Updated Results of CPX-351 in Combination with Gemtuzumab Ozogamicin (GO) in Relapsed Refractory (R/R) Acute Myeloid Leukemia (AML) and Post-Hypomethylating Agent (Post-HMA) Failure High-Risk Myelodysplastic Syndrome (HR-MDS)

Background:

Treatment outcomes in patients (pts) with AML and HR-MDS who progress after HMA +/- venetoclax (VEN) based regimen remains poor and warrants new therapeutic strategies. CPX-351 (Vyxeos™) is a dual liposomal formulation of cytarabine and daunorubicin, at a fixed synergistic 5:1 molar ratio, approved by US Food and Drug Administration (FDA) for the treatment of newly diagnosed therapy-related AML or AML with myelodysplastic related changes (AML MRC). GO (Mylotarg™) is a humanized IgG4 antibody directed against CD33 and conjugated to the DNA toxin calicheamicin and approved by the FDA for the treatment of newly diagnosed or R/R CD33-expressing AML. We hypothesized that the combination of CPX-351 and GO could be synergistic in this high risk pt population and have potent anti-leukemic activity.

Aim:

To determine the safety and efficacy of CPX-351 in combination with GO in patients with R/R AML or post-HMA failure HR-MDS.

Methods:

We present here the updated results of the Phase 2 single-arm pilot study (NCT03672539) of CPX-351 and GO (CPX-GO) combination therapy in pts with CD33 positive R/R AML or post-HMA failure HR-MDS.

Pts received induction cycle (C) CPX-351 (daunorubicin 44 mg/m² and cytarabine 100 mg/m²) administered via intravenous (IV) infusion on days (D) 1, 3, and 5. GO was administered at a dose of 3 mg/m² (capped at 4.5 mg) IV on D1. Pts not attaining complete remission (CR) or CR with incomplete count recovery (CRi) after C1, could receive a 2nd induction cycle of CPX at the same dose, but only on D1 and D3 with GO 3 mg/m² on D1. Pts attaining CR/CRi could receive up to 2 consolidation cycles, after a minimum of 4 weeks from the start of the previous cycle with CPX (daunorubicin 29 mg/m² and cytarabine 65 mg/m²) IV on D1 and D3 and GO at 3 mg/m² on D1. GO was only administered during the second consolidation cycle if there was evidence of measurable residual disease (MRD+) as measured by flow cytometry. GO could also be administered as a single agent for maintenance treatment on D1 every 6 weeks, in case of persistent detection of MRD. CD33 genotyping to assess for GO sensitivity was available for a subset of pts. Response was denoted as per the ELN 2017 criteria for AML and IWG 2018 criteria for MDS.

Results:

From Nov 2018- Apr 2022, 31 pts were treated on the trial, 29 (93%) pts with AML and 2 pts (7%) with HR-MDS. Baseline disease, treatment and response characteristics are mentioned in Table 1. The median lines of prior AML therapy were 2 (range, 1-6). Eleven pts (38%) had baseline de novo AML, while 15 (52%) and 3 (10%) pts had secondary (s-AML) and therapy-related (t-AML) AML respectively. Twenty-three AML pts (79%) had prior VEN exposure. Both the prior HR-MDS patients had bone marrow (BM) blasts >15% and one pt had prior VEN exposure. Overall, 21 pts with AML (72%) were adverse risk and only 1 pt (3%) had favorable risk AML (diploid karyotype with NPM1 mutation). Four pts (all AML; 13%) had undergone a prior stem cell transplantation (SCT).

The median number of cycles on trial were 1 (range, 1-6). At the time of data cutoff, none of the pts were on active study therapy and the median follow up was not reached. Sixteen pts (52%) had an overall response; 9 pts had CR/CRh/MLFS (8 AML and 1 MDS= responders), after a median of 1 cycle of therapy, and 7 pts had a partial response (all AML). CD33 genotype did not affect response rates (no response in 5 pts with C/C vs 2/6 responders in C/T genotype, p=0.45). Amongst prior VEN exposed pts, 5/24 (21%) responded versus 3/7 (43%) non VEN exposed pts (p=0.15)

One pt could directly proceed to a SCT in CR after the trial therapy. For relapse free survival (RFS) analysis partial responders were grouped under non-responders. The median RFS and overall survival (OS) for the whole cohort was 6.3 and 6.8 months (mos) respectively, and median OS was 12.9 mos for responders (Fig. 1).

One pt had a grade 2 infusion reaction to GO. Overall, adverse events at least possibly attributed to the trial medications were seen in 5 pts, of which 2 events were ≥ grade 3 (mucositis and seizures). No pt had hepatic veno-occlusive. Eight pts (all non-responders) (26%) died on the trial from infections/leukemia progression and there were no deaths in remission.

Conclusion:

In a cohort composed predominantly of heavily pre-treated and VEN exposed adverse risk AML and HR-MDS pts, CPX-GO led to overall response (CR/CRh/MLFS/PR) rates of 52% and year-long median survival in responders (CR/CRh/MLFS) with no major non-hematological adverse event.