A novel therapy with a rational design for AML

In this issue of Blood, Liu et al¹ report the results of a phase 1 clinical trial using pegcrisantaspase (PegC) in combination with venetoclax in relapsed and refractory acute myeloid leukemia (AML). This study presents safety and preliminary efficacy of this combination, along with proof of principle correlative studies showing that response to PegC and venetoclax correlated with protein synthesis inhibition.

AML is a devastating disease that predominantly affects the aging population, with a median age of diagnosis of 69 years. The standard treatment since 1973 has been induction chemotherapy with cytarabine and an anthracycline. However, this regimen is limited by significant treatment-related mortality in older patients, particularly those over 75 years of age.² A new era began in 2017 with the Food and Drug Administration (FDA) approval of targeted therapies such as midostaurin and enasidenib against AML. This was followed by the approval of the frontline AML therapy with venetoclax, an inhibitor to the antiapoptotic protein BCL-2, and a hypomethylating agent for older adults with AML.³ Although these new therapies have improved outcomes in AML, no targeted therapy has been curative to date. Hematopoietic stem cell transplantation remains the most curative form of treatment, but older patients may not be eligible for transplant due to comorbidities. Once patients relapse, survival is dismal,^4,5 leaving a large unmet need for treatment for this population.

PegC is a pegylated form of Erwinia asparaginase and it is derived from Erwinia chrysanthemi. It has been previously used for patients with hypersensitivity to Escherichia coli–derived asparaginase products in other hematologic malignancies such as acute lymphoblastic leukemia.⁶ PegC has been shown to deplete cells of glutamine as well as affect protein translation.^7,8 In this phase 1 clinical trial, a 3 + 3 design was used for the dose escalation phase. Venetoclax was used per FDA label, and PegC was escalated at doses 500, 750, and 1000 IU/m² and given on days 1 and 15 on 28-day cycles. The primary end points were maximum tolerated dose and incidence of regimen-limiting toxicities.

A total of 27 patients were enrolled in the trial, 25 of whom underwent treatment. The median number of prior treatment regimens was 3, and median age was 63. In the treatment group, 68% of patients enrolled had previously received venetoclax-based regimens. The most common adverse events were hyperbilirubinemia (60%) and decreased antithrombin III (52%), although there were no reports of grade ≥3 hemorrhage or thrombosis. The most common grade ≥3 adverse events were febrile neutropenia (56%), sepsis (24%), elevated aspartate aminotransferase (24%), and hyperbilirubinemia (20%). Based on drug-limiting toxicities, the recommended phase 2 dose for the expansion phase was PegC 750 IU/m² intravenously on days 1 and 15, with venetoclax 400 mg orally daily on a 28-day cycle. Regarding responses, 33% of the patients achieved either a complete remission (CR) or a CR with incomplete blood count recovery (CRi) or a CR with partial blood count recovery (CRh). Importantly, half of those patients had no measurable residual disease, signifying a deep remission. Given many patients on this study had relapsed after more than 1 therapy, deep remissions in this patient population are very compelling.

In an era when venetoclax in combination with a hypomethylating agent is the standard-of-care frontline AML therapy for older patients, it is important to assess efficacy of new regimens for patients who have relapsed after receiving venetoclax with a hypomethylating agent. In this study, responses were seen in patients who previously received venetoclax-based therapies. These patients also showed a 33% rate of CR, CRi, or CRh, suggesting that prior venetoclax and azacitidine treatment does not preclude responses with PegC in combination with venetoclax. It is worth noting, however, that patients who received prior venetoclax-based therapies were more likely to have measurable residual disease, suggesting remissions were not as deep and could lead to earlier relapse. Patients with RUNX1 mutations appeared to have the best responses, with 4 of 4 patients achieving CR or CRi. Median overall survival for patients who responded was 100 days.

Various groups have shown that AML cells are dependent on glutamine,⁹ making it a good target for this disease. PegC has been shown to decrease glutamine levels in AML cells. The authors also found that glutamine depletion using the combination of PegC and venetoclax in preclinical models led to a decrease in cap-dependent protein translation by inhibiting phosphorylation of the translation initiation factor 4EBP1 (see figure).⁸ In patient-derived xenograft models, the combination of PegC and venetoclax decreased protein expression of the antiapoptotic protein MCL-1, with no decrease in BCL2, the main target of venetoclax. Given that venetoclax resistance in AML has been shown to correlate with MCL-1 dependence,¹⁰ this suggests that PegC plus venetoclax can overcome venetoclax resistance via this mechanism. For patients who previously received venetoclax-based therapy, one could question whether venetoclax adds benefit in combination with PegC. However, preclinical data showed more robust protein synthesis inhibition and decreased MCL-1 expression in AML cell lines and a patient-derived xenograft with the combination therapy compared with single-agent PegC.⁸ This suggests the combination of PegC and venetoclax could be more efficacious than single-agent PegC, although this has not been investigated in the context of a clinical trial.

In this study, the authors show promising data for this combination therapy in a population with very limited treatment options. Responses to this regimen are very compelling, and prior venetoclax-based therapies do not preclude responses. With regard to safety, patients with relapsed and refractory AML often exhibit prolonged neutropenia, and toxicities from this regimen may limit its use in some patients. High rates of febrile neutropenia and sepsis could be a limiting factor for frailer patients with AML. In conclusion, promising results from this trial suggest this should be studied in larger phase 3 clinical trials.

Conflict-of-interest disclosure: M.L.A. declares honoraria from Bristol Myers Squibb.