Venetoclax and decitabine vs intensive chemotherapy as induction for young patients with newly diagnosed AML Free

Acute myeloid leukemia (AML) is the most common form of acute leukemia in adults, representing >80% of cases, with a 5-year overall survival (OS) rate of 31.7%.¹ For the past 50 years, the combination of cytarabine and an anthracycline (7+3 regimen) has been the standard induction therapy for young patients with newly diagnosed AML eligible for intensive chemotherapy, yielding complete remission (CR) rates of 46% in the adverse-risk group and 80% to 90% in favorable and intermediate-risk groups.^2-5 Despite these outcomes, the 7+3 regimen is associated with significant toxicities, including myelosuppression, neutropenic fever, infections, and bleeding, with early mortality rates reported between 10% and 29%.^6,7 Therefore, there is a critical need for safer and more effective induction regimens for young patients with AML.

Venetoclax (VEN), a selective inhibitor of B-cell lymphoma 2 (BCL-2), is currently standard therapy when combined with hypomethylating agents (HMAs) for older patients with AML or those unfit for intensive chemotherapy.⁸ The VEN-HMA combination has dramatically altered the therapeutic landscape of AML. In the VIALE-A study, VEN plus azacitidine demonstrated a composite CR (CRc) rate of 66.4%, with a low early mortality rate of 6%, and extended OS in older patients with intermediate and adverse genetic risk.⁹ Furthermore, VEN combined with decitabine (DEC; VEN-DEC) has shown efficacy comparable to VEN-azacitidine,¹⁰ achieving a CRc rate of 83% with reduced infections in a phase 2 trial involving patients with newly diagnosed AML with adverse risk.¹¹ In unfit patients with AML with favorable-risk features, such as NPM1 mutations and CBFB::MYH11 fusions, retrospective studies have indicated promising responses to VEN-HMAs.^12,13 These findings suggest that VEN-HMAs could be beneficial as frontline induction therapy for younger patients across different genetic groups as well.

To further evaluate this possibility, a phase 2 trial conducted at MD Anderson Cancer Center compared VEN-DEC10 (10-day DEC) with intensive chemotherapy with intermediate- or high-dose cytarabine in patients with intermediate and adverse risk, revealing superior CRc rates, reduced 30-day mortality, and prolonged OS in the VEN-DEC10 arm.⁷ However, some studies have reported better outcomes with intensive chemotherapy in specific subgroups. For example, a study from the University of Colorado¹⁴ suggested that intermediate-risk patients had better responses to intensive chemotherapy. In addition, our retrospective analysis indicated that patients with RUNX1::RUNX1T1 exhibited higher response rates to 7+3 regimen than VEN-HMAs.¹⁵ These retrospective findings highlight inconsistencies and underline the need for prospective comparisons to accurately determine treatment outcomes.

To address this, we conducted a multicenter, open-label, randomized trial to compare the efficacy and safety of VEN-DEC with cytarabine and idarubicin as induction therapy in young patients with AML eligible for intensive chemotherapy. This study was designed as a noninferiority trial. Here, we present the results of this study.

Eligible patients were aged 18 to 59 years with a confirmed diagnosis of previously untreated AML in accordance with World Health Organization criteria.¹⁶ Cytoreduction with hydroxyurea, cytarabine at a cumulative dose of <1.0 g, or leukapheresis was permitted. Exclusion criteria included acute promyelocytic leukemia, central nervous system involvement, solitary myeloid sarcoma, AML secondary to myelodysplastic neoplasia and treated with HMAs, previous exposure to BCL-2 inhibitors, New York Heart Association grade 3 to 4 cardiovascular disease, and active and uncontrolled infection. Full eligibility criteria are provided in the supplemental Appendix, available on the Blood website.

This phase 2b, multicenter, randomized, open-label, controlled trial evaluated the noninferiority of VEN-DEC (study group, VEN-DEC) compared with the standard 7+3 regimen (control group, IA-12) as induction therapy in newly diagnosed AML. Patients were simply randomized 1:1 to receive either VEN-DEC or IA-12. All participants were hospitalized from day 1 of induction to neutrophil recovery to monitor for treatment-related complications. Prophylaxis for tumor lysis syndrome (TLS) included uric acid-lowering agents, intravenous hydration, and routine laboratory assessments was performed during induction.

In the VEN-DEC group, VEN was administered orally, starting at 100 mg on day 1 and escalating to 400 mg by day 3, which continued until day 28 of the first cycle. Dose adjustments were made for CYP3A inhibitors (supplemental Tables 1 and 2). For subsequent cycles, a fixed VEN dose of 400 mg daily was used. DEC was administered IV at 20 mg/m² daily on days 1 to 5 of each 28-day cycle. Patients in the IA-12 group received idarubicin (12 mg/m², days 1-3) and cytarabine (100 mg/m², days 1-7) IV, with dose adjustments for liver function (supplemental Table 3). Bone marrow assessments were conducted at the end of cycle 1 induction therapy (±3 days), following the 2022 European LeukemiaNet (ELN-2022) risk classification.¹⁷ Patients achieving partial remission (PR) underwent reinduction with the initial regimen and repeated bone marrow assessment. Consolidation therapy, including 1 to 4 cycles of high-dose cytarabine (2 g/m² every 12 hours on days 1-3) based chemotherapy, was given to patients in CR or CR with incomplete hematologic recovery (CRi). Allogeneic hematopoietic stem cell transplantation (allo-HSCT) was recommended for adverse-/intermediate-risk group and favorable-risk patients with a high relapse risk as dynamically assessed by physicians. Treatment continued until disease progression, unacceptable toxicity, or consent withdrawal or as defined by protocol (supplemental Figure 8).

All patients in the intention-to-treat population were included in the efficacy analysis. The primary end point was the CRc rate after induction therapy, including both initial and reinduction therapy. Secondary end points included the incidence of grade ≥3 infections during induction, duration of severe myelosuppression, event-free survival (EFS), OS, and measurable residual disease (MRD) negativity after induction. Exploratory analyses examined induction efficacy across genetic subgroups. A fixed sequence testing procedure was applied for the primary and secondary end points. Statistical significance for secondary end points was contingent upon achieving significance for the primary end point. These end points were tested sequentially in the specified order, without adjustment for multiplicity, using a nominal 2-sided type I error rate of 5%.

Response assessments were conducted at the end of induction and before each cycle of consolidation chemotherapy or conditioning for allo-HSCT, until disease progression or consent withdrawal. CR was defined as <5% bone marrow blasts, no circulating blasts, no extramedullary disease, absolute neutrophil count ≥1.0 × 10⁹/L, and platelet count ≥100 × 10⁹/L. CRi was defined as CR with absolute neutrophil count <1.0 × 10⁹/L or platelets <100 × 10⁹/L. CRc encompassed both CR and CRi. PR was defined by 5% to 25% bone marrow blasts and ≥50% reduction from baseline. Risk stratification followed ELN-2022.¹⁷ MRD negativity,¹⁷ defined as <0.1% CD45-expressing cells with the target immunophenotype by multiparameter flow cytometry, was assessed centrally in patients with CRc. All MRD assessments were conducted to achieve a sensitivity threshold of 0.1%. To confirm responses without MRD negativity, a subsequent assessment was required at least 4 weeks apart.

OS was measured from randomization to death. EFS was defined as the time from randomization to disease progression, initiation of new antileukemic therapy, confirmed relapse, or death. Patients who withdrew consent were excluded from follow-up, but those who discontinued for other reasons were followed for survival.

Safety analysis included all patients who received at least 1 dose of study medication. Adverse events (AEs) were defined as those occurring from the first dose until 30 days after induction or before consolidation therapy. AEs were graded using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0.

The noninferiority design was based on our phase 2 trial data, where VEN-DEC demonstrated a favorable toxicity profile and faster platelet recovery than intensive chemotherapy.¹¹ Sample size was calculated to establish noninferiority, assuming CRc rates of 83% for VEN-DEC and 70% for IA-12. With a noninferiority margin of 5%, a sample size of 172 patients provided 80% power with a 1-sided type I error of 2.5%. Allowing for a 10% dropout rate, the final target was 188 patients in the intention-to-treat population. The primary end point was assessed using the Farrington-Manning method,¹⁸ and secondary end points were analyzed without adjustment for multiplicity. Confidence intervals (CIs) were calculated using the Clopper-Pearson method. Comparisons of patient characteristics and AEs were made using the χ² test or Fisher exact test for categorical variables and the Mann-Whitney test for continuous variables. Probabilities for time-dependent events were calculated using the Kaplan-Meier method and analyzed using the log-rank test for univariable comparisons. The hazard ratio (HR) between treatment groups was estimated using the Cox proportional-hazards model. The clinical data cutoff date was 30 April 2024. Data were analyzed using SPSS (version 29.0) and R (version 4.4). The study followed the Declaration of Helsinki and received ethical approval (ethical committee of the First Affiliated Hospital of Soochow University [approval number 2021305]). A written informed consent was obtained from all patients. The trial was registered at ClinicalTrials.gov (identifier NCT05177731).

Between 8 March 2022 and 5 February 2024, 255 patients were screened at 3 centers in China, and 188 were randomized (Figure 1; supplemental Table 7): 94 to the VEN-DEC group and 94 to the IA-12 group. The median age was 45 years in the VEN-DEC group (53% male) and 40 years in the IA-12 group (66% male). At randomization, 32% and 34% of patients in the VEN-DEC and IA-12 groups, respectively, had white blood cell counts >30 × 10⁹/L. Baseline characteristics in both groups showed no statistically significance detailed in Table 1.

Of the 188 patients, 182 (97%) completed induction therapy per protocol. This included 143 patients who achieved CRc after 1 induction cycle, 20 who achieved PR and underwent reinduction, and 19 who were withdrawn for not achieving PR. In the VEN-DEC group, 1 patient did not receive reinduction owing to a serious AE (SAE). In the IA-12 group, 3 patients did not undergo reinduction owing to SAEs, and 2 were withdrawn by their physicians (supplemental Table 4).

In the VEN-DEC group, 43 of 94 patients (46%) completed 4 cycles of consolidation chemotherapy compared with 34 patients (36%) in the IA-12 group. A total of 82 patients (44%) underwent allo-HSCT (supplemental Figure 9), with 40 patients (43%) in the VEN-DEC group and 42 patients (45%) in the IA-12 group. The rate of allo-HSCT in first CR was 83% (68/82), including 31 patients in the VEN-DEC group and 37 in the IA-12 group. Transplantation rates were 71% (30/42) in the adverse-risk group, 57% (24/42) in the intermediate-risk group, and 27% (28/104) in the favorable-risk group. Among the 28 favorable-risk patients who underwent allo-HSCT, 18 received a transplant during first CR (including 5 with MRD positive) and 10 received a transplant during greater than or equal to second CR or active disease.

CRc was achieved after induction therapy (including the initial induction and reinduction cycle as needed) in 89% of patients (95% CI, 81-95) in the VEN-DEC group and 79% of patients (95% CI, 69-87) in the IA-12 group. Of these, 78% of patients (95% CI, 68-86) in the VEN-DEC group and 75% (95% CI, 64-83) in the IA-12 group achieved CRc after initial induction. The difference in CRc between VEN-DEC and IA-12 was 10.6% (95% CI, 0.2-21.3). With a noninferiority margin of 5%, the P value from the Farrington-Manning test for noninferiority was .0021 (Table 2), demonstrating noninferiority of VEN-DEC at the 2.5% significance level. In the VEN-DEC group, CR was achieved by 83% (95% CI, 74-89), whereas 6% (95% CI, 3-13) achieved CRi. In contrast, all patients in the IA-12 group achieved CR. MRD negativity after induction was observed in 80% of patients (67/84) in the VEN-DEC group and 76% (56/74) in the IA-12 group.

The median time to response and duration of response were similar between the groups. The median time to first CRc was 43 days (95% CI, 40-46) in the VEN-DEC group and 38 days (95% CI, 35-41) in the IA-12 group (P = .26). The mean duration of CR was 18.1 months (95% CI, 15.9-20.3) in the VEN-DEC group and 19.1 months (95% CI, 17.2-21.1) in the IA-12 group (P = .29).

Subgroup analyses revealed statistically significant differences favoring VEN-DEC in certain populations at the 2-sided 5% level (Figure 2). The detailed results of interaction P values for subgroup comparisons are provided in supplemental Table 5. In the ELN-2022 adverse-risk group, CRc rates were 91% (95% CI, 72-99) for VEN-DEC compared with 42% (95% CI, 20-67) for IA-12 (P < .001; P for interaction = .017). In patients with U2AF1, CRc was 100% (95% CI, 40-100) for VEN-DEC compared with 14% (95% CI, 0-58) for IA-12 (P = .015; P for interaction = .008). Conversely, in patients with RUNX1::RUNX1T1 rearrangement, CRc rates were 44% (95% CI, 14-79) for VEN-DEC and 88% (95% CI, 64-99) for IA-12 (P = .028; P for interaction <.001).

Trends favoring VEN-DEC were observed in the following subgroups, but without significant interaction effects. In patients older than 40 years, CRc rates were 91% (95% CI, 81-97) for VEN-DEC and 75% (95% CI, 60-86) for IA-12 (P = .032; P for interaction = .239). In patients with epigenetic modifier mutations (IDH1/IDH2, DNMT3A, ASXL1, and TET2), CRc rates were 91% (95% CI, 76-98) for VEN-DEC and 67% (95% CI, 48-82) for IA-12 (P = .033; P for interaction = .129). Detailed mutational landscapes are provided in supplemental Figure 1. No significant difference in MRD negativity rates was observed between 2 groups after induction therapy (supplemental Figure 2).

All 188 patients were included in the safety analysis for induction therapy, including 13 patients in the VEN-DEC group and 7 in the IA-12 group who received reinduction with the same regimen. All patients experienced at least 1 AE. Common AEs are summarized in Table 3. The incidence of treatment-related SAEs was significantly lower in the VEN-DEC group (19 patients, 20%) than the IA-12 group (39 patients, 42%; P = .003). SAEs of grade ≥3 were more common in the IA-12 group, including pneumonia (15% vs 32%; P = .009), febrile neutropenia (10% vs 31%; P < .001), and sepsis (7% vs 25%; P = .002). Septic shock occurred in 1% (1 patient) in the VEN-DEC group and 6% (6 patients) in the IA-12 group. Early deaths within 100 days of induction were 1% (1 patient) in the VEN-DEC group and 4% (4 patients) in the IA-12 group.

The incidence of hematologic AEs (grade ≥3) was similar between VEN-DEC and IA-12, including leukopenia (97% vs 98%; P = 1.0), neutropenia (86% vs 89%; P = .657), thrombocytopenia (56% vs 68%; P = .097), and anemia (26% and 28%; P = .869), but febrile neutropenia (43% vs 69%; P < .001) was significantly less frequent in the VEN-DEC group. The median duration of grade 4 neutropenia was longer in the VEN-DEC group (23 days [interquartile range (IQR), 17-28]) than the IA-12 group (19 days [IQR, 16-23]; P = .001). Conversely, the median duration of grade 4 thrombocytopenia was shorter in the VEN-DEC group (13 days [IQR, 0-20]) than the IA-12 group (19 days [IQR, 14-26]; P < .001).

In the VEN-DEC group, 56% of patients (41/73) received granulocyte colony-stimulating factor (G-CSF) compared with 70% (49/70) in the IA-12 group. The median duration of G-CSF administration was significantly shorter in the VEN-DEC group, with a median of 2.3 days (95% CI, 1.6-3.2) vs 4.0 days (95% CI, 3.1-5.0) in the IA-12 group (P = .010). The median volume of red blood cell transfusions was also lower in the VEN-DEC group at 6 units (IQR, 4-10) than 10 units (IQR, 6-12) in the IA-12 group (P = .012). Similarly, the median platelet transfusion volume was reduced in the VEN-DEC group at 25 units (IQR, 0-40) vs 35 units (IQR, 25-45) in the IA-12 group (P < .001).

The overall incidence of infections was significantly lower in the VEN-DEC group than the IA-12 group (Table 3). Notably, grade 3 or higher infections occurred in 30 patients (32%) in the VEN-DEC group, significantly fewer than the 63 patients (67%) in the IA-12 group (P < .001). Gastrointestinal AEs of any grade were common, primarily manifesting as nausea and vomiting, which affected 35% of patients in the VEN-DEC group and 27% in the IA-12 group (P = .269). Diarrhea occurred in 9% of the VEN-DEC group and 20% of the IA-12 group (P = .036), whereas abdominal pain was reported in 1% and 13% of the respective groups (P = .002). Both the VEN-DEC and IA-12 regimens had an equivalent incidence of TLS during induction therapy, with 5 cases reported in each group (5/94 [5.3%]). Among these, 1 patient in each group experienced life-threatening TLS (supplemental Table 9).

Four patients discontinued induction therapy due to AEs, with 1 in the VEN-DEC group (due to intracranial infection) and 3 in the IA-12 group (due to septic shock, syncope, and pneumonia with gastrointestinal bleeding). VEN dose adjustments occurred in 3 patients (3%) in the VEN-DEC group, with the treatment duration reduced from 28 to 21 days. In addition, 71% of patients (67/94) in the VEN-DEC group received a reduced VEN dose of 100 or 200 mg/d owing to concomitant use of posaconazole or voriconazole.

The median follow-up was 12.1 months (range, 0.33-26.5). At data cutoff, 13 patients (14%) in the VEN-DEC group and 11 patients (12%) in the IA-12 group remained on consolidation therapy. The median EFS was not reached in either group. The 1-year EFS was 64.4% (95% CI, 54.5-76.1) in the VEN-DEC group and 62.6% (95% CI, 52.8-74.2) in the IA-12 group, with an HR of 0.91 (95% CI, 0.55-1.50; P = .714; Figure 3). At 6 months, the EFS rates in both groups were comparable, remaining stable between 73% and 76%.

The median OS was also not reached in either group. The 1-year OS was 83.1% (95% CI, 74.5-92.5) in the VEN-DEC group and 83.5% (95% CI, 75.5-92.4) in the IA-12 group, with an HR for death of 1.15 (95% CI, 0.56-2.35; P = .705; Figure 3). Among patients with intermediate genetic risk, the 1-year OS was 81.1% (95% CI, 65.9-99.9) in the VEN-DEC group and 71.1% (95% CI, 50.0-100) in the IA-12 group (HR for death, 0.39; 95% CI, 0.11-1.38; P = .129). For patients with adverse genetic risk, the 1-year OS was 70.4% (95% CI, 50.2-98.8) in the VEN-DEC group and 69.1% (95% CI, 49.5-96.5) in the IA-12 group (HR for death, 1.07; 95% CI, 0.31-3.75; P = .91). However, in patients with favorable genetic risk, OS was significantly lower in the VEN-DEC group than the IA-12 group after 11 months (landmark P = .024; supplemental Figure 3).

At the last follow-up, although the overall number of deaths was similar between the VEN-DEC and IA-12 groups (17% vs 15%; respectively), the causes of death differed. Most deaths in the VEN-DEC group (12%) were caused by disease progression, whereas in the IA-12 group deaths were primarily caused by treatment-related SAEs (9%; P = .013) (supplemental Table 6). Of the 11 patients in the VEN-DEC group who died from relapse, 8 had basic leucine zipper (bZIP) in-frame (inf) mutated CEBPA (CEBPA^bZIP-inf) at diagnosis. The 1-year relapse-free survival for patients with CEBPA^bZIP-inf mutation was significantly lower in the VEN-DEC group (52.5%; 95% CI, 33.2-83.0) than the IA-12 group (85.1%; 95% CI, 68.0-100) (HR for relapse-free survival, 5.43; 95% CI, 1.14-25.75; P = .017), which may have affected OS in this subgroup (HR for death, 2.92; 95% CI, 0.30-28.10; P = .331; supplemental Figure 4).

Patients who achieved CRc after induction had significantly better OS in both groups (P < .001; supplemental Figure 5). The 1-year OS was 87.2% in the VEN-DEC group and 92.2% in the IA-12 group for these patients compared with 57.1% and 52.8% for those who did not achieve CRc. OS was similar between patients who underwent allo-HSCT and those who did not in both treatment arms (supplemental Figure 6). However, in the adverse-risk group, patients who received allo-HSCT showed significantly improved OS in both treatment groups. The median OS was not reached in either group for patients who underwent transplant, whereas it was 6.6 months in the VEN-DEC group and 4.6 months in the IA-12 group for patients who did not undergo transplant (P = .002; supplemental Figure 7).

To our knowledge, this randomized, head-to-head study is the first to compare VEN with HMAs to the conventional 7+3 regimen in young and fit patients with AML. Our results demonstrated that VEN-DEC achieved a not inferior CRc rate to IA-12 in patients with untreated, intensively chemotherapy-eligible AML. The high response rates for VEN-DEC and IA-12 were 89% and 79%, respectively, consistent with previously reported rates of 81% for VEN-DEC10⁷ and 70% to 83% for IA-12 in large-scale clinical studies.^4,5,19 In subgroup analyses, patients with adverse genetic risk, U2AF1 or epigenetic modifier mutations, and those older than 40 years potentially exhibited more favorable responses to VEN-DEC than IA-12. These data suggest that VEN-DEC may be a viable alternative frontline treatment for young, fit patients with AML, offering valuable guidance on selecting initial regimens for newly diagnosed cases.

Conventional intensive chemotherapy has historically yielded unsatisfactory CR rates of <50% in patients with ELN high risk.²⁰ In contrast, previous studies have demonstrated improved prognoses with VEN-HMA in adverse-risk patients. Retrospective studies from the University of Colorado¹⁴ and Mayo Clinic²¹ showed prolonged survival for patients with AML with adverse risk treated with VEN-HMA compared with intensive chemotherapy. In addition, VEN has been shown to abrogate the adverse impact of splicing factor mutations in AML,²² and VEN-HMA also achieved superior survival compared with the 7+3 regimen in patients with SF3B1 mutation.²³ In alignment with these studies, VEN-DEC demonstrated a 83% CRc rate in ELN adverse-risk AML in our phase 2 trial.¹¹ In the present study, VEN-DEC achieved a CRc rate of 91%, significantly higher than the 42% observed in the IA-12 group, further highlighting VEN-DEC’s advantage in adverse-risk AML. These findings have motivated the initiation of a randomized multicenter study comparing VEN-HMA with the 7+3 regimen in patients with ELN adverse-risk AML (NCT05939180).

Interestingly, our study observed a significantly lower CRc rate in patients with RUNX1::RUNX1T1 fusion treated with VEN-DEC (44%) than IA-12 (88%). This is consistent with results from several retrospective studies where response rates to VEN-HMA ranged from 25% to 40.9% in this subgroup.^24,25 These findings concordant with previous reports that patients with RUNX1::RUNX1T1 fusion are associated with low expression of BCL-2²⁶ and high expression of BCL-XL.²⁷ Given that patients with RUNX1::RUNX1T1 are considered curable with intensive chemotherapy and gemtuzumab ozogamicin and are often excluded from clinical trials, our results provide prospective evidence supporting the use of the 7+3 regimen over VEN-HMA in this patient subgroup.

In our study, patients with favorable-risk genetics in the VEN-DEC group exhibited a higher relapse rate and inferior OS than the IA-12 group. Notably, 47% of relapsed patients (9/19) in the VEN-DEC group harbored CEBPA^bZIP-inf mutation, which may have contributed to their poor survival outcomes. This aligns with retrospective data from our institution, where patients with CEBPA^bZIP-inf mutation treated with VEN-HMA achieved a 100% CRc rate after 1 cycle but experienced a high relapse rate of 46.2% compared with 22.1% for those treated with intensive chemotherapy.²⁸ Although a retrospective pooled analysis identified CEBPA^bZIP-inf mutation as a high-benefit group for VEN and azacitidine, the analysis included only 4 patients with these mutations.²⁹ Therefore, our findings suggest that patients with CEBPA^bZIP-inf mutation may benefit more from 7+3 induction therapy, highlighting the need for prospective studies in larger patient cohorts.

The safety profiles of both VEN-DEC and IA-12 were consistent with the established toxicities of these regimens.^30,31 VEN-DEC was associated with a more favorable safety profile, including a lower incidence of febrile neutropenia, fewer severe infections, and reduced supportive care needs, such as G-CSF use and blood product transfusions. These findings are in concordance with a multicenter, real-world study involving >70 year older patients with AML,³² which reported comparable nonhematologic toxicities but a significantly lower incidence of febrile neutropenia, documented infections, and grade 3 to 4 anemia and thrombocytopenia in patients treated with VEN-HMA than those receiving the 7+3 regimen. The favorable safety profile of VEN-DEC may ensure the patients in this group to receive subsequently intensive consolidation treatments including high-dose cytarabine and allo-HSCT.

There are several limitations to this study. First, the short-term end point of treatment response was selected as the primary end point rather than survival. This study aimed primarily to evaluate the feasibility of VEN-HMA as a frontline treatment in young/fit patients with AML. At the time of study initiation, there were no established guidelines supporting continued VEN maintenance in this population. Therefore, high-dose cytarabine consolidation was applied uniformly to all participants in accordance with ELN-2017 and National Comprehensive Cancer Network guidelines, which may have influenced survival outcomes and limited the comparability between induction regimens beyond the initial treatment phase. Second, the use of CRc as the primary end point may not fully capture the nuances of treatment response, particularly given the presence of CRi in the VEN-DEC arm. This difference reflects the myelosuppressive effects of VEN and highlights the complexity of treatment outcomes. Third, the short follow-up time may limit the interpretation of Kaplan-Meier survival curves and their differentiation between treatment groups, restricting conclusions on the long-term impact of induction therapy on relapse and survival. Finally, owing to the limited sample size, the exploratory findings from subgroup analyses should be interpreted with caution and warrant further prospective investigation in larger cohorts.

In conclusion, VEN-DEC demonstrated noninferior response rates and a superior safety profile compared with IA-12 for the frontline treatment of young/fit patients with untreated AML. Subgroup analyses suggest that VEN-DEC regimen is favored in patients with adverse-risk genetics, whereas the IA-12 regimen may be more suitable for those with RUNX1::RUNX1T1 fusion or CEBPA^bZIP-inf mutations.

The authors acknowledge the statistical analysis support provided by Huan Yi from the Medical Department at AbbVie, China. The authors are grateful to all the patients and their families for their participation in this study.

This study was supported by grants from the Suzhou Health Commission Key Disease Program (LCZX202301), the Clinical Research Program of The First Affiliated Hospital of Soochow University (BXLC001), and the National Natural Science Foundation of China (grant 82470174).

Contribution: S.C., H.D., and J.L. contributed to the concept and design of the study; J.L., S.X., Ying Wang, X. He, X. Hu, M.M., Y.Z., X.Y., M. Xu, Y.S., F.D., Q.W., M. Xue, Yun Wang, A.D., X.D., and S.C. were responsible for the provision or treatment of research patients; J.L., H.D., and J.X. collected and assembled the data; J.L., N.D., Z.T., and J.X. performed data analysis and interpretation; J.L. and H.D. wrote the first draft of the manuscript, which was read and revised by Y.X., S.C., and D.W.; and all authors approved the final manuscript for submission.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Yang Xu, Department of Hematology, The First Affiliated Hospital of Soochow University, National Clinical Research Center for Hematologic Diseases, 188 Shizi St, Suzhou 21500, China; email: yangxu@suda.du.cn; Hai-ping Dai, Department of Hematology, The First Affiliated Hospital of Soochow University, National Clinical Research Center for Hematologic Diseases, 188 Shizi St, Suzhou 21500, China; email: daihaiping8@126.com; De-pei Wu, Department of Hematology, The First Affiliated Hospital of Soochow University, National Clinical Research Center for Hematologic Diseases, 188 Shizi St, Suzhou 21500, China; email: drwudepei@163.com; and Su-ning Chen, Department of Hematology, The First Affiliated Hospital of Soochow University, National Clinical Research Center for Hematologic Diseases, 188 Shizi St, Suzhou 21500, China; email: chensuning@suda.edu.cn.