Hydroxyurea in the Early Management of Acute Myeloid Leukemia and Hyperleukocytosis.

BACKGROUND: Patients with acute myeloid leukemia (AML) and hyperleukocytosis are at high risk of early mortality due to pulmonary, renal, and central nervous system complications. Leukapheresis and low-dose continuous infusion of cytarabine and hydroxyurea (HU) have been used but their advantages and limitations are not well characterized. The University hospital in Casablanca, Morocco has a limited number of beds, and hence admissions must be prioritized. Here we report the effects of HU on the white blood cell (WBC) count and early mortality rate of patients with hyperleukocytic AML. METHODS: Between April 2003 and December 2006, patients with AML were enrolled on the AML-MA2003 protocol (2 induction courses of cytarabine and daunorubicin and 2 postremission courses that include intermediate-dose cytarabine). Patients with AML and hyperleukocytosis (WBC count >50 x 109/L) were immediately started on HU (50 mg/kg/day orally x 4 days), regardless of hospital bed availability. Response was evaluated after 4 days of HU; patients were considered responders if >50% reduction of the initial WBC count was observed. RESULTS: Ninety of 260 (34.6%) patients enrolled had hyperleukocytosis. Three patients were excluded, induction therapy started on the day of admission. Therefore, 87 patients (48 females, 39 males) were evaluable. The mean age was 32 years (range, 2–60); 29% were children (ages 2–20 years). The mean initial WBC count was 104x109/L (range 50–260 x109/L); 37 (42.5%) patients had WBC counts > 100x109/L. The French-American-British subtypes were M1 (45%), M2 (26%), M4 (12%), M5 (7%), and M0, M3, M6 and M7 (2.5% each); 5 cases were unclassified. Karyotypes determined for 65/87 cases revealed 13 (20%) favorable karyotypes [9 had the t(8;21); 3 had inv16; 1 had the t(15;17)], 30 (46%) intermediate-risk karyotypes, including normal karyotypes, and 22 (34%) unfavorable-risk karyotypes. Sixty-two (71%) patients were classified as responders. In an additional 3, the WBC count was reduced 25%–50%. In 22 (25%) patients, including 4 whose WBC counts increased, HU showed no cytoreductive effect. The mean WBC count after 4 days of HU was 24 x 109/L (range, 1.5–125 x109/L); 15 (17%) patients’ WBC counts remained >100x109/L. Four patients developed acute tumor lysis syndrome (TLS) (hyperuricemia and renal dysfunction) in response to HU. There were 5 (8%) early deaths (mean, 7 days after the start of HU; range, 4–14 days). All 5 patients had WBC counts > 100x109/L at diagnosis, and only 1 was a responder. This mortality rate does not differ from that (9%) observed among the 170 protocol patients who did not have hyperleukocytosis. Causes of death included infection (n=1), pulmonary and CNS leukostasis (n=1 each), TLS (renal failure and hyperkalemia, n=1), and intracranial hemorrhage (n=1). Among several factors (age, sex, FAB type, karyotype, WBC counts), only WBC count of ≤ 100x109/L was significantly associated with response to HU (P=0.01). The complete remission rate after first course of induction therapy was 43.5% for responders and 16% for non-responders (P=0.02). CONCLUSION: HU given orally for 4 days rapidly reduces the WBC count in pediatric and adult AML with hyperleukocytosis. The early mortality rate in this high-risk group treated with HU compares favorably with rates reported for similar patients. It remains to be determined whether initial response to HU is associated with overall outcome in AML.