Remission Rates In Childhood Acute Myeloid Leukemia (AML) Utilizing a Dose-Intensive Induction Regimen with or without Gemtuzumab Ozogamicin (GO): Initial Results From the Children's Oncology Group Phase III Trial, AAML0531

Abstract 182

Acute Myeloid Leukemia - Therapy, excluding Transplantation: Pediatric and Adult AML Therapy

Therapy for childhood AML has evolved in North America over the past decade by pursuing a dose-intensive rather than an intensive-timing strategy. COG AAML0531 was a recently completed Phase III trial that was based upon the MRC AML 12 dose-intensive regimen which reported a complete remission (CR) rate of 93% in their ADE arm with 3% induction (Ind) deaths (Gibson, Blood 100:35a, 2002). While survival outcomes remain “blinded” as the last patients (pts) complete therapy, remission (REM) outcomes can now be examined. Utilizing a data cutoff of March 31, 2010, 1009 non-M3 AML pts were enrolled and after excluding 36 ineligible and 5 DS pts 968 eligible de novo AML were examined. Pts were randomized to standard therapy with or without GO at 3 mg/m² given on Ind I day 6 and intensification II day 7. Induction consisted of 2 therapy courses (Ind I: ADE10 & Ind II: ADE8), cytarabine 100 mg/m²/dose bid × 10 days (8 days for Ind II), etoposide 100 mg/m²/day on days 1–5, and daunomycin 50 mg/m²/day on days 1–3-5. Pts remained on the trial regardless of REM status after Ind I; however pts not in CR after Ind II were taken off protocol. There were 968 pts that began Ind I and 851 pts that began Ind II, with 25 & 7 withdrawals, 50 & 29 who were still in Ind or had not yet submitted data, and 19 & 4 deaths in each course, respectively. CR (defined in the protocol as <5% morphologic blasts (blasts) & extramedullary disease (EMD) resolved) was achieved in 70% (628/900) & 86% (731/854) by the end of each Ind course, respectively. This is similar to that seen in the intensively-timed CCG-2961 after 2 Ind courses. The impact of REM status after Ind I upon CR rates after Ind II was then assessed. After Ind I (ADE10), partial REM (PR) (5-15% blasts) was seen in 12% (104/900) and persistent disease (PD) (>15% blasts) was seen in 14% (126/900) of whom 81% (79/98) & 60% (65/109) entered CR after Ind II (ADE8), respectively. The extent of PD after Ind I and its impact upon CR after Ind II was then examined. For those whose PD was defined by only residual EMD and whose marrow was in PR or CR, 97% achieved a CR after Ind II. For those whose PD was defined by marrow blasts >15%, 42% achieved CR following Ind II. We examined whether the degree of marrow disease (15-30% vs >30% blasts) impacted CR in the PD pts but found no significant difference in CR by amount of PD (52% vs 36%, p=.234). Among the 25 pts who withdrew from Ind I, response after Ind I was assessable in 19 (4 CR, 1 PR, 14 PD) although no impact upon Ind II outcome could be ascertained. Diagnostic characteristics were analyzed for CR rate after Ind II with selected risk factors listed in the table. Significant prognostic factors for REM after 2 courses were found by univariate analysis to include WBC>100,000 (OR=2.7, p<.001), FLT3-ITD (OR=2.9, p<.001), low risk cytogenetics (OR=0.2, p<.001). In a multivariate model in 464 pts who had all three risk factors reported to date, the same risk factors were independently predictive of outcome: WBC (OR=2.5, p=.002), FLT3-ITD (OR = 2.0, p=.034), and low risk cytogenetics (OR = 0.24, p=.007). FLT3 analysis for this abstract includes only those clinically available (after a trial amendment in the 3^rd year). This will be updated with the FLT3 research sample analyses for those pts enrolled prior to the amendment. Overall toxic mortality by the end of Ind II of 2.7% (2.1% in ADE10, and 0.5% in ADE8) are similar to reported outcomes in the COG pilot trial, AAML03P1, (2.6%) and better than that seen in CCG-2961 (14.1% & 10%, pre-& post-amendment). These data provide an important background for the next COG Phase III trial utilizing this same standard Ind due to open soon and provides a platform for an early comparison between outcomes of the MRC trials and those in COG with identical Ind courses.