Hypomethylating agent/venetoclax versus intensive chemotherapy in adults with relapsed or refractory acute myeloid leukaemia

Abstract

Outcomes of patients with relapsed or refractory acute myeloid leukaemia (r/r AML) remain poor with long-term survival ranging around 10%–20%. Treatment options are limited and outside of clinical trials and in the absence of a targetable mutation, there is no consensus on an optimal regimen.1, 2 Intensive chemotherapy (IC) has frequently been used in the salvage setting but is associated with significant toxicities, restricting its use. Venetoclax (Ven), combined with hypomethylating agents (HMA) such as azacitadine (Aza) or decitabine (Dec), is approved as front-line treatment for elderly or unfit patients with AML.3 HMA/Ven is now also being used frequently in the salvage setting. However, outcomes of HMA/Ven compared to IC for patients with r/r AML are largely unknown. Within this context, we conducted a retrospective study to compare outcomes of patients with r/r AML treated with either HMA/Ven or IC at our institution. We identified patients treated with HMA/Ven. Variables captured included age, gender, bone-marrow blast percentage, risk stratification per cytogenetics and European LeukemiaNet (ELN) 2017 classification, time to relapse (primary refractory/<6 months vs ≥6 months) and line of therapy. Treatment with HMA consisted of either Aza 75 mg/m2 for seven days or Dec 20 mg/m2 for five days. Using 1:1 matching, we identified patients with r/r AML treated with IC. Matching followed a hierarchal algorithm based on age, ELN risk stratification, time to relapse and line of therapy. Outcomes of interest included complete remission (CR) rate, CR with incomplete haematological recovery (CRi) rate, composite CR (CR + CRi) rate and overall survival (OS). Measurable residual disease (MRD) was assessed using flow cytometry validated to a sensitivity level of 0.1%. Statistical analysis is provided in Appendix S1. We identified 53 patients treated with HMA/Ven and matched them to 53 patients that received IC. Baseline characteristics are provided in Table 1 and were well balanced. Regimens in the IC and HMA/Ven cohorts are provided in Appendix S1. The rate of CR was higher for IC (49% vs 23%, p = 0.02) whereas rate of CRi was higher for HMA/Ven (36% vs 8%, p = 0.001). There was no difference in composite CR rates (IC 57% vs 59% HMA/Ven, p = 0.8). The rate of MRD-negative composite CR was 46% with HMA/Ven and 43% with IC (p = 0.2). Four patients (8%) with HMA/Ven and six (11%) with IC achieved the morphological leukaemia-free state (MLFS). Additionally, there was no difference in 30-day mortality (IC 13% vs 11% HMA/Ven, p = 0.6) between the two cohorts. More patients in the IC cohort (41%), compared to the HMA/Ven (20%) cohort proceeded to transplant (p = 0.04). The median OS (mOS) for the entire cohort was 16 months with IC and eight months with HMA/Ven (p = 0.1) (Figure 1A). The mOS for patients with primary refractory/<6 months relapse was nine months for IC and six months for HMA/Ven (p = 0.4). The mOS for patients relapsing after at least six months was 15 months for IC and 10 months for HMA/Ven (p = 0.5). Relapse-free survival (RFS) for the entire cohort was nine months with IC and five months with HMA/Ven (p = 0.2). There was no significant difference in survival based on age, ELN risk stratification or cytogenetics. There were 20 patients (38%) that received HMA/Ven and 24 patients (46%) that received IC as salvage therapy after not responding to one cycle of intensive induction with traditional 7 + 3 chemotherapy. All patients in the IC cohort received a high-dose cytarabine-based regimen [fludarabine, cytarabine, granulocyte colony-stimulating factor (FLAG) = 16; FLAG with idarubicin (FLAG-ida) = 2; cladribine, cytarabine, granulocyte colony-stimulating factor, mitoxantrone (CLAG-M) = 6]. The mOS for this population was 20 months for the IC cohort and five months for the HMA/Ven cohort (p = 0.03) (Figure 1B). In this analysis, we compared outcomes of HMA/Ven to IC in r/r AML and report that while the rate of CR was higher with IC and rate of CRi was higher with HMA/Ven, the composite responses were similar. Furthermore, OS was not statistically different between both cohorts for the entire population but favoured the IC cohort for patients refractory to one cycle of intensive induction chemotherapy. A HMA/Ven is now being used frequently in the salvage setting with reported CR/CRi rates of 35%–46%.4-8 However, durability of response, in the absence of transplant, remains suboptimal. The mOS of HMA/Ven in r/r AML has ranged from three to 9 nine months.1, 4 As a bridge to transplant, the 1y OS has ranged from 68%–77%.4, 9 Our finding of a modest mOS rate despite high remission rates, with HMA/Ven, is likely explained by a low rate of transplant in this cohort. Overall, there was no statistically significant difference in survival between the two cohorts. However, we observed a significant survival benefit for patients receiving high-dose cytarabine-based therapy after not responding to one round of 7 + 3, when compared to those who received HMA/Ven. The treatment of patients with AML, eligible for IC, has typically consisted of induction therapy with 7 + 3. Around 10%–40% of patients may not achieve CR to one round of an anthracycline-containing induction therapy.10 The decision process to give those patients a second round of IC versus venetoclax-based therapy is not well defined. The only other study comparing HMA/Ven to IC for r/r AML utilized 10 days of Dec, instead of five, with the initial IC regimens already incorporating high-dose cytarabine plus an anthracycline with or without purine analogues. That study demonstrated a higher rate of CRi and a modest improvement in OS with Dec/Ven, compared to IC.1 Of note, in our subset of patients refractory to one cycle of induction therapy, 71% patients receiving IC and 29% patients receiving HMA/Ven proceeded to transplant, perhaps providing some insight into the survival difference. The retrospective nature of our study limits our findings. Our study may have been underpowered to detect a significant difference in survival outcomes between the two treatment arms as well as based on age, ELN risk stratification or cytogenetics. Most notably, the selection bias associated with either switching to HMA/Ven or proceeding with a second round of chemotherapy in patients refractory to 7 + 3 may have influenced outcomes. For patients refractory to the first cycle of intensive induction chemotherapy, a significant survival benefit was observed in those receiving IC compared to HMA/Ven. If tolerable, a second round of induction, preferably with a high-dose cytarabine-based regimen, may provide better long-term outcomes for these patients. These results warrant prospective studies comparing the two treatments with inclusion criteria of relapsed AML or primary refractory disease, defined as refractory disease after two cycles of IC (or one cycle if it incorporates high-dose cytarabine). Omer Jamy and Sravanti Rangaraju contributed towards the conception of the presented study, data collection and analysis. Omer Jamy, Karen Lin, Sarah Worth, Kimo Bachiashvili, Sravanti Rangaraju, Pankit Vachhani and Ravi Bhatia contributed towards drafting, revising, and approving the manuscript. No funding sources to disclose. Pankit Vachhani has received consulting fees from AbbVie, Amgen, Blueprint Medicines, CTI biopharma, Genentech, Incyte, Novartis, and Pfizer. All other authors have no relevant conflict of interest to disclose. The data that support the findings of this study are available on request from the corresponding author. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts