Efficacy of Azacitidine Plus Venetoclax in Acute Myeloid Leukemia Transformed from Myelodysplastic Syndrome after Failure of Azacitidine Monotherapy

Abstract

Objective

Patients with acute myeloid leukemia (AML) transformed from myelodysplastic syndrome (MDS) have a poor prognosis, including those treated with azacitidine during the MDS phase; there is no standard for the care of these patients. Recently, azacitidine plus venetoclax (AZA/VEN) was reported to prolong the survival in treatment-naïve AML patients compared with AZA monotherapy. However, the results of AZA/VEN for AML transformed from MDS, particularly after AZA monotherapy, remain unclear. The present study therefore compared the clinical results of AZA/VEN treatment in these patients.

Methods

Data from MDS patients diagnosed at 10 institutions in Nagasaki Prefecture were collected. Thereafter, patients with transformed AML following AZA monotherapy during the MDS phase were selected, and their treatment response and survival were analyzed.

Results

The overall response (OR) rate, overall survival (OS), and event-free survival (EFS) were compared among patients treated with AZA/VEN (n=13), chemotherapy (intensive and low-intensity, n=35), AZA monotherapy (mAZA, n=15), and best supportive care (BSC, n=43) after AML transformation. The corresponding OR rates were 38.5%, 20.0%, and 6.7% for the AZA/VEN, chemotherapy, and mAZA groups, respectively (p=0.235). The respective median OS and EFS were 10.7 and 8.9 months for AZA/VEN, 3.2 and 2.0 months for chemotherapy, and 3.8 and 2.7 months for mAZA, and 1.7 months for BSC (OS only) (p=0.000023 for the OS and p=0.026 for the EFS),

Conclusion

Our findings suggest the superiority of AZA/VEN for AML patients with transformation from MDS following AZA monotherapy.

Introduction

Myelodysplastic syndrome (MDS) is a clonal hematopoietic stem cell disorder characterized by dysplastic morphology, cytopenia due to ineffective hematopoiesis, and a tendency for leukemia transformation (1). High-risk MDS patients (HR-MDS), usually evaluated using the International Prognostic Scoring System (IPSS) (2) and its revised version (IPSS-R) (3), tend to develop acute myeloid leukemia (AML). Most guidelines recommend allogeneic hematopoietic stem cell transplantation (allo-SCT) as the first-line treatment for HR-MDS, if possible, to cure MDS (1,4-6).

Unfortunately, most HR-MDS patients are not eligible for allo-SCT, as the majority of these patients are elderly. Instead, they are treated with hypomethylating agents, such as azacitidine (AZA), which has been shown to prolong the survival, although a cure for MDS is not expected (7,8). Ultimately, many HR-MDS patients (40-50%) experience AML transformation after AZA treatment, and this AML is AZA-resistant if they have a history of AZA exposure (9,10). Currently, there is no standard treatment for AML transformed from MDS, and the survival rate is very low (11-13).

The current VIALE-A clinical trial evaluated the efficacy of venetoclax (VEN) in combination with AZA in previously untreated AML patients ineligible for intensive chemotherapy and demonstrated the significant superiority of AZA plus VEN (AZA/VEN) versus AZA monotherapy in terms of the survival (14). Based on the results of this study, AZA/VEN has been approved as a treatment for AML patients ineligible for intensive care treatment in Japan. The VIALE-A trial included secondary AML patients with a history of MDS, demonstrating the efficacy of AZA/VEN. However, patients with secondary AML who received AZA monotherapy during the MDS phase were not included, and there is no clear evidence that this combination is useful for these patients.

In light of these circumstances, we conducted a retrospective analysis of MDS patients treated at various facilities in the Japanese prefecture of Nagasaki to examine the effectiveness of various treatments for AML that progressed from MDS, specifically for those who were exposed to AZA during the MDS phase. This study evaluated the clinical effectiveness of adding VEN to AZA treatment in these patients.

Materials and Methods

1 Study design and patients

This was a retrospective study of patients ≥16 years old who were first diagnosed with MDS between January 2000 and December 2022 at 10 institutions in Nagasaki Prefecture (listed in the Acknowledgment). This involved creating a Nagasaki-MDS database and then analyzing treatment results for MDS-transformed AML using this database. The main purpose of this study was to investigate the effectiveness of AZA/VEN treatment in patients with transformed AML (from MDS), particularly in those who received AZA therapy during the MDS phase.

This study was conducted in accordance with the ethical standards of the Declaration of the World Medical Association Helsinki and was approved by the Institutional Review Board of Nagasaki University Hospital (approval no. 19090906-5), and the corresponding committee of each participating institute.

The diagnosis and classification of MDS and AML were based on the revised 4th edition of the World Health Organization (WHO) classification (WHO 2016) (15). Patients with de novo MDS, therapy-related MDS, and secondary MDS with a history of aplastic anemia were included in this database; however, patients with myelodysplastic/myeloproliferative neoplasms, including chronic myelomonocytic leukemia, were excluded.

Management after AML transformation was divided into six groups: (1) AZA/VEN, (2) conventional chemotherapy [C-Chemo; intensive chemotherapy, low-dose cytosine arabinoside (LDAC), or LDAC plus aclarubicin], (3) LDAC plus VEN (LDAC/VEN), (4) clinical trial, (5) AZA monotherapy (mAZA), and (6) best supportive care (BSC).

The data were updated at the end of January 2024. Clinical data for each patient, including the diagnosis, treatment, leukemia transformation, and prognosis, were collected from the medical records of each hospital.

2 Response criteria and outcome definitions

The best response of transformed AML to treatment was evaluated according to the European Leukemia Net consensus guidelines (16). The overall response (OR) was defined as a combination of complete remission (CR), CR with incomplete hematologic recovery (CRi), and partial remission. Relapse was defined as an increase in bone marrow blasts of ＞5% or the occurrence of extramedullary AML in patients who achieved CR or CRi. The overall survival (OS) was defined in this study as the period from the start of AML treatment (for groups other than BSC), or the date of the AML diagnosis (for BSC groups) to the date of death or last follow-up. The event-free survival (EFS) was defined as the period from the start of AML treatment to the date of disease progression, treatment failure (discontinuation or change of treatment due to failure to achieve OR), death, or last follow-up. The relapse-free survival (RFS) was defined as the period from the achievement of OR to relapse, death, or the last follow-up.

3 Statistical analyses

Clinical covariates were compared using the Kruskal-Wallis rank-sum test for continuous variables and Fisher's exact test for categorical variables. The OS, EFS, and RFS were estimated using the Kaplan-Meier method and compared between the groups to detect significant differences using the log-rank test. Statistical significance was set at p＜0.05. All statistical analyses were performed using the EZR software program, version 1.64 (17).

Results

1 Analysis cohort

At 10 institutions in Nagasaki Prefecture, 1,588 patients were newly diagnosed with MDS between January 2000 and December 2022 using WHO2016, creating the Nagasaki-MDS database. First, we excluded patients who received allo-SCT (n=79), those who did not receive AZA (n=1,175), and those who received AZA monotherapy or AZA/VEN as their first treatment following AML transformation, with no treatment during the MDS phase (n=22), to assess the efficacy of treatment for transformed AML following AZA administration during the MDS phase (Fig. 1). The remaining 312 patients were treated with AZA for MDS, and 17 were excluded due to participation in a clinical trial or prior treatment with AZA/VEN or LDAC/VEN before AML transformation. Therefore, 295 patients were treated with AZA monotherapy for MDS, and 113 ultimately underwent transformation to AML (Fig. 1).

Figure 1.

Patient selection flowchart. MDS: myelodysplastic syndrome, allo-SCT: allogeneic hematopoietic stem cell transplantation, AZA: azacitidine, VEN: venetoclax, AML: acute myeloid leukemia

Following AML transformation, we observed the first course of treatment in 113 patients. Forty-three and 15 patients received BSC and mAZA therapy, respectively. The remaining 55 patients received interventional treatment for AML, such as AZA/VEN (n=13), C-Chemo (n=35), LDAC/VEN (n=2), and clinical trials (n=5) (Fig. 2). Because the number of patients treated with LDAC/VEN was small (n=2), we excluded them from further analyses. Patients who participated in the clinical trials (n=5) were also excluded.

Figure 2.

Initial treatment strategy after AML transformation. After AML transformation, 13 patients were treated with AZA/VEN, 35 with conventional chemotherapy, 2 with LDAC/VEN, and 5 in a clinical trial. Meanwhile, 15 patients received AZA monotherapy, and 43 patients received BSC. AML: acute myeloid leukemia, AZA: azacitidine, VEN: venetoclax, LDAC: low-dose cytosine arabinoside, BSC: best supportive care

The patient characteristics of the three remaining treatment groups are shown in Table 1. Of note, the BSC group was excluded from this table because sufficient examinations, such as bone marrow aspiration, were often not performed at the time of AML transformation. The age at AML transformation was significantly different among the groups (p=0.011), with the C-Chemo group being the youngest (median age 72.0 years old), followed by the AZA/VEN (median age 75.0 years old), and mAZA groups (median age 77.0 years old). The distribution of MDS subtypes at the diagnosis according to WHO 2016 differed marginally among the 3 groups (p=0.056); MDS other than excess blasts (e.g. MDS with single lineage dysplasia, MDS with multilineage dysplasia, and MDS with ring sideroblasts and single lineage dysplasia) was more common in the mAZA group (40.0%) than in the AZA/VEN and C-Chemo groups (7.7% and 22.8%, respectively). Owing to the different MDS subtypes, the proportion of blasts in the bone marrow also differed slightly (p=0.053). Cytogenetic risk at transformation differed among the 3 groups, with poor-risk karyotypes being found less frequently in the mAZA group (6.7%) than in the AZA/VEN and C-Chemo groups (38.5% and 45.7%, respectively, p=0.056). Of note, all patients in the AZA/VEN group received AZA for seven days per cycle, whereas the duration of VEN administration was determined by the physician's choice.

Table 1.

Comparison of Patient Characteristics between the Three Treatment Groups.

Parameter	AZA/VEN	Conventional chemotherapy	AZA monotherapy	p value
No. of patients	13	35	15
Age at AML transformation, median (range)	75.0 (65-89)	72.0 (55-79)	77.0 (63-84)	0.011a
Age ≥75 years at AML transformation
Yes, n (%)	7 (53.8)	11 (32.4)	10 (66.7)	0.055b
No, n (%)	6 (46.2)	24 (67.6)	5 (33.3)
Sex
Male, n (%)	7 (53.8)	21 (60.0)	11 (73.3)	0.566b
Female, n (%)	6 (46.2)	14 (40.0)	4 (26.7)
Therapy-related MDS
Yes, n (%)	0	5 (14.3)	1 (6.7)	0.453b
No, n (%)	13 (100.0)	30 (85.7)	14 (93.3)
Classification by WHO2016 at MDS diagnosis
MDS-SLD, n (%)	1 (7.7)	0	0	0.056b
MDS-MLD, n (%)	0	6 (17.1)	6 (40.0)
MDS-RS-SLD, n (%)	0	2 (5.7)	0
MDS-EB-1, n (%)	4 (30.8)	13 (37.1)	2 (13.3)
MDS-EB-2, n (%)	7 (53.8)	14 (40.0)	7 (46.7)
MDS-U with 1% blood blasts, n (%)	1 (7.7)	0	0
Complex karyotype (≥3 clonal abnormalities) at MDS diagnosis
Yes, n (%)	3 (23.1)	15 (42.9)	2 (13.3)	0.107b
No, n (%)	10 (76.9)	20 (57.1)	13 (86.7)
Complex karyotype (≥3 clonal abnormalities) at AML transformation
Yes, n (%)	5 (38.5)	14 (40.0)	1 (6.7)	0.211b
No, n (%)	7 (53.8)	15 (42.9)	9 (60.0)
Not evaluated, n (%)	1 (7.7)	6 (17.1)	5 (33.3)
Courses of AZA before AML transformation, median (range)	12 (1-33)	8 (1-23)	5 (1-18)	0.099a
Percentage of blasts in bone marrow at AML transformation†
≥20 to <30%, n (%)	8 (61.5)	5 (14.3)	7 (46.7)	0.053b
≥30 to <50%, n (%)	2 (15.4)	12 (34.3)	3 (33.3)
≥50%, n (%)	1 (7.7)	8 (22.9)	1 (6.7)
Not evaluated, n (%)	2 (15.4)	10 (28.6)	4 (26.7)
Cytogenetic risk at AML transformation by NCCN guidelines
Intermediate risk karyotype, n (%)	7 (53.8)	13 (37.1)	10 (66.7)	0.056b
Poor risk karyotype, n (%)	5 (38.5)	16 (45.7)	1 (6.7)
7 or 7q deletion	0	2	0
Complex (≥3 clonal abnormalities)	5	14	1
Not evaluated, n (%)	1 (7.7)	6 (17.1)	4 (26.7)
Regimens of conventional chemotherapy
Intensive chemotherapy, n (%)		9 (25.7)
LDAC, n (%)		7 (20.0)
LDAC plus aclarubicin, n (%)		19 (54.3)

AZA: azacitidine, VEN: venetoclax, AML: acute myeloid leukemia, MDS: myelodysplastic syndromes, WHO: World Health Organization, MDS-SLD: myelodysplastic syndrome with single lineage dysplasia, MDS-MLD: myelodysplastic syndrome with multilineage dysplasia, MDS-RS-SLD: myelodysplastic syndrome with ring sideroblasts and single lineage dysplasia, MDS-EB-1: myelodysplastic syndrome with excess blasts 1, MDS-EB2: myelodysplastic syndrome with excess blasts 2, MDS-U: myelodysplastic syndrome, unclassifiable, NCCN: National Comprehensive Cancer Network, LDAC: low-dose cytosine arabinoside

^aCalculated using Kruskal-Wallis rank sum test.

^bCalculated using Fisher’s exact test.

^†16 patients who were diagnosed with AML transformation based only on the percentage of blasts in peripheral blood.

2 Difference in initial treatment strategy after approval of VEN for AML patients

The co-administration of VEN with AZA or LDAC was approved in Japan in March 2021 for the treatment of AML patients. Prior to this approval, 87 MDS patients underwent transformation to AML, and 33 of the 87 patients (37.9%) received C-Chemo as the first-line treatment for transformed AML (Fig. 3A). In contrast, after VEN became available for AML, 13 of the 26 patients with transformation (50.0%) were treated with AZA/VEN as the first-line treatment for AML, and only 2 patients (7.7%) were treated with C-Chemo (Fig. 3B). AZA monotherapy was chosen as the treatment for AML before the approval of VEN for AML. Most patients continued mAZA reluctantly because of their inability to tolerate C-Chemo. However, only a few patients who discontinued AZA for a period of time included those who attempted retreatment. An analysis categorized by age showed that the proportion of patients ≥75 years old increased following the approval of VEN. Before the approval of VEN, 44.9% (22 of 49) of patients ＜75 years old at AML transformation did not receive active treatment, including AZA/VEN, LDAC/VEN, C-Chemo, or clinical trials. However, after the approval of VEN, 88.9% (8 of 9) received active treatment. Among patients ≥75 years old with AML transformation, only 28.9% (11 of 38) received active treatment before the approval of VEN, whereas after its approval, 52.9% (9 of 17) received active treatment (Supplementary material 1).

Figure 3.

Changes in initial treatment strategy following approval of VEN. (A) Before VEN became available for AML, 33 of the 87 patients (37.9%) received conventional chemotherapy. (B) In contrast, after VEN became available for AML, 13 of 26 patients (50.0%) were treated with AZA/VEN, and only 2 patients (7.7%) were treated with conventional chemotherapy. AZA monotherapy was selected only before the approval of VEN for AML. AML: acute myeloid leukemia, VEN: venetoclax, AZA: azacitidine, LDAC: low-dose cytosine arabinoside, BSC: best supportive care

4 Response to treatment for AML and the survival by treatment group

The effectiveness of treatment of MDS-transformed AML after AZA exposure was analyzed in three treatment groups: AZA/VEN (n=13), C-Chemo (n=35, 33 patients before and 2 after approval of VEN for AML), and mAZA groups (n=15). The OR rates were 38.5%, 20.0%, and 6.7% in the AZA/VEN, C-Chemo, and mAZA groups, respectively, with no statistically significant difference (p=0.235, Table 2). CR was observed in 38.5% and 17.1% of the patients in the AZA/VEN and C-Chemo groups, respectively, but not in the mAZA group. In the AZA/VEN group, among the patients whose best response to mAZA in the MDS phase was CR, three out of eight achieved CR with AZA/VEN. In contrast, among those who did not observe CR to mAZA in the MDS phase, two out of five achieved CR with AZA/VEN.

Table 2.

Best Treatment Response by Treatment Groups.

Best treatment response	AZA/VEN	Conventional chemotherapy	AZA monotherapy	p value
OR, n (%)	5 (38.5)	7 (20.0)	1 (6.7)	0.235a
CR, n (%)	5 (38.5)	6 (17.1)	0
CRi, n (%)	0	1 (2.9)	1 (6.7)
PR, n (%)	0	0	0
Not response, n (%)	8 (61.5)	26 (74.3)	14 (93.3)
Not evaluated, n (%)	0	2 (5.7)	0

AZA: azacitidine, VEN: venetoclax, OR: overall response, CR: complete remission, CRi: CR with incomplete hematologic recovery, PR: partial response

^aCalculated using Fisher’s exact test.

The OS curves for the three treatment groups, including the BSC group, were calculated and are shown in Fig. 4. Median survival times differed significantly among groups: 10.7 [95% confidence interval (CI), 2.9-20.5 months], 3.2 (95% CI, 1.9-3.9), 3.8 (95% CI, 2.3-5.8), and 1.7 months (95% CI, 0.7-2.2) for AZA/VEN, C-Chemo, mAZA, and BSC groups, respectively (Fig. 4, p=0.000023). The median EFS for the AZA/VEN, C-Chemo, and mAZA groups was 8.9 (95% CI, 1.7-17.4), 2.0 (95% CI, 1.5-2.4), and 2.7 months (95% CI, 1.3-4.3), respectively. These differences were also significant among the three treatment groups (Fig. 4B, p=0.026). Regarding both the OS and EFS, the treatment results with C-Chemo were similar to those with mAZA. No statistically significant differences in the OR, OS, or EFS were observed across chemotherapy regimens in the C-Chemo group (Supplementary material 2, 3). No statistically significant differences in OR rate or EFS were observed between patients ＜75 and those ≥75 years old at AML transformation (Supplementary material 4, 5). In patients with a poor-risk karyotype for AML transformation, the EFS also seemed to be superior in the AZA/VEN group (Supplementary material 6). Although the number of patients who achieved CR was small, the RFS among the treatment groups was significantly different (Fig. 4C, p=0.010).

Figure 4.

(A) The overall survival by treatment group. The survival curves differed significantly among the 4 groups (p=0.000023). (B) The event-free survival by treatment group. The survival curves differed significantly among the 3 groups (p=0.026). (C) The relapse-free survival by treatment group. The survival curves differed significantly among the 3 groups (p=0.010). OS: overall survival, AZA: azacitidine, VEN: venetoclax, BSC: best supportive care, EFS: event-free survival, RFS: relapse-free survival

Discussion

The efficacy of AZA/VEN in AML patients transformed from MDS with a history of AZA exposure remains unclear. In this study, we demonstrated that AZA/VEN provided better results than chemotherapy or continuation of AZA monotherapy with regard to the OS, EFS, and RFS, although the number of patients was small. The survival rate of patients with AZA-resistant MDS has been reported to be dismal regardless of primary or secondary resistance, including transformation to AML (11-13). Notably, this poor prognosis appears to be unaffected by the intensity of chemotherapy administered after AML transformation (11). In this study, a comparable pattern emerged; however, the small sample size in each group limited the ability to analyze patient backgrounds. Consequently, the potential impact of clinical judgment on the composition of patient groups cannot be dismissed, which makes it challenging to reach definitive conclusions.

After VEN became available for AML treatment in combination with AZA or LDAC, these became the choice for these patients for the following two reasons. Because mild adverse events were expected, patients who had previously limited treatment options, such as mAZA or BSC, would have alternative treatment options. Furthermore, these treatments are expected to provide survival benefits rather than poor outcomes with C-Chemo. The VIALE-A trial included patients who were ineligible for intensive chemotherapy because of coexisting conditions, ≥75 years old, or both. However, according to real-world data, AZA/VEN is used in patients ＜75 years old (18). In our study, approximately half of the patients in the AZA/VEN group were ＜75 years old (Table 1). The use of VEN combination regimens increased in patients both ＜75 and ≥75 years old, resulting in very few patients receiving C-Chemo as an active treatment (Supplementary material 1).

However, several concerns have been raised concerning the efficacy of AZA/VEN and LDAC/VEN in transformed AML. AML from MDS was thought to result in AZA resistance following AZA exposure, and no strong anti-leukemia effect was observed with VEN monotherapy, as shown in a phase 2 study (19). A previous in vitro study reported an increase in BCL-2 and decrease in MCL-1 following hypomethylating agent treatment (20). These reports may have enhanced the use of AZA/VEN, even with AZA resistance. Our study also included cases in which AZA/VEN was effective in patients who responded to AZA during the MDS phase, as well as those who did not. The number of cases was limited, and no clear trend was observed in relation to the clinical efficacy of AZA for prior MDS and the subsequent efficacy of the AZA/VEN combination in conditions that had progressed to AML. The results of this study suggest that the combination of AZA and VEN has clinical efficacy in AML patients after MDS, even with a history of AZA treatment. Fewer events, including relapses, in the AZA/VEN group suggested the presence of an anti-leukemic effect (Fig. 4B, C). Considering that AML in this situation had already acquired AZA resistance, the AZA/VEN results could not be explained by a simple synergistic effect of AZA and VEN. It would be interesting to further analyze the mechanism of action of these agents in combination.

In this study, we found that, in the Nagasaki cohort, conservative therapy, including mAZA and BSC, was chosen as treatment for 51.4% of patients with AML after AZA treatment in the MDS phase (Fig. 2). The proportion receiving conservative therapy decreased after the approval of VEN for AML (Fig. 3), but 34.6% of these patients received BSC, suggesting a difficult situation for these patients for active treatment after acquiring AZA resistance. This might be attributed to the increased proportion of elderly patients following the approval of VEN. However, after the approval of VEN, even among patients ≥75 years old, approximately half were able to select active treatments, such as AZA/VEN or C-Chemo (Supplementary material 1). A comparison of the response to AZA/VEN or C-Chemo was made between patients ≥75 and ＜75 years old, but the difference was not clear, due in part to the limited number of cases in each group.

MDS is a disease of the elderly, and this could have implications for the choice of treatment and management, particularly for diseases. We assume that AZA/VEN is a more popular treatment for AML transformed from AZA-treated MDS than is conventional chemotherapy. As the combination of AZA and VEN is being evaluated for higher-risk MDS, the situation will change if this combination is shown to be effective.

Several limitations associated with the present study warrant mention. This was a retrospective analysis of a small number of patients in each group, which made it difficult to perform multivariate, propensity score matching, and stratified analyses. Therefore, confounding factors such as disease status were not adjusted for. This is a principal limitation of our study. An important objective for future research is to include a larger cohort of patients treated with AZA/VEN who have clinical characteristics similar to those in our study to improve the validity of the findings.

Author’s disclosure of potential Conflicts of Interest (COI).

Y. Miyazaki received honoraria from AbbVie, Astellas Pharma, Chugai Pharmaceutical, Novartis, Bristol Myers Squibb, and Daiichi-Sankyo. The other authors have no conflicts of interest to declare.

Supplementary Material

Changes in initial treatment strategy following approval of VEN by age at AML transformation

Abbreviations: AZA, azacitidine; VEN, venetoclax; LDAC, low-dose cytosine arabinoside; AML, acute myeloid leukemia; BSC, best supportive care.

Best treatment response by chemotherapy regimens in conventional chemotherapy group

Abbreviations: LDAC, low-dose cytosine arabinoside; OR, overall response; CR, complete remission; CRi, CR with incomplete hematologic recovery; PR, partial response. a Calculated using Fisher's exact test.

Overall survival (A) and event-free survival (B) according to chemotherapy regimens in conventional chemotherapy group

The survival curves did not differ significantly between the three groups.OS, overall survival; EFS, event-free survival; LDAC, low-dose cytosine arabinoside.

Best treatment response by treatment groups and age at AML transformation

Abbreviations: AZA, azacitidine; VEN, venetoclax; OR, overall response; CR, complete remission; CRi, CR with incomplete hematologic recovery; PR, partial response. a Calculated using Fisher's exact test.

The event-free survival among patients <75 (A) and ≥75 years old (B) at AML transformation.

The survival curves did not differ significantly among the three groups. EFS, event-free survival; AZA, azacitidine; VEN, venetoclax.

The event-free survival by treatment group in patients with a poor-risk karyotype at AML transformation according to NCCN guidelines.

The survival curves differed significantly among the three groups (P=0.016). EFS, event-free survival; AZA, azacitidine; VEN, venetoclax; NCCN, National Comprehensive Cancer Network.

Funding Statement

This study was supported by the Japan Society for the Promotion of Science KAKENHI under Grant JP22K08481.