Diagnostic and follow-up eosinophilia is not predictive of outcome in childhood acute myeloid leukemia.

BACKGROUND

Peripheral blood eosinophilia (PBE) and bone marrow eosinophilia (BME) are associated with various tumors. Eosinophilia is frequently paraclonal (because of cytokines and not part of the malignant clone), but in some cases of acute myeloid leukemia (AML), it may also be clonal.

METHODS

We retrospectively evaluated 80 pediatric AML patients treated from June 2004 to December 2008 at our cancer center for PBE and BME at diagnosis and postinduction.

RESULTS

At diagnosis, PBE was present in 18.8% patients whereas at postinduction it was observed in 7.7% patients. No patient had abnormal eosinophils. At diagnosis, PBE correlated with the absence of gum hypertrophy whereas BME correlated with the French-American-British-M2 subtype and absence of gum hypertrophy. PBE/BME did not have a significant correlation with event-free survival. After completion of therapy, on follow-up up to 2 years every 3 months, one-third of the patients had eosinophilia; however, it did not correlate with disease-free survival.

CONCLUSIONS

This is the first study to evaluate the impact of eosinophilia at presentation, postinduction, and on follow-up with outcome in pediatric AML. Although eosinophilia was observed in one-fifth of the patients at diagnosis and in one-third of the patients on follow-up, it seems to be reactive in nature with no impact on outcome and thus it should not be a cause of alarm.