Venetoclax with decitabine or azacitidine in the first‐line treatment of acute myeloid leukemia

Abstract Treatment paradigms for acute myeloid leukemia (AML) have evolved at a rapid pace in recent years. The combination of venetoclax with a hypomethylating agent prolonged survival in clinical trials when compared to hypomethylating agent monotherapy. However, little is known about the performance of venetoclax‐based regimens outside of clinical trials, given conflicting safety and efficacy data. Even less is known about the impact of the hypomethylating agent backbone. In this study, we demonstrate that decitabine‐venetoclax is associated with a significantly higher rate of grade three or higher thrombocytopenia, but lower rates of lymphocytopenia compared to azacitidine‐venetoclax. There was no difference in response or survival across ELN 2017 cytogenetic risk categories in the overall cohort. Significantly more patients succumb to relapsed or refractory disease than death from any other cause. We demonstrated that a Charlson comorbidity index score threshold of seven identifies exceptionally high‐risk patients, providing evidence for clinical use to reduce the risk of early treatment‐related mortality. Lastly, we provide evidence that measurable residual disease negativity and an IDH mutation predict a significant survival benefit outside clinical trials. Taken together, these data illuminate the real‐world performance of venetoclax and decitabine or azacitidine in the treatment of AML.

In the VIALE-A trial, venetoclax and azacitidine demonstrated significantly longer overall survival at 14.7 months compared to 9.6 months with azacitidine alone [10]. The use of venetoclax with a hypomethylating agent has since become routine in the treatment of AML. Despite this, outcomes outside clinical trials and analyzed with respect to the selection of the hypomethylating agent backbone remain unclear. Large population-based studies investigating azacitidine versus decitabine as monotherapy showed a small survival benefit for decitabine-treated patients. However, the significance of the survival benefit disappeared after analyzing patients that completed the intended schedule of chemotherapy [15].
The efficacy of different hypomethylating agent backbones in combination with venetoclax requires clarification. Retrospective analyses are conflicting, with no definitive differences in response rates between azacitidine-venetoclax and decitabine-venetoclax in the firstline setting [16]. In one retrospective study, the median overall survival significantly favored the azacitidine-venetoclax group at 12.3 months compared to 2.8 months with decitabine-venetoclax [17]. In contrast, other retrospective analyses showed a non-significant survival benefit favoring decitabine-venetoclax over azacitidine-venetoclax [18].
Moreover, treatment-related adverse events and results stratified by ELN cytogenetic risk outside clinical trials are lacking.

Objectives
The two primary objectives of this study were to retrospectively determine the composite complete remission rate and overall survival of patients with newly diagnosed AML treated with venetoclax with decitabine or azacitidine. The secondary objectives were to determine patient-and disease-related predictors of survival, assess the survival in patients that achieved a response negative for measurable residual disease (MRD), and characterize toxicities associated with venetoclax and decitabine or azacitidine.

Patient eligibility
The Institutional Review Board of Virginia Commonwealth Univer-

Treatment regimens
Patients were treated with venetoclax starting on day one of treatment and continuing until the end of the 28-day cycle or shorter duration, adjusted for toxicity or drug-drug interactions. Venetoclax was administered in 28-day cycles with decitabine 20 mg/m 2 in 5-or 10-day courses or azacitidine 75 mg/m 2 in 5-or 7-day courses. Venetoclax and decitabine or azacitidine were then administered as maintenance in 28-day cycles until intolerability, disease progression, or death, with cycle delays allowed for adverse events or count recovery.

Data collection and entry
We designed a REDCap instrument to retrospectively capture patient data [19]. The instrument was programmed to include cytogenetic and molecular profiles, response, and toxicity for each phase of treatment, including induction, maintenance, and relapse. Built-in score calculators and survival computation were programmed into the instrument during development to standardize data entry among investigators and minimize the likelihood of analytical errors.
The lead investigator reviewed the data set at two pre-specified time points and cross-checked entries for accuracy with the electronic medical record (Cerner Millennium and Epic). A minimum of two investigators standardized and cross-checked response and toxicity grading. Data discrepancies were resolved following a review by the lead investigator.

Safety analysis
Toxicities were graded using the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 [20,21]. Treatment-related adverse events were included if they occurred between the first dose and 28 days following treatment discontinuation. Quantitative toxicities were graded and recorded throughout each patient's treatment phase, excluding electrolyte aberrations. In instances where complete records were unavailable, toxicities were marked as unavailable for the phase of treatment to reduce bias.

2.6
Cytogenetic, molecular, and measurable residual disease analyses AML was defined using the fourth edition World Health Organization criteria, with a minimum of one bone marrow biopsy demonstrating at least 20% or greater myeloblasts [22]. The cytogenetic risk was defined as recommended by the European LeukemiaNet 2017 guidelines [23]. PCR assays obtained at diagnosis had a sensitivity of 10 −4 for NPM1 and 10 −2 for CEBPA, FLT3-ITD, and FLT3-TKD. Next-generation sequencing (NGS) was performed using an in-house NGS assay with a sensitivity of 2.7 × 10 −2 .
MRD negativity was defined using an assay at a minimum sensitivity threshold of 10 −3 , including PCR-based MRD assays and multiparameter flow cytometry (MFC; University of Washington Medical Center).
Mutations frequently associated with clonal hematopoiesis, including DNMT3A, TET2, and ASXL1, were not considered MRD if detected on a remission NGS assay [24]. Similarly, germline mutations, such as DDX41, GATA2, and TP53, were excluded as MRD [24]. MRD-negative results with suboptimal sample quality, as indicated in the result report, were excluded from MRD analysis.

Response assessment
Response assessments were performed in accordance with the modified International Working Group response criteria for AML [25].
Complete remission (CR) was defined as an absolute neutrophil count (ANC) of greater than 1000 cells/mm 3 , a platelet count of greater than 100,000 cells/mm 3 , transfusion independence, and a bone marrow biopsy with less than 5% blasts. CR with incomplete hematologic recovery (CRi) was defined as all the criteria for CR except for neutropenia (ANC ≤ 1000 cells/mm 3 ) or thrombocytopenia (platelets ≤ 100,000 cells/mm 3 ). CR with partial hematologic recovery (CRh) was defined as all the criteria for CR except for lower ANC (>500 cells/mm 3 ) and platelet (>50,000 cells/mm 3 ) thresholds. Progressive disease was defined as outlined by the European LeukemiaNet guidelines [23].
Composite complete remission (CRc) included patients that achieved CR, CRi, or CRh.

Statistical analysis
Patients treated between January 1, 2018, and January 1, 2022, were included in the study. The clinical data cutoff date was August 1, 2022, and patients alive at that time were censored. Means between groups were compared using the nonparametric Mann-Whitney test. Remission rates were reported with 95% confidence intervals using the Wilson method and compared between groups using Fisher's exact test. The overall survival was estimated for each cohort using the Kaplan-Meier method and compared using the log-rank test. The hazard ratio was estimated using the Cox proportional hazards model.
Cox proportional hazards assumptions were assessed using graphical methods and tested using Schoenfeld residual analysis with a level of significance of 0.01. All reported p-values were two-sided, with statistical significance evaluated at the 0.05 alpha level. Data analysis was performed with GraphPad Prism 9.4.1 for Macintosh.

Toxicity
The most common grade three or higher non-hematologic toxicities in the overall cohort were neutropenic fever (43.7%) and infection (40.8%). The most common bacterial infection was Enterococcus (17.2%), and the most common viral infection was severe acute respiratory syndrome coronavirus 2 (17.2%). Respiratory failure (16.9%), arrhythmia (7.0%), and hemorrhage (7.0%) were the subsequent most common grade three or higher adverse events. There were no significant differences in high-grade non-hematologic toxicities between the decitabine and azacitidine cohorts, presented in Table 2.   Table 3, and a heat map of molecular profiles among responders and non-responders is depicted in Figure 3.

Survival
The median overall survival for the entire cohort was 6.0 months, and the median duration of follow-up was 33.0 months, shown in Figure 4A.
In the decitabine-venetoclax cohort, the median overall survival was 8.3 months compared to 2.6 months in the azacitidine-venetoclax group, approaching statistical significance (p = 0.080, Figure 4B). The overall survival data are summarized in Table 4.
In the entire cohort, the median overall survival was 7. Next, we compared patients with a diagnosis of MDS preceding AML or AML-MRC with de novo AML. The median overall survival significantly favored the de novo cohort at 11.5 months compared to 6.0 months in the preceding MDS/AML-MRC group (p = 0.01, Figure 4C).
Patients exposed to a prior hypomethylating agent for preceding MDS or CMML had a median overall survival of 6.8 months compared to 6.0 months for those without previous hypomethylating agent exposure Patients with an ECOG score of 0 -1 had significantly prolonged survival at 8.6 months compared to patients with an ECOG score of 2 -3 at 4.6 months (p = 0.034, Figure 4E). Patients with CCI scores <7 had a median overall survival of 8.2 months compared to 2.1 months for scores ≥7 (p = 0.002, Figure 4F). Therefore, we identified a CCI score threshold of ≥7, which distinguished a cohort of patients with a significantly higher risk of death.

MRD status
Patients that achieved CR, CRi, and CRh were analyzed for survival with respect to MRD positivity. Patients that responded and were MRD negative had a significantly prolonged median overall survival of 17.7 months compared to 8.2 months for patients that were MRD positive (p = 0.01, Figure 4D).

DISCUSSION
We highlight several novel observations in the real-world setting, in addition to evidence supporting clinical trial findings for toxicity, response, and survival. We demonstrate that decitabinevenetoclax may result in significantly more severe thrombocytopenia than azacitidine-venetoclax, which may extend periods of transfusion dependence and cycle delays for count recovery. In contrast, the degree of severe lymphocytopenia was significantly lower in the decitabine-venetoclax cohort -implying more profound cytopenias may not universally be associated with decitabine-venetoclax across TA B L E 3 Response of patients treated with venetoclax and decitabine or azacitidine.

Azacitidine-venetoclax (N = 23) C Significance
Complete remission (CR) 3 (4.5) 2 (4.   Our subset analyses identified that IDH mut AML had significantly improved survival compared to IDH wild-type, suggesting venetoclax may be beneficial in specific molecular cohorts [27]. In contrast, signaling mutations, such as NRAS mut or KRAS mut , appeared to be associated with reduced survival. Our results suggest that a revised risk F I G U R E 3 Heat map of mutations in venetoclax-sensitive and venetoclax-resistant patients. Patients that achieved complete remission (CR), CR with incomplete hematologic recovery (CRi), or CR with partial hematologic recovery (CRh) are shown in the left panel. Patients that were refractory to venetoclax, died during treatment, or had an evaluable response that was no better than a morphological leukemia-free state are shown in the right panel. Due to the decreased survival outside of clinical trials, the selection of therapy candidates needs to be improved. We demonstrated that ECOG scores of 0 to 1 were associated with significantly improved survival compared to ECOG scores of 2 to 3. More strikingly, we identified that a CCI score threshold of seven differentiated patients is more likely to have favorable outcomes despite numerous comorbidities. This threshold may guide candidates for venetoclax-based therapy and reduce the rates of treatment-associated mortality and early death.
Overall, the combination of venetoclax and a hypomethylating agent is associated with lower remission rates and overall survival outside Josh Boron collected the data and revised the manuscript. Valerie