Benchmarking treatment effects for patients over 70 with acute myeloid leukemia: A systematic review and meta-analysis

Objectives: The majority of patients with acute myeloid leukemia (AML) are aged 70 and over. However, there is uncertainty about how and whether older patients with AML should receive cytotoxic treatment.

Materials and methods: Medline and Cochrane library search was performed for studies in newly diagnosed AML which enrolled at least 20 patients per arm (for randomized controlled trials), or 50 patients (for non-randomized studies) over the age of 70. References were hand searched for additional eligible studies. Study investigators were contacted to maximize relevant data. Dual independent data extraction was done using standardized data collection forms. Data was collected on study and treatment characteristics, baseline patient information, and outcomes. Study methodological quality was assessed. The primary outcome was 1 year overall survival (OS). Impact of treatment [intensive chemotherapy (INT), low-dose chemotherapy (LOW), hypomethylating agents (HMA), or best supportive care (BSC)], cytogenetics, performance status, and comorbidity were assessed.

Results: The search produced 11,846 references of which 38 randomized controlled trials and 30 non-randomized studies met inclusion criteria, representing 13,381 patients, with a worldwide distribution. One-year OS with INT was 37% (31-42%), with LOW 11% (6-18%), with HMA 35% (18-54%) and with BSC 17%(13-21%). Two-year OS was 22% (18-26%), 11% (7-15%), 22% (16-28%), 6% (2-12%), respectively. We present subgroup data based on the studies including cytogenetics, performance status, and comorbidity. Formal direct comparisons with adjustment for all prognostic factors were not possible.

Conclusions: In this largest to date series of AML patients aged 70 and older, we provide benchmarks for treatment efficacy and effectiveness that may be used for decision analysis models and for the future development of clinical trials focusing on these patients.