ALBERT

Description: Abstract 1071 Background: Treatment of children with acute myeloid leukemia (AML) is associated with considerable toxicity. Children's Oncology Group (COG) AAML0531 trial adopted a modified AML Medical Research Council backbone and reviewed adverse event reports in real time to maximize accurate toxicity data. As of March 31, 2010, AAML0531 had randomized 968 patients with de novo AML to gemtuzumab ozogamicin (GMTZ) versus no GMTZ. Accurate description of toxicities with the backbone regimen as well as increment toxicities of new agents is important to optimize supportive care. Objectives were to describe hepatic, cardiac and infectious toxicities and duration of neutropenia in all patients, and to compare toxicities between treatment arms. Methods: AAML0531 included those ≥ 1 month to ≤ 30 years with de novo AML and used a 5 cycle chemotherapy regimen: Induction (Ind) I and II: cytarabine, daunorubicin, and etoposide (ADE 10+3+5 and 8+3+5); Intensification (Int) I: cytarabine and etoposide; Int II: mitoxantrone and high-dose cytarabine; and Int III: high-dose cytarabine and L'asparaginase. Subjects were randomized to receive or not receive GMTZ 3 mg/m2 during Ind I and Int II. Common Terminology Criteria for Adverse Events v3.0 toxicities were collected prospectively. Hepatic (ALT, bilirubin and veno-occlusive disease (VOD)), cardiac (left ventricular systolic dysfunction (LVSD)) and infection (sterile site bacterial and fungal) toxicities were targeted. All grades of cardiac and VOD toxicities were captured; all other severe (≥ grade 3) toxicities were recorded. Cumulative incidence (CI) of toxicities was calculated only during protocol chemotherapy. Duration of neutropenia was reported from beginning of cycle to absolute neutrophil recovery ≥ 500/uL for patients alive during the cycle. Results: Table describes the prevalence of severe (≥ grade 3) toxicities and illustrates that severe high ALT occurred in 3–10%, high bilirubin was less common and VOD was reported in ≤ 1% of cycles. Severe LVSD also was rare although compliance with suggested cardiac monitoring ranged from 50–80% per course of treatment. In terms of grade 1 or 2 LSVD there were no differences in any course except for Int II where grade 1 or 2 LVSD was higher with GMTZ (8.0% versus 2.4%; P