Combination of CPX-351 and gemtuzumab ozogamicin (GO) in relapsed/refractory (R/R) Acute Myeloid Leukemia and post-hypomethylating agent (HMA) failure high-risk myelodysplastic syndrome (HR-MDS)

Abstract

Abstract Background: The outcome of patients (pts) with R/R AML or MDS after failing HMA and/or BCL-2 inhibitor combinations is poor and lacks effective treatment options. CPX-351 (Vyxeos™) is a liposomal formulation of cytarabine and daunorubicin approved for therapy-related AML and AML with myelodysplasia-related changes. Gemtuzumab ozogamicin (Mylotarg™) a CD33-targeted antibody-drug conjugate, is approved in both frontline and R/R AML. We hypothesized that combining CPX-351 and GO (CPX-GO) may offer synergistic efficacy in patients with CD33+ R/R myeloid malignancies. Methods: Here we present updated results of this single-arm, pilot study (NCT03672539) evaluating CPX-GO in patients with CD33+ R/R AML or post-HMA failure HR-MDS (>10% blasts). Induction consisted of CPX-351 (daunorubicin 44 mg/m² + cytarabine 100 mg/m² IV on days 1, 3, 5) plus GO 3 mg/m² (capped at 4.5 mg) IV on day 1. Patients not achieving complete remission (CR) or CR with incomplete count recovery (CRi) after cycle 1 (C1) could receive re-induction (CPX D1, 3 + GO D1). Responders could receive up to 2 consolidation cycles (CPX 29/65 mg/m² D1, 3), with GO added in cycle 2 only if measurable residual disease (MRD) was present. GO maintenance (D1 every 6 weeks) was allowed in MRD-positive patients. Responses were assessed by ELN 2017 (AML) and IWG 2018 (MDS). Overall survival (OS) was calculated from the start of treatment to the date of death or last follow up. Progression-free survival (PFS) was defined as the time from treatment initiation to disease progression, change in therapy, or death from any cause. Results: From Nov 2018 to Jul 2025, 51 patients were treated (median age 68 [21–77]; 68% male). Among 50 patients included in this analysis, diagnoses included R/R AML (46) and HR-MDS (4). Median prior therapies: 2 (range 1–6); 41 patients (82%) had received prior Venetoclax in combinations with HMA's and/or chemotherapy and 9 patients had prior stem cell transplant (SCT). Complex karyotype was present in 18 (36%), 4 (8%) pts had MECOM rearrangement (r), and 3 pts had KMT2Ar, 1 pt had multiply relapsed core-binding factor AML with additional cytogenetic changes. Most frequent molecular mutations were RUNX1 found in 14 patients (28%), ASXL1 in 13 (26%), TET2 and NRAS were both found in 12 patients (24.5%) and TP53 in 9 (18%). Patients received a median of 1 cycle (range 1–6) with 3 pts receiving GO maintenance therapy. One patient did not complete the induction cycle and was not included in the assessment. We observed an overall response rate (ORR) of 30% (n=15), including CR (n=6), CRi (n=7), PR (n=1) and MLFS 1. All responding patients were TP53 wild type. The median overall survival for the cohort is 4.8 months (95% CI approximately 4.1–5.3 months). Among the responders (CR, CRi, PR, MLFS), the median overall survival was 9.7 months (95% CI, 7.2–12.4 months), median PFS was 3.6 months (95% CI, 2.8–4.9 months). Among responders, the median time to ANC >0.5 x109/L was 43 days (Range 28-81) and PLT >50 x109/L was 36.5 days (Range 11-76). At data cutoff 5 pts are alive, with one still on trial therapy. Thirty-day mortality was 8% (n=4); 60 days mortality rate was 20% (n=10), all grade 5 events were unrelated to therapy. Febrile neutropenia was the most common grade ≥3 event, in 24 pts (48%), followed by sepsis-bacteremia in 20 pts (40%) and pneumonia in 10 pts (20%). There was no treatment-related grade ≥3 non-hematological toxicity. Conclusion: The combination of CPX-351 and GO is feasible and shows modest activity. OS was meaningful in responders, considering multiple prior lines of therapy. The combination showed a tolerable safety profile with minimal treatment-related toxicity. Infections stood out as the most common serious complications.