The efficacy and safety of venetoclax and azacytidine combination treatment in patients with acute myeloid leukemia and myelodysplastic syndrome: systematic review and meta-analysis

ABSTRACT

Objectives:

The meta-analysis sought to evaluate the efficacy and safety of a combination of venetoclax (Ven) and azacitidine (AZA) in the treatment of acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS).

Methods

We searched PubMed, Excerpta Medica Database (EMBASE), Cochrane Library, and Web of Science for eligible studies from inception to June 2022. We used the Cochrane Risk of Bias 2.0 (RoB 2.0) and Methodological Index for Non-Randomized Studies (MINORS) to evaluate the quality of the included literature. The inverse variance method was used to calculate the pooled proportion and 95% confidence interval (CI).

Results

The meta-analysis included nineteen studies with a total of 1615 patients. The pooled overall CR/CRi (complete response (CR)/complete response with incomplete blood count recovery (CRi)) rate for AML and MDS was 57.9% (95% CI 49.5-65.9%, I² = 83%). Subgroup analyses showed that the rate of pooled CR/CRi was 67.5% (95% CI 61.1-73.3%, I² = 54%) for the new-diagnosed (ND) AML group, 30% (95% CI 20-44.1%, I² = 66%) for relapsed/refractory (R/R) AML, and 67.6% (95% CI 52.6-79.8%, I² = 65%) for MDS, respectively. One randomized controlled trial (RCT) showed that CR/CRi was 64.7% in ND-AML patients. A total of 9 studies reported adverse events, with neutropenia being the most common of grade 3–4 adverse events, with a rate of 53.7% (95% CI 61.1-73.3%, I² = 54%).

Conclusion

The present meta-analysis demonstrated that the Ven + AZA regimen is efficacious for the treatment of AML and MDS, with it being more effective for ND-AML than R/R AML. The most common adverse effects of this regimen are grade 3–4 neutropenia and neutropenia with fever.

KEYWORDS:

• Meta-analysis
• acute myeloid leukemia
• myelodysplastic syndrome
• venetoclax
• azacitidine

Introduction

Acute myeloid leukemia (AML) is a heterogeneous group of hematopoietic malignancies characterized by a decrease in normal blood cells due to blocked proliferation and apoptosis of primitive cell clones, with the median age of diagnosis being 68 years [1,2]. Older patients are often unable to tolerate intensive chemotherapy due to having more comorbidities and lower physical status at the time of diagnosis. Even though a few elderly patients are evaluated as fit patients (eligible for intense chemotherapy) by the geriatric assessment (GA) tools, owing to the higher incidence of adverse risk cytogenetic abnormalities in older patients with AML, which leads to lower rates of complete remission and long-term survival with intense chemotherapy [3,4]. Myelodysplastic syndromes (MDS) are a heterogeneous group of hematopoietic stem/progenitor cell-derived clonal disorders with a high risk of conversion to AML, where the median age of diagnosis is approximately 70 years [5,6]. The primary therapy for MDS, especially for high-risk MDS, is hypomethylating agents, but the overall results are poor, with a 5-year survival rate of approximately 31% [7]. In recent years, there has been a greater understanding of the molecular mechanisms of AML/MDS, which has promoted the application of targeted agents such as: FLT3 inhibitors, IDH inhibitors, and selective BCL-2 inhibitors in elderly AML/MDS patients, particularly selective BCL-2 inhibitors have shown promising results in AML/MDS clinical studies [8–10].

Anti-apoptotic BCL-2 family proteins are widely expressed in myeloid malignancies. Venetoclax (Ven) is a BH3 mimetic that selectively binds to BCL-2 and activates cell apoptosis via mitochondrial outer membrane permeability [11]. A combination of Ven and azacitidine (AZA) is recommended for the treatment of elderly or unfit AML patients. In recent years, several retrospective analyses or clinical trials have suggested the efficacy of Ven + AZA in AML treatment. However, the evidence from these studies was limited because the efficacy varied widely among studies and some studies were small in size [12–14]. Given the efficacy of Ven + AZA in AML, the investigation of this combination has been promoted in MDS. A phase Ib clinical trial of Ven + AZA for de novo high-risk MDS showed an effective response time of 30 days and an overall response rate of 74% [15]. However, the studies on Ven + AZA for MDS that have been reported so far are small-scale and exploratory.

Therefore, the objective of this study is to conduct a systematic review and meta-analysis to evaluate the efficacy and adverse events of Ven + AZA in patients with AML/MDS.

Materials and methods

Data sources and literature search strategy

This study has been conducted in the International Prospective System for Registration Review (Registration No. CRD42022347624). Two researchers jointly participated in the research process. Relevant articles published before June 2022 were searched in Pubmed, Cochrane, Web of Science, and EMBASE respectively. The keywords were ‘venetoclax or GDC-0199 or RG7601 or Venclexta or ABT-199’, ‘Azacitidine or Azacytidine or Vidaza or NSC102816 or 5 azacytidine’ ‘leukemia, myeloid, acute or leukemia or acute myeloid leukemia or ANLL or AML’, ‘Myelodysplastic syndrome or MDS’. The full search strategy is shown in Table S1.

Inclusion and exclusion criteria

Inclusion criteria: (1) Participants: Adult patients with AML and MDS treated with Ven + AZA (including newly diagnosed, refractory/relapsed patients); (2) Intervention: Ven + AZA; (3) Comparator: Intensive chemotherapy group, reduced-intensity chemotherapy group, supportive therapy group or no control group; (4) Outcome: The study provides the primary outcome on CR/CRi rates, secondary outcomes on ORR and median OS, and the rate of 3–4 grade hematologic adverse events; (5) Study design: Cohort study, randomized controlled trials (RCTs), or non-randomized controlled trials (NRCTs).

Exclusion criteria: (1) Case reports, reviews, meta-analyses, comments, conference abstracts, and non-human studies. (2) Treatment regimens that included drugs other than Ven in combination with AZA. (3) The study results did not include one of the primary outcome measures. (4) Original data could not be obtained. (5) Duplicate publications.

Two investigators (YFD and CHL) independently read the title, abstract and full text to decide whether the literature was included. Any discrepancy in the inclusion literature was resolved by consensus.

Definition of treatment response

CR/CRi consisted of complete response (CR) and CR with incomplete blood count recovery (CRi). ORR included CR/CRi, morphologic leukemia-free status (MLFS), and partial remission (PR). The criteria for response refer to the 2003 International Working Group, 2017 European Leukemia Net, and the International Response Unified Criteria revised by the MDS International Working Group in 2006 [16–18]. Overall Survival (OS) was defined as the time between treatment initiation and death or was censored at the last follow-up.

Adverse events of different grades: Adverse events were graded according to the National Cancer Institute Common Terminology Standard for Adverse Events, VERSION 4.03 [19].

Article quality assessment

In this meta-analysis, the risk of bias was assessed with the RoB 2.0 for RCTs, and NRCTs were assessed using the methodological index for non-randomized studies (MINORS) guidelines [20,21]. MINORS has 12 items, the first 8 were related to non-comparative studies, while the remaining 12 items were related to comparative studies. The items are scored as 0 (not reported), 1 (reported but inadequate), or 2 (reported and adequate). The assessment of methodological quality is presented in Figure S1 and Table S2. Two investigators (YFD and CHL) independently evaluated each study for risk bias, and any discrepancy was resolved through discussion.

Data extraction and statistical analysis

Two investigators (YFD and CHL) independently collected the data using a pre-designed format. Discrepancies were resolved through consensus. Review Manager (version 5.4.1) was used for this meta-analysis. For dichotomous data CR/CRi, ORR, and adverse events, the generic inverse variance was used for dichotomous variable analysis without a control group, and a 95% confidence interval (CI) was reported for each measurement. This method uses the calculation method of ratio type data, where the odds ratio (OR) value calculated by Revman software is converted into the following method to obtain the final result: 95% CI: Pf = OR/ (1 + OR); 95% CI lower limit conversion: LL_OR= LL_OR/(1 + LL_OR); 95% CI upper limit conversion: UL_OR= UL_OR/(1 + UL_OR). Q test and I² test (I²≤ 25%: no heterogeneity; 26-50%: low heterogeneity; 51-75%: intermediate heterogeneity; >75%: high heterogeneity) for heterogeneity analysis. For p <0.1 or I² >50%, there was statistical heterogeneity among the references, and the random effects model was used. When p > 0.1 or I² < 50%, the fixed effects model was used for analysis [22]. To determine the causes of heterogeneity, subgroup and sensitivity analysis were conducted on the results with high heterogeneity. Sensitivity analysis is performed by sequentially removing individual studies to assess the impact of each study on the statistical results (p <0.05 was considered statistically significant).

Results

Literature search results

A total of 2070 articles were retrieved. From these results, 647 duplicates and 1335 irrelevant papers were excluded. After reading the full text of the remaining 88 papers, 69 were excluded because of unavailable data or a lack of primary outcome. Finally, 1 RCT and 18 NRCTs met the inclusion criteria for this meta-analysis. Using the Revman software, a flow chart of literature screening was prepared (Figure 1). A total of 1477 patients with AML and 138 patients with MDS were included in the study.

Figure 1. PRISMA flow chart of study selection process.

Characteristics of included studies

A total of 1 RCT and 18 NRCTs involving 1615 patients were eligible for this meta-analysis [23–41], as shown in Table 1. The included studies were published between 2017 and 2022. The sample size of the trials ranged from 6 to 439 patients (median number 33) with ages ranging between 19 and 91 years, and a male proportion of 59%. Eleven of these studies originated from North America, five from Asia, and two from Europe. The last study was a multicenter RCT involving multiple countries. The use of this regimen (Ven + AZA) refers to Table 2. Seven studies published the number of cycles (from 1 to 12) of the Ven + AZA therapy.

Table 1. Clinical characteristics of patients included in the studies.

Study	Disease	Patients (numbers)	Median Age (years)	median follow up time(months)	Efficacy			Adverse effects (Grade 3/4)				Early death ≤60days (%)
					CR/CRi	ORR	Median os months	Anemia (%)	Thrombocytopenia (%)	Neutropenia (%)	Febrile neutropenia (%)
Pollyea DA et al. [23]	ND AML	84	75 (61-90)	29(0.4-42)	71%	N	16.4(11.3-24.5)	30%	25%	20%	39%	N
DiNardo CD et al. [24]	ND AML	29	74(65-86)	15.1(9.8-31.7)	76%	83%	NR (9.0-NR)	31%	N	N	38%	2(8%)
DiNardo CD et al. [25]	ND AML	431	76 (49–91)	20.5(0.1-30.7)	64.7%	66.4%	14.1(10.7-19.3)	26%	45%	42%	42%	N
Cherry EM et al. [26]	ND AML	143	69.5 (22-91)	N	71.3%	76.9%	16.1(11.2-21.3)	N	N	N	N	N
Byrne M et al. [27]	R/R AML	12	64(34-73)	N	42%	58%	4.5(0.4-12.5)	N	N	N	N	N
Labrador J et al. [28]	R/R AML	30	67 (41–76)	5.5(0.7-10.4)	17.9%	32.1%	4(2.6-5.8)	N	N	N	N	N
Garcia JS et al. [29]	High-Risk MDS	57	71(26-85)	13(11.3-15.6)	77%	77%	NR (16.2-NR)	N	30%	51%	46%	N
DiNardo CD et al. [30]	ND AML	22	75(71-80)	12.7(9.0-15.8)	59%	82%	14.2(9.3-NR)	5%	59%	36%	45%	14%
Matthews AH et al. [31]	ND AML	439	75(36-88)	8.8(N)	65%	N	11(N)	N	N	N	N	13%
Ucar MA et al. [32]	R/R AML	35	65(19-85)	N	20%	51.4%	9.13(7.65-10.62)	N	N	N	N	N
Zeidan AM et al. [33]	R/R MDS	24	76(44-91)	N	50%	50%	N	0	43%	57%	N	N
Mirgh S et al. [34]	ND AML R/R AML	ND:16 R/R:8	60(30-69)	8(N)	ND:75% R/R:12.5%	ND81.2% R/R: 12.5%	8 (2-26)	49.2%	38%	70%	33%	8%
Taniguchi S et al. [35]	ND AML R/R AML	ND:5 R/R:1	75 (66–80)	11.4(1.9–29.7)	ND:100% R/R:0	ND:100% R/R:0	15.7 (6.2-NR)	N	50%	50%	67%	N
Pollyea DA et al. [36]	ND AML	33	75(65-89)	11.7(4.9–15.6)	85%	91%	NR	N	N	N	N	N
Winters AC et al. [37]	ND AML	30	72(33-85)	21.9(1.2-37.7)	63.3%	73.3%	29.3 (12.8-NR)	63.3%	16.7%	53.3%	N	13%
Wei AH et al. [38]	High-Risk MDS	59	71(26-85)	N	70.2%	70.2%	NR	20%	39%	61%	34%	N
Aktimur SH et al. [39]	R/R AML	30	67(33-84)	8(1-12)	50%	63.3%	7(7.8-10.1)	N	N	83.3%	70%	N
Lou Y et al. [40]	R/R AML	48	61(19-73)	5.1(1.7-15.8)	47.9%	47.9%	9.6(N)	N	N	N	N	18.8%
Labrador J et al. [41]	ND AML R/R AML	ND:38 R/R:39	69(22-86)	N	ND:45% R/R:21%	ND:58% R/R:37%	ND:9.4(N) R/R 5.9(N)	N	N	N	N	15%

Abbreviations: ND, newly diagnosed; R/R, relapsed/refractory; AML, acute myeloid leukemia; MDS, myelodysplastic syndrome; CR: Complete remission; CRi: complete remission with incomplete hematological recovery; OS: overall survival; N: not mentioned; NR: not reached.

Table 2. The treatment regimen of Included Studies in the meta-analysis.

First author	Year	Country	Diagnose	Treatment regimen
Pollyea DA et al. [19,23]	2021	American	Newly diagnosed AML: De novo AML Secondary AML	AZA: 75 mg/m^2; days 1-7; Ven: 100 mg day 1→200 mg day 2→400 mg day 3-28; Or Ven: 50 mg day 1→100 mg day 2→200 mg day 3-28(coadministered with moderate CYP3Ai or PgPi); Or Ven: 10 mg day 1→20 mg day 2→50 mg day 3-28(coadministered with moderate CYP3Ai or PgPi).
DiNardo CD et al. [20,24]	2019	American	Newly diagnosed AML: De novo AML Secondary AML	AZA: 75 mg/m^2/d, days 1-7; Ven was administered with a short ramp-up from 20 mg on day 2 or 100 mg on day 2 to a target dose of 400 mg during the first cycle.
DiNardo CD et al. [21,25]	2020	American	Newly diagnosed AML: De novo AML Secondary AML (History of myelodysplastic syndrome or CMML; Therapy-related AML)	AZA: 75 mg/m^2; days 1-7; S.C. or I.V. Ven: 100 mg day 1→200 mg day 2→400 mg day 3-28; Or Ven: 50 mg day 1→100 mg day 2→200 mg day 3–28 (coadministered with moderate CYP3Ai or PgPi); Or Ven: 10 mg day 1→20 mg day 2→50 mg day 3–28 (coadministered with moderate CYP3Ai or PgPi).
Cherry EM et al. [22,26	2021	American	Secondary AML Treatment-related AML Prior therapy for MDS or MPN Allogeneic stem cell transplantation	Not mentioned.
Byrne M et al. [23,27]	2020	American	Post-hematopoietic cell transplantation relapse of AML	AZA: 75 mg/m^2; days 1-7; Ven was started at 100 mg daily and escalated to 400 mg daily, which was reduced by 75% when used in combination with strong inhibitors of CYP3A4.
Labrador J et al. [24,28	2022	Spanish	R/R-AML	AZA: 75 mg/m^2; days 1-7; S.C. Ven was administrated at 400 mg/d which was reduced when in combination with inhibitors of CYP3A4.
Garcia JS et al. [25,29	2020	American	High-risk MDS	AZA: 75 mg/m^2; days 1-7; S.C. or I.V. Ven was initially given at a dose of 400 mg for 28 days in a 28-day or 14-day cycle.
DiNardo CD et al. [26,30	2018	American	Newly diagnosed AML	AZA: 75 mg/m^2; days 1-7; S.C. or I.V. Ven was administered with a short ramp-up from 20 mg day 2 or 100 mg day 2 to a target dose of 800 mg during the first cycle.
Matthews AH et al. [27,31	2022	American	Newly diagnosed AML: De novo AML Secondary AML Therapy-Related	Not mentioned.
Ucar MA et al. [28,32	2021	Turkey	R/R-AML	Not mentioned.
Zeidan AM et al. [29,33	2019	American	High-risk MDS	AZA: 75 mg/m^2; days 1-7; Ven: 100 mg day 1→200 mg day 2→400 mg day 3-14.
Mirgh S et al. [30,34	2021	India	Newly diagnosed AML R/R-AML	AZA: 75 mg/m^2; days 1-7; S.C. Ven: 50 mg × 2 days→100 mg × 2 days→200 mg × 2 days→400 mg for 14–21 days.
Taniguchi S et al. [31,35]	2021	Japan	Newly diagnosed AML R/R-AML	AZA: 75 mg/m^2; days 1-7; subcutaneously or intravenously Ven: 100 mg day 2→200 mg day 3→400 mg day 4-28. From cycle 2 onward, patients received ven 400 mg orally once daily on days 1–28 of each 28-day cycle.
Pollyea DA et al. [32,36]	2017	American	newly-diagnosed AML	Not mentioned.
Winters AC et al. [33,37]	2019	American	newly-diagnosed AML	AZA: 75 mg/m^2; days 1-7; S.C. or I.V. Ven: 100 mg day 1→200 mg day 2→400 mg day 3-28.
Wei AH et al. [34,38]	2019	American	High-risk MDS	AZA: 75 mg/m^2; days 1-7; S.C. or I.V. Ven was administered with a ramp-up of 100 mg on day 2 to a target dose of 400 mg or 800 mg during the first cycle.
Aktimur SH et al. [35,39]	2020	Turkey	R/R AML: Primary AML Secondary AML	Not mentioned.
Lou Y et al. [36,40]	2020	China	R/R AML	AZA: 75 mg/m^2; days 1-7; S.C. Ven was started at a daily dose of 100 mg with a 3-day ramp to the target dose of 400 mg for 28 days.
Labrador J, et al. [37,41]	2022	Spain	newly-diagnosed AML R/R AML	AZA: 75 mg/m^2; days 1-7; S.C. Ven: 100 mg day 1→200 mg day 2→400 mg day 3-28.

Abbreviations: AZA, azacytidine; Ven, venetoclax; AML, acute myeloid leukemia; R/R, relapsed/refractory; CYP3Ai, CYP3A inhibitor; PgPi, PgP inhibitor; S.C., Subcutaneous; IV, Intravenous.

Note: The treatment regimen in the table is usually a 28-day cycle, with some studies of 14 or 21 days.

Assessment of article quality

The Cochrane Collaboration determined that one of the RCTs in this meta-analysis was not at a high risk of bias. The remaining 18 NRCTs were assessed using MINORS, two of which had a control group we assessed with 12 items, and the remaining 16 were single-arm studies that we could only assess with the first 8 items.

Efficacy of overall study

It mainly included the primary outcome on CR/CRi, as well as secondary outcomes on ORR and OS. Analysis of 18 NRCTs using random-effects models showed that the overall CR/CRi after Ven + AZA treatment for AML or MDS was 57.9% (95% CI, 49.5-65.9%, I²= 83%; figure 2) and overall ORR was 65.4% (95% CI, 56.5-73.3%, I² = 76%; figure 3). Both results showed high heterogeneity. A phase III RCT from DiNardo et al. suggested that CR/CRi was 64.7% and ORR achieved 66.4% [25]. Because more than half of the patients in some studies included in this meta-analysis survived during the follow-up period, the median overall survival could not be reached. The longest follow-up in this meta-analysis was from a retrospective analysis of primary AML, with a median survival of 29.3 months (95% CI, 12.8-NR) [37]. Another study involving relapsed/refractory (R/R) AML reported a median follow-up of 5.1 months (1.7-15.8 months) and median overall survival of 5.3 months (not reported 95% CI) [40]. The included MDS groups did not obtain a median survival during the follow-up period.

Figure 2. Forest plots of overall CR/CRi rate with AML/MDS.

Figure 3. Forest plots of overall ORR rate with AML/MDS.

Subgroup analysis

According to their disease characteristics, the patients were divided into three different study subgroups: newly diagnosed AML (ND-AML), R/R AML, and MDS. The results were as follows: ND-AML pooled CR/CRi 67.5% (95% CI,61.1-73.3%, I² = 54%), R/R AML pooled CR/CRi 30.1% (95%CI, 20.0-44.1%, I²= 66%), high-risk MDS 67.6% (95% CI, 52.6-79.8%. I²= 65%) (Figure 4). The results of ORR were as follows: ND-AML pooled ORR 77.0% (95% CI, 68.7-83.7%, I² = 48%), R/R AML pooled ORR 45.7% (95% CI, 35.9-55.9%, I²= 46%), R/R or High-risk MDS ORR 67.6% (95%CI, 52.6-79.8%, I²= 65%) (Figure S2). AML was classified into favorable/intermediate and poor prognosis groups according to cytogenetic risk stratification. The CR/CRi rates for the favorable/intermediate and poor groups were 59% (95% CI, 39.4-76.1%. I²= 65%) and 62.1% (95% CI, 48.4-73.1%. I²= 0%), respectively (Figure S3). The results of the subgroup analysis according to race were as follows: North America pooled CR/CRi 63.8% (95% CI,63.1-73.0%, I² = 44%), Asia pooled CR/CRi 45.9% (95%CI, 31.0-61.7%, I²= 65%), and Europe pooled CR/CRi 26.5% (95% CI, 14.5-43.2%. I²= 52%) (Figure S4).

Figure 4. Forest plots of CR/CRi rate in subgroups of AML/MDS.

Adverse effects

The analysis of grade 3–4 adverse reactions and mortality within 60 days was performed. Grade 3–4 adverse reactions were analyzed for anemia, neutropenia, thrombocytopenia, and febrile neutropenia. A total of 9 studies reported grade 3–4 adverse reactions. Seven studies reported mortality within 60 days. The combined results were as follows: Anemia was 33.3% (95% CI, 20.6-48.9%, I² = 83%), neutropenia was 53.7% (95% CI, 38.7-67.9%, I² = 86%), thrombocytopenia was 34.6% (95% CI, 27.0-43.5%, I² = 54%), Febrile neutropenia was 42.9% (95% CI, 34.6-51.0%, I² = 56%) (Figure 5). The pooled mortality within 60 days was 13.8% (95%CI, 11.5-16.7%, I² =  0; Figure 6). A study DiNardo et al. reported that its grade 3–4 hematologic adverse effects were 26% for anemia, 45% for thrombocytopenia, and 42% for both neutropenia and neutropenia with fever, respectively [25]. The study, however, did not report 60 days mortality (7% mortality at 30 days).

Figure 5. Forest plots of pooled rate with grade 3–4 adverse events.

Figure 6. Forest plots of the pooled mortality within 60 days.

Sensitivity analysis

Sensitivity analysis was used to test the robustness of our results. We performed sensitivity analysis by removing one piece of literature at a time, which showed that the pooled rate of the remaining literature was similar to the overall meta-results (Figure S5). In addition, we found that Labrador J. et al. had the greatest impact on CR/CRi, with a CR/CRi changed to 0.60 (0.52-0.68) after removing that study [32]. We also divided the studies into 3 subgroups based on disease characteristics, which showed a decrease from high to moderate heterogeneity.

Discussion

This meta-analysis included 19 papers with a total of 1615 patients and systematically evaluated the efficacy and safety of Ven + AZA in the treatment of AML and MDS. We conducted a systematic review on a phase Ⅲ RCT from DiNardo et al. study, in which the CR/CRi and ORR were 64.7% and 66.4% respectively [25]. We combined the 18 NRCTs in this study and found that the CR/CRi and ORR rates were 57.9% and 65.4% respectively, and the combined I² values were 83% and 76% respectively, indicating a high heterogeneity of the combined data. We excluded the literature one by one to perform the sensitivity analysis showing results were stable and found that the study by Labrador J. et al. had a significant impact on the CR/CRi rate [32]. It could be related to that study cohort being refractory/relapsed AML patients and carrying more adverse molecular biological variations (e.g. P53/FLT3). Our subgroup analysis revealed that heterogeneity was reduced across all subgroups. We hypothesized that the source of heterogeneity was related to cytogenetics and mutations in each patient subgroup. Subgroup analysis showed that the CR/CRi and ORR rates were higher in the ND-AML and MDS groups than in the R/R-AML group, while the CR/CRi and ORR rates in the ND-AML group were comparable to the MDS group. The AML groups were classified into favorable/intermediate and unfavorable prognostic groups according to cytogenetic risk stratification. According to our findings, the favorable/intermediate and poor prognosis groups had similar CR/CRi rates of 59% and 62.1% for Ven + AZA, respectively. The similarity in remission rates between the two groups could be explained by a small sample size in the poor prognosis group, which reduced statistical validity. We analyzed grade 3–4 hematologic adverse events and found that neutropenia and neutropenia with fever were the most common, with incidences of 53.7% and 42.9%, respectively, which were higher than anemia and thrombocytopenia. We further found that the early mortality rate (< 60 days) of the regimen was 13.8%. The results of the adverse reaction analysis showed heterogeneity, which could be attributed to the age of the population included in each center, the cycles of the treatment regimen, and differences in treatment interventions.

Ven was approved in 2018 for the treatment of AML, and studies of Ven in combination with hypomethylating agents for AML/MDS have followed. A previous meta-analysis of Ven combined with hypomethylating agents for AML and MDS showed CR and ORR rates of 56% and 68%, respectively [42]. This is consistent with the findings of our study. It appears that either AZA or decitabine (DEC) in combination with Ven for MDS and AML have similar efficacies. A network meta-analysis comparing single-agent AZA and DEC in AML and MDS revealed that DEC was more likely to achieve better CR and ORR, but AZA could achieve better OS in patients ≥75 years of age or with high-risk-MDS. DEC was likely to achieve better CR, ORR, and OS in elderly AML patients, but at relatively high toxicity [43]. This meta-analysis suggests that the Ven + AZA regimen for AML had a CR/CRi of 60.4%, which is slightly lower than that reported in the DiNardo et al. phase Ⅲ clinical trial, which may be related to the fact that some of the literature we included were from real-world studies. Our study also suggests that the Ven + AZA regimen is more effective than single-agent AZA and conventional chemotherapy regimens in elderly patients with AML [25, 44]. According to geographical subgroups, we observed higher efficacy in the North American group than in the Asian and European groups. This could be due to the different mutations backgrounds races, as well as the inclusion of fewer studies in Asia and Europe [45]. However, more clinical evidence is needed to confirm this in the future. This meta-analysis showed that the remission rate and median OS of Ven + AZA for ND-AML were superior to R/R-AML efficacy. This could be because R/R-AML is more likely to carry an unfavorable karyotype and poor prognostic mutations which leads to Ven + AZA resistance [46,47]. The mechanism of resistance to Ven is still unclear, however recently it has been found that RAS/MAPK/MCL-1 may be one of the resistance mechanisms [48]. Due to the different follow-up times for each study, a quantitative analysis of median overall survival was not possible. The longest follow-up in this study was from a retrospective analysis of ND-AML reported by Winters. et al., with a median survival of 29.3 months (95% CI, 12.8-NR) [37]. Another study on R/R AML by Labrador. et al reported a median overall survival of 4 months (2.6-5.8 months) [28]. We found that the CR/CRi of Ven + AZA for high-risk MDS was 67.6%, which is higher than the efficacy of MDS treated with hypomethylating agents alone [49,50]. A randomized, double-blind, phase Ⅲ clinical trial of Ven + AZA versus AZA + placebo for high-risk MDS is currently underway (NCT04401748). In the future, Ven + AZA may become one of the effective treatments for high-risk MDS. This study looked at the adverse effects of this regimen and found that the common adverse effects were neutropenia (53.7%) and neutropenia with fever (42.9%). This is similar to the results of the meta-analysis by Liu. et al describing a combination of Ven and hypomethylating agents in the treatment of AML/MDS [42]. According to this study, AZA does not increase the incidence of adverse reactions as compared to decitabine. We calculated the early mortality (<60 days) rate of the regimen as 13.8%. A phase III clinical trial comparing AZA monotherapy with Semi-Intensive Fludarabine and Cytarabine in the treatment of elderly AML was reported by Vives et al, which showed a 60 days mortality rate of 20% in the AZA monotherapy group [51]. This study and the other reported studies suggest that Ven combined with AZA does not increase early mortality.

The current study has several advantages. Firstly, we were the first to pool data focusing solely on Ven in combination with a single hypomethylating agent (AZA), making the study highly specific. Second, to ensure the reliability of our results, we used a comprehensive search strategy, clearly defined selection criteria, performed a strict quality assessment, and reported in accordance with the PRIMSA statement. Third, we investigated the efficacy of Ven + AZA in patients with AML/MDS for different geographic regions, which has seldom been reported in the past.

Every meta-analysis has flaws, and our study also has several limitations. First, this meta-analysis included a single RCT and lacked prospective studies, which may affect the reliability of the results. As a result, Therefore, more RCTs and prospective clinical studies are required to confirm our findings. Second, because more than half of the participants in the meta-analysis were from the United States, it is unclear whether the evidence from this study can be extended to other populations. Third, the data in this meta-analysis is highly heterogeneous. We used sensitivity analysis to obtain a few factors that caused heterogeneity, but identifying every cause of heterogeneity is difficult. Fourth, despite a comprehensive search of several databases, available literature on MDS is scarce (3 articles), which may limit statistical power and affect the robustness of the results. We hope to update our meta-analysis by adding more studies in the future. Finally, it is difficult to quantitatively synthesize data because of the small sample sizes of some analyzed subgroups, and more sample sizes must be accumulated for analysis.

Conclusion

The present meta-analysis demonstrated that the Ven + AZA regimen is efficacious for the treatment of AML and MDS, with it being more effective for ND AML than R/R AML. The most common adverse effects of this regimen are grade 3–4 neutropenia and neutropenia with fever, but it does not increase early mortality in patients. Additional data from prospective clinical studies are needed to further evaluate the efficacy of Ven + AZA in different genetic prognostic stratification of AML/MDS.

Ethics approval and consent to participate

This study did not directly involve human subjects and therefore did not require ethical approval for institutional board approval.

Availability of data and material

The original manuscripts contained in the study are contained in the article/supplementary material and inquiries can be directed to the corresponding author.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.