Daunorubicin and cytarabine liposomal (CPX-351) plus glasdegib for newly diagnosed secondary Acute Myeloid Leukemia: Results of University of California hematologic malignancies consortium trial 1913

Abstract

Abstract Intro Secondary acute myeloid leukemia (sAML) from antecedent hematologic disorders or prior cytotoxic therapy is associated with adverse risk cytogenetics, poor prognosis, and inferior response to induction therapy. Patients (pts) receiving CPX-351 who achieved complete remission (CR) and CR with incomplete hematologic recovery (CRi) had superior overall survival (OS) than the standard 7+3 which led to its approval as a preferred regimen for sAML. Nevertheless, CR rate with CPX-351 is modest. Glasdegib, a hedgehog pathway inhibitor, shows synergistic activity with low-dose cytarabine and may delay relapse by targeting leukemic stem cells. As CPX-351 and glasdegib have complementary actions, combining them could improve responses/remission duration in sAML. Method A University of California Hematologic Malignancies Consortium (UCHMC) investigator-initiated phase II trial (NCT04231851) enrolled adults with newly diagnosed AML with MDS related changes or therapy-related AML as defined by WHO2016 criteria. Pts were treated with CPX-351 (100units/m² IV on days 1,3,5) and glasdegib (100mg orally daily on day 6-28 or count recovery if in remission). Day 20 bone marrow (BMBx) results guided re-induction if>5% blasts remained. Responders based on BMBx results received up to 2 CPX-351 consolidation cycles (65units/m² days 1,3) plus glasdegib 100mg daily, followed by optional maintenance glasdegib 100mg daily for up to 12 months (mos) if not proceeding to allogeneic hematopoietic cell transplant (alloHCT). Primary endpoints were 6-month event-free survival (EFS; defined as time from enrollment to relapse/progression, treatment refractory, or death); secondary endpoints included post-induction overall response rate (ORR), 6 month OS, relapse-free survival (RFS), alloHCT rate, and safety. Assuming a historical control of 50% 6 month EFS, the hypothesis was that this combination increases 6 month EFS to 65%. Results From 2020–24, 29 pts (median age 68.5, range 25–82) were treated at 4 UCHMC centers. Per 2017 ELN criteria, 90% had adverse-risk. Subtypes included AML from antecedent hematologic malignancy (41%), MDS-related changes (41%), and therapy-related AML(17%). TP53 mutations were present in 4. Grade 5 adverse events (AEs) occurred in 5 (sepsis[2], respiratory failure, neutropenic colitis, mucormycosis). There were 41 grade 3–4 non-hematologic treatment-related AEs including febrile neutropenia and sepsis. The safety profile was manageable and consistent with expected toxicity. Post-induction therapy, 13/29 evaluable pts achieved CR and 2 CRi(ORR 51.6%). Two required a second induction. One had MRD negative (MRDneg) CR and the other had persistent disease. Among responders, 73%(11/15) were MRDneg by multiparametric flow cytometry(MFC). Notably, 3/4 with TP53 mutations (TP53mut) achieved CR/CRi with MRDneg by MFC and clearance of the TP53mut post induction. Median time to count recovery was 40 days(range 27–61). Ten had persistent disease and received salvage therapy. Four died during induction, yielding a 13% 60-day mortality rate and BMBX disease results were not obtained. Among responders, 13 received consolidation, 6 underwent alloHCT and 1 went to alloHCT post induction so 7/15 (47%) responders went to alloHCT. Three received only maintenance glasdegib. Median follow up duration was 24 mos. 6-month EFS and OS probabilities were 31%(95%CI 15-48) and 62%(95%CI 42-76), respectively. This combination did not improve 6-month EFS over a baseline 6-month EFS of 50% and the primary end point was not met. OS was 10.8 mos in all pts(IQR 4.3, 25.7). For responders, median OS was 17.5 mos when censored at the time of alloHCT. Median OS was improved at 27.5 mos for alloHCT compared to 5.1 mos for non-alloHCT responders. Three of 4 TP53mut responders had a median OS of 13.6 mos. Relapse occurred in 47% of responders(median time 186 days, range 56–709). Seven relapsed post-alloHCT, 4 relapsed 1 year post-alloHCT and 2 relapsed 2 years post-alloHCT. Achieving MRDneg may be a key determinant of durable response in sAML. Among responders, 73% were MRDneg, including all 3 TP53mut pts with CR/CRi. Thus clearance, not just morphologic remission, may be a critical endpoint in sAML. Conclusion Despite encouraging safety/activity, the study did not meet the primary endpoint of EFS. The CPX-351/glasdegib combination may offer benefit in sAML, particularly as a bridge to alloHCT. Encouraging responses with MRDneg rates were observed in TP53mut pts.