Venetoclax and Hypomethylating Agent in Previously Untreated Higher‐Risk Myelodysplastic Syndromes and Genotype Signatures for Response and Prognosis: A Real‐World Study

Higher-risk myelodysplastic syndrome (MDS), as defined by the revised international prognostic scoring system (IPSS-R), requires more aggressive treatment. Despite the use of hypomethylating agents (HMAs), response rates remain low, underscoring the need for more effective therapies. Venetoclax is a potent, orally bioavailable inhibitor of the B-cell lymphoma 2 (BCL-2) protein, which acts synergistically with HMAs to target malignant myeloid cells. The combination of venetoclax and HMAs has been approved for the treatment of newly diagnosed acute myeloid leukemia (AML) [1]. Clinical trials have shown that venetoclax combined with HMAs exhibits efficacy not only in newly diagnosed patients but also in those with relapsed or refractory MDS [2, 3]. However, real-world data on this combination remain limited, particularly in previously untreated high-risk MDS populations, where genotype signatures for response and prognosis are not yet fully elucidated. Our study aims to address this gap by providing real-world evidence on the efficacy, safety, and genotype-based outcomes of venetoclax plus HMA therapy in this patient population. This retrospective, real-world study was conducted at Peking University People's Hospital and Beijing Qinghe Hospital, analyzing the data from a cohort of previously untreated patients with higher-risk MDS who were administered venetoclax in combination with a HMA between January 2020 and February 2024. Eligible participants were adults aged 18 years or older, with a morphological diagnosis of MDS as per the 2016 World Health Organization (WHO) classification, and were stratified as higher-risk according to the IPSS-R (score > 3.5), with an Eastern Cooperative Oncology Group (ECOG) performance status of 0–2. Patients were excluded if they had incomplete treatment response data, a diagnosis of chronic myelomonocytic leukemia (CMML), or therapy-related MDS. Additionally, patients who were eligible for transplantation were typically not considered for venetoclax and HMA treatment at our center, but those scheduled for transplantation more than 2 months later could receive venetoclax and HMAs while waiting for transplantation if marrow blasts were ≥ 5%. Approval was obtained from the institutional ethics review committee. Baseline data, including demographics, disease characteristics, and treatment patterns, were extracted from medical records. Venetoclax was given at 400 mg daily on Days 1–14 of a 28-day cycle, while decitabine (20 mg/m2, d1-5) or azacitidine (75 mg/m2, d1-7) were administered as per standard protocols. Effectiveness outcomes included the objective response rate (ORR), complete response (CR), and event-free survival (EFS), with responses assessed after each treatment cycle according to the International Working Group (IWG) 2006 criteria. EFS was defined as the time to relapse, AML progression, or death, and overall survival (OS) was time to death from any cause. Statistical analyses were conducted using IBM SPSS Statistics version 22.0. ORR and CR were calculated using the Clopper-Pearson method, and survival analyses were estimated by the Kaplan–Meier method. Subgroup analyses were conducted based on key factors such as age, IPSS-R, and gene mutations. Logistic and Cox regression were used to identify factors influencing outcomes, with multivariate analyses including variables with a p value < 0.2 from univariate analysis. A total of 103 previously untreated higher-risk MDS patients treated with venetoclax and HMAs were included. The median age was 61 years (range: 18–77). The median IPSS-R score was 6.0, with 50.5% classified as high risk and 35.0% as very high risk. ASXL1 mutations were found in 32.2%, followed by RUNX1 (18.9%), U2AF1 (14.4%), and DNMT3A (12.2%). Monoallelic TP53 mutations were present in 13.3%, and biallelic in 1.1% (Table S1). Transplanted patients were generally younger, with a median age of 48 years compared with 63 years in nontransplanted patients (p < 0.001). A higher proportion of transplanted patients were diagnosed with intermediate blast phase 2 (IB2) (83.3% vs. 54.8%, p = 0.022). Key molecular features, including TP53, RUNX1, and NPM1 mutations, were similar between the two subgroups (Table S2). The median time to transplantation was 4.0 months (range: 0.9–64.9). Of the 103 patients, 97 (94.2%) were treated with decitabine and 6 (5.8%) with azacitidine, with a median of 4 treatment cycles (range: 1–14). By the cutoff date of May 27, 2024, 59 patients had discontinued treatment, primarily due to stem cell transplantation (50.8%) and disease progression (30.5%). Other reasons included noncompliance (6.8%), adverse events (AEs, 5.1%), early death from pulmonary infection (3.4%), and stable disease after four cycles (3.4%). The ORR was 79.6% (95% confidence interval [CI]: 70.5–86.9), with a CR rate of 34.0% and marrow complete response (mCR) of 28.2%. After one treatment cycle, the ORR was 68.9%, with a CR of 7.8% and mCR of 33.0%. At a median follow-up of 23.3 months, 27 patients progressed to AML, with a median time to progression of 6.3 months. The median EFS was 14.7 months, and the 1-, 2-, and 3-year EFS rates were 60.5%, 42.1%, and 42.1%, respectively (Figure S1A). Twenty-seven patients (26.2%) died, primarily from disease progression (55.6%) and pulmonary infections (37.0%). The median OS was not reached (NR), with 1-, 2-, and 3-year OS rates of 87.1%, 67.1%, and 58.8%, respectively (Figure S1B). Patients with TP53 mutations had a significantly lower CR rate compared with those without (7.7% vs. 41.6%, p = 0.027), while those with NPM1 mutations had a higher CR rate (87.5% vs. 31.7%, p = 0.003). TP53-mutated patients had shorter median EFS (6.0 vs. 15.7 months, p = 0.015) and OS (21.3 months vs. NR, p = 0.027). Achieving a response after one cycle or as best response was associated with significantly longer EFS and OS (Table 1, Figure S1C–J). There was no significant OS difference between transplanted and nontransplanted patients (NR vs. 26.8 months, p = 0.137). The 1-, 2-, and 3-year OS rates for transplanted patients were 89.7%, 77.0%, and 72.2%, respectively, compared with 86.0%, 60.8%, and 49.6% for nontransplanted patients (Figure S1K). Causes of death differed, with disease progression being more common in nontransplanted patients (n = 14, 70.0%), while pulmonary infection was the leading cause in transplanted patients (n = 5, 71.4%, p = 0.027). The restricted mean survival time analysis showed growing trend of difference between the two groups over time (Table S3). No factors were independently associated with ORR (all p > 0.05). TP53 (hazard ratio [HR] = 4.356, 95% CI: 1.596–11.892, p = 0.004) and RUNX1 mutations (HR = 4.365, 95% CI: 1.675–11.376, p = 0.003) were independent risk factors for shorter EFS, while achieving an objective response was linked to better EFS (HR = 0.193, 95% CI: 0.045–0.829, p = 0.027). For OS, only achieving an objective response was independently associated with improved survival (HR = 0.221, 95% CI: 0.084–0.581, p = 0.002) (Table S4). The most common AEs were anemia (98.1%; grade ≥ 3: 55.3%), neutropenia (98.1%; grade ≥ 3: 53.3%), thrombocytopenia (97.1%; grade ≥ 3: 49.5%), and myelosuppression (86.4%; grade ≥ 3: 47.6%) (Table S5). This study shows that RUNX1 mutations are associated with poorer response, EFS, and OS in higher-risk MDS patients treated with venetoclax and HMAs, consistent with prior findings [4]. Although the ORR in TP53-mutated patients was 69.2%, the CR rate was only 7.7%, with significantly shorter survival, highlighting the limited benefit of venetoclax plus HMAs in this subgroup. The prognostic significance of ASXL1 mutations in MDS remains an area of active investigation, with conflicting findings reported in different studies. Notably, prior studies [5, 6] involving venetoclax plus HMAs in MDS populations showed improved outcomes in ASXL1-mutated patients. However, differences in patient characteristics, such as a higher proportion of MDS with excess blasts or relapsed/refractory cases, and the specific focus on comparing venetoclax plus HMA to HMA alone, may explain the discrepancies with our findings, where ASXL1 mutations were not associated with response or survival. The impact of transplantation on patient prognosis was assessed in this study. Although the 3-year OS rate favored transplanted patients (72.2% vs. 49.6%), this difference was not statistically significant, likely due to the small sample size and baseline disparities, such as younger age among transplanted patients. Non-relapse mortality was the main cause of death in transplanted patients, while disease progression dominated in nontransplanted patients. Most transplant-related events occurred early, likely contributing to the observed survival trend. Despite these limitations, our findings suggest that venetoclax plus HMA is an effective alternative for patients ineligible for immediate transplantation. This study has several limitations. The small sample size and short follow-up may limit generalizability and underestimate long-term outcomes. The absence of a control group hinders direct comparison with other treatments, and baseline differences between transplanted and nontransplanted patients may introduce bias. In conclusion, venetoclax plus HMAs shows promise in treating untreated higher-risk MDS, with favorable response rates and a manageable safety profile. This study identified genotype signatures and response as survival predictors, but further validation through clinical trials is required to strengthen these results. Approval was obtained from the institutional ethics review committee, which also waived the requirement for informed patient consent due to the retrospective nature of the study. The authors declare no conflicts of interest. Data available on request from authors. Data S1. Supporting Information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.