML diagnosis. Health care claims in the 6 months pre-index were examined, and patients with any diagnosis of a solid tumor (not specified in SEER) were also excluded. Patients were followed until their date of death or end of observation period (i.e., 12/31/2007). Study measures included AML-directed treatments (i.e., chemotherapy, radiation therapy, hematopoietic stem-cell/bone marrow transplants [HSCT/BMT]), best supportive care received, and post-AML diagnosis survival time. Patient survival time was assessed overall and for patients receiving chemotherapy during follow-up versus patients receiving best supportive care only. Temporal changes in treatment utilization and survival were assessed by evaluating these measures separately for AML cases diagnosed in 1997–1999, 2000–2003, and 2004–2007. Multivariate logistic regressions were undertaken to assess predictors of receipt of chemotherapy, including patient demographics, comorbidities, and year of AML diagnosis.

6,888 patients met the study inclusion criteria. Mean (SD) and median age were 78.3 (7.2) and 78.0 years respectively. Over 43% of patients received chemotherapy at any point post-diagnosis. Chemotherapy use increased slightly over time: 40.7%, 42.3%, and 46.0% of patients diagnosed with AML in the periods in 1997–1999, 2000–2003, and 2004–2007, respectively. Fifty-six percent of patients received only best supportive care post-diagnosis, and the percentage slightly decreased over time. Among patients receiving only best supportive care, rates of hospice care increased substantially over time: 32.9%, 42.7%, and 49.1% in each of the respective time periods. Rates of HSCT/BMT procedures were low with an increase over time: 0.67%, 2.06%, and 2.49%. Nearly all patients (97.1%) died during the observable follow-up, and median survival time was 2.6 months. Among patients who received chemotherapy, 93.9% died during follow-up and the median survival was 6.5 months with 5.7, 6.4, and 7.0 months among patients diagnosed in 1997–1999, 2000–2003, and 2004–2007, respectively. Among patients who received only best-supportive care, 99.5% died during follow-up and median survival was 1.5 months with little change over time. Younger patients (65–74 years vs >= 75 years), patients with fewer comorbidities (Charlson Comorbidity Index [CCI] ≤1 vs > CCI >1), patients with a post-AML secondary cancer diagnosis, and patients diagnosed with AML in more recent years were found to be more likely to receive chemotherapy.

Findings from our analysis showed an increasing trend in rates of chemotherapy treatment and utilization of hospice care among Medicare patients with AML. However, a large portion of elderly patients remain untreated. Median survival among patients who received chemotherapy was found to increase over time. Patients who received chemotherapy, when compared to those who did not, had a lower mortality rate and an over 3-fold longer median survival.

Yu:Boehringer Ingelheim Pharmaceuticals, Inc.: Employment.