Epidemiology and Survival Outcomes in Elderly Acute Myeloid Leukemia Patients: An NCDB Study

Background: Recent data suggests that more elderly AML patients are receiving treatment today compared to no treatment. With this in mind, there have been no recent studies looking at prognostic variables and survival outcomes in this patient population. We are conducting this retrospective study utilizing the National Cancer Database to help better prognosticate these patients and improve clinical outcomes.

Methods: A total of 43,270 patients with AML over the age of 60 were identified utilizing the National Cancer Database. Patients were identified by the histology codes 9840, 9861, 9867, 9871-4, 9896, 9897, and 9920. Survival tables and Kaplan-Meier curves were utilized to calculate both 1- and 3-year survival probabilities with log-rank analysis to compare the variables. Variables analyzed include cytogenetics, sex, facility type, race, insurance status, income and education quartiles, and treatment including adjuvant therapies to chemotherapy and type of bone marrow transplant.

Results: Approximately 55.3% of the cohort was male and Caucasian patients represented 88.1% of the cohort. The median age for the cohort was 73 years of age. The most common insurance was Medicare at 71.8% of the cohort followed by private insurance at 20.3%, Medicaid at 2.8%, and not insured at 1.7%. Approximately 34% of the cohort lived in zip codes with incomes of >$63,333, followed by 23.4% living in $50,354-63,332, 23.1% living in $40,227-50,353, and 17.9% living in <$40,227. The majority (43.5%) elected to receive treatment at academic and research programs followed by 37.5% at comprehensive community cancer programs, 11.9% at integrated network cancer programs and 7.1% at community cancer programs. Approximately 68% of the cohort elected to have chemotherapy with 1,897 patients electing to have an adjuvant bone marrow transplant and 229 patients elected to receive adjuvant immunotherapy.

The overall 1- and 3-year survival probabilities for the cohort was 29.0% and 11.3%, respectively. Asian patients had the best survival outcomes, followed by Caucasians and African Americans. The type of AML resulted in significant difference in survival outcomes with AML with abnormal marrow eosinophils having the best 1- and 3- year survival probabilities of 48.7% and 29.5%, respectively, while acute myelomonocytic leukemia had the worst 1-year survival at 27.2% and AML with minimal differentiation had the worst 3-year survival at 9.4%. Patient that received treatment at academic or research centers had the best 1- and 3-year survivals of 35.6% and 14.4%, respectively. Patients that received treatment at community cancer programs had the worst survival outcomes (1-year:22.3% and 3-year:8.3%). Patients with private insurance had the highest 1- and 3-year survival probabilities of 41.8% and 20.4% followed by Medicaid (1-year:36.7% and 3-year:16.5%), not insured (1-year:33.9% and 3-year:17.4%), and Medicare (1-year:24.8% and 3-year:8.3%). As income and education levels increased, the probability of survival also increased. Patients that received an adjuvant bone marrow transplant to chemotherapy had the highest survival probability (1-year:76.3% and 3-year:46.7%) followed by adjuvant immunotherapy (1-year:53.8% and 3-year:18.8%), and chemotherapy alone (1-year:38.5% and 3-year:15.0%). Patients that didn't receive chemotherapy had 1- and 3-year survival probabilities of 9.6% and 3.5%, respectively.

Conclusion: To date, this is the largest study on the prognostic factors in patients over 60 years of age with AML. Major prognostic factors include type of AML, facility type, race, insurance status, median income quartile, educational level, and adjuvant therapy.