Background and objectives: The prognosis of elderly patients with acute myelogenous leukemia (AML) is usually dismal, while the true survival of older patients not included in clinical trials is not known. We retrospectively evaluated the impact on survival of an aggressive versus a non-aggressive approach in 1005 patients aged >60 years registered in the database of the GIMEMA cooperative group.

Design and methods: Group A patients (n=621) received aggressive treatment, while group B patients (n=384) underwent non-aggressive therapy. The groups were different for risk factor distribution: the patients in group B had a higher median age, worse performance status (PS) and a higher proportion of previous myelodysplastic disease.

Results: The overall median survival was 7 and 5 months in groups A and B, respectively (p min of 0.0001). At multivariate analysis the following factors were associated with a significantly shorter survival: age >71 years (RR=1.27; 95% CI=1.07-1.50), PS=2-4 (RR=1.44; 95% CI=1.24-1.68), white cell count > 10,000 mL (RR=1.37; 95% CI=1.06-1.75), and heart dysfunction requiring treatment (RR=1.26; 95% CI=1.05-1.50). No difference in survival was associated with aggressive or non-aggressive treatment (RR=1.1; 95% CI=0.94-1.32). Patients aged min of 70 years, with no heart disease, but a white cell count > 10,000/mL showed a significantly better survival when treated aggressively (median survival 7 vs 3 months, p = 0.011).

Interpretation and conclusions: Despite an obvious selection of patients with a worse prognosis in group B, the difference in survival between the two groups was marginal. Multivariate analysis failed to demonstrate a significant survival benefit in aggressively treated patients. All these considerations indicate that elderly patients with AML are overall unlikely to benefit from aggressive treatment, so that this should be offered only to selected patients.