Significance of Prior HSCT on the Outcome of Salvage Therapy with CPX-351 or Conventional Chemotherapy Among First Relapse AML Patients

CPX-351 is a liposomal formulation containing a 5:1 molar ratio of cytarabine (Ara-C) and daunorubicin (DNR) designed to maximize anti-tumor synergy (Mayer L, et al. Mol Can Ther, 2006;5:1854). In the Phase 1 study 18/45 AML patients were in 1^st relapse with 8 patients ≤65 years of age. All 8 achieved aplasia and 4/8 achieved CR after CPX-351 treatment. (Feldman E, et al. Blood 2008;112:Abst 2984). These observations provided the rationale for a randomized Phase 2b study in 1^st relapse AML patients comparing CPX-351 against investigator's choice of salvage regimen. An initial safety and efficacy summary has been submitted to this meeting. This report summarizes the impact of prior HSCT on patient response and 60 day mortality.

Patients ≤65yo with AML in 1^st relapse after an initial CR lasting >1 month with ECOG PS= 0–2, S_Cr <2.0 mg/dL, total bilirubin <2.0 mg/dL, ALT/AST <3 × ULN, and LVEF ≥50% were eligible. Patients were randomized 2:1 to receive CPX-351 (100 u/m²; D 1, 3, 5) or investigators choice of intensive salvage treatment. Up to 2 induction and 2 consolidation courses were allowed. Post remission treatment with HSCT was permitted. Patients were stratified using the European Prognostic Index (EPI) (Breems DA, et al. J Clin Oncol, 2005 Mar 20;23:1969–1978). The primary efficacy endpoint is % survival at 1 year.

By November 12, 2010, 126 patients were accrued at 35/46 sites in the US, Canada, France, Poland. Data for response, acute safety, and 60 day deaths are available. Patients in both arms were well balanced by Relapse Free Interval (RFI), age at 1^st relapse and cytogenetics. More patients with prior HSCT were assigned to the CPX-351 arm (27% v. 18%). Table 1 shows patient characteristics organized by presence or absence of prior HSCT. Among patients with prior HSCT, the CPX-351 arm had more patients with RFI ≤6 months (36% v. 0%) and younger patients (age <45, 36% v. 13%). The group with no prior HSCT was balanced for overall risk by EPI, RFI, age at 1^st relapse, and adverse cytogenetics.

For the overall study, the CPX-351 arm had higher aplasia (90% v. 60%) and CR + CRi rate (51% v. 42%) and similar 60 day mortality (15% v. 16%) compared to control.

Table 2 shows patient response and early mortality organized by presence or absence of prior HSCT. In the 96 patients (76%) with no prior HSCT, CPX-351 produced an increased rate of aplasia compared to control (90% v. 57%), a response rate that was greater than control (56% v. 41%), and a lower 60 day mortality rate (10% v. 16%). HSCT had little observed impact on outcomes in the control arm but appears to be associated with reduced aplasia (90% to 55%) and response (56% to 36%) and a disproportionate contribution to early deaths (6/12) in the CPX-351 arm. The higher proportion of patients with RFI ≤6 months on the CPX-351 arm (36% v. 0%) may be a contributing factor.

Eighty six/126 (68%) patients had unfavorable risk by EPI with equal proportions randomized to each arm of the study. Prior HSCT had little impact on outcomes in the control arm but seemed to have a much larger adverse impact on the CPX-351 arm. This observation of a change in outcome associated with prior HSCT may be due to the higher proportion of patients with RFI ≤6 months on the CPX-351 arm (36% v. 0) and is being evaluated further. Additional data (EFS, % survival at 1 year) are expected later this year.

Cortes:Celator: Membership on an entity's Board of Directors or advisory committees. Louie:Celator: Employment.