Background: Immunosuppressive therapy has been used for successful treatmen
t of severe aplastic anemia, but little information is available on outcome
in older patients.
Objective: To evaluate outcome in patients older than 50 years of age who r
eceived immunosuppressive therapy for aplastic anemia.
Design: Retrospective cohort study.
Setting: 56 centers of the European Group for Blood and Marrow Transplantat
ion (EBMT).
Patients: 810 patients with aplastic anemia reported between 1974 and 1997.
Patients were evaluated according to age group: 60 years of age or older (
n = 127), 50 to 59 years of age (n = 115), and 20 to 49 years of age (n = 5
68; reference group).
Intervention: Antilymphocyte globulin, cyclosporine, or both.
Measurements: Survival, cause of death, response to treatment, relapse rate
, and risk for late complications were analyzed in all patients and by age
group.
Results: The 5-year survival rate was 57% (95% CI, 46% to 66%) in patients
50 to 59 years of age and 50% (CI, 39% to 60%) in patients 60 years of age
or older compared with 72% (CI, 68% to 76%) in patients younger than 50 yea
rs of age (P < 0.001). Response to therapy, relapse rate, and risk for clon
al complications were similar in all three age groups (P > 0.2). Age was si
gnificantly associated with an increased risk for death (relative risk comp
ared with patients 20 to 49 years of age, 1.80 [CI, 1.29 to 2.52] for patie
nts 50 to 59 yea rs of age and 2.57 [CI, 1.87 to 3.53] for patients greater
than or equal to 60 years of age), mainly because of bleeding or infection
(P = 0.02). Response to immunosuppressive therapy in all patients at 12 mo
nths was 62% (CI, 58% to 66%); no difference was seen among the age groups
in multivariate analysis (P > 0.2). Sixty-six of the 379 responding patient
s (17%) subsequently had relapse. The risk for clonal disorders at 10 years
was 20% (CI, 15% to 27%).
Conclusions: Response to immunosuppression in aplastic anemia is independen
t of age, but treatment is associated with increased mortality in older pat
ients.