Age affects the treatment outcome in adults with acute myeloid leukemia (AML) but its effect in pediatric AML is less certain. We reviewed outcome in 423 patients, 21 years of age or younger, who were treated for newly diagnosed AML (excluding acute promyelocytic leukemia) at St. Jude Children’s Research Hospital (n=288) or M.D. Anderson Cancer Center (n=135) between 1983–2002. Patients were divided into those treated between 1983–1989 and 1990–2002. During these two time periods, distinct sets of protocols were used in both institutions, with higher intensity treatment used in the more recent era. After accounting for the effects of cytogenetics, Down syndrome (favorable), increased WBC (unfavorable), FAB type M7 (unfavorable) and early treatment era (unfavorable), every additional year of age conveyed a 4.4 % increase in the risk of death and a 3.3% increase in the risk of any adverse event (death, failure to achieve complete remission, or relapse), with p< 0.001 for the effect of age on survival or event-free survival (EFS). When patients were divided into those above or below the median age (10 years), there was a significant interaction between age and treatment era (p=.05), with patients below 10 having better EFS when treated in the recent era. For patients below 10, the relative risk of an event was 1.6 folds higher in the early treatment era than in the recent era ( p=.016). For patients 10 and older, the improvement in EFS according to treatment era was not significant. Treatment site did not influence the EFS or survival, and the effects of age were similar at both sites. These results suggest that age is an independent prognostic factor in childhood AML with younger children benefiting from more intensive treatment than their older counterparts.

Author notes

Corresponding author

Sign in via your Institution