STAG2 突变预测低强度诱导治疗急性髓系白血病移植后良好生存
STAG2 mutation predicts superior posttransplant survival in acute myeloid leukemia treated with low-intensity induction
在接受低强度诱导治疗(包括维奈克拉)的急性髓系白血病患者中,STAG2 突变与移植后生存率显著提高相关。一项新研究在独立的亚洲移植队列中验证了这一发现,尽管样本量较小。STAG2 突变患者的生存优势可能因不同研究中的诱导方案差异而有所不同。
To the Editor:
Allogeneic hematopoietic stem cell transplantation (HSCT) is increasingly offered to patients with acute myeloid leukemia (AML) who achieve remission on low-intensity, venetoclax-based induction, but the molecular determinants of posttransplant outcome in this population are poorly defined. [1] In a recent report of 111 patients with newly diagnosed AML who received frontline venetoclax plus a hypomethylating agent (Ven-HMA) and subsequently underwent allogeneic HSCT, age ≥ 65 years, STAG2 mutation (STAG2MUT) and DNMT3A mutation (DNMT3AMUT) were each associated with superior posttransplant overall survival (OS). [2] Combined into a three-tier score, these variables separated 3-year OS of 89%, 59%, and 19%. Two of the three components — older age and DNMT3AMUT — invert conventional prognostic expectations, and the third, STAG2MUT, is derived from a subgroup of 12 patients. Given the counterintuitive prognostic associations of older age and DNMT3AMUT, we therefore applied the model to an independent Asian cohort transplanted after low-intensity induction for external validation.
We retrospectively identified 46 consecutive adults with newly diagnosed myelodysplastic neoplasm /AML or AML treated at National Taiwan University Hospital (NTUH) who received low-intensity induction and subsequently underwent a first allogeneic transplant. [3] Induction was low-dose cytarabine (LDAC) plus venetoclax [4] in 26 patients (56.5%), LDAC alone in 13 (28.3%), HMA plus venetoclax in four (8.7%), and HMA alone in three (6.5%). Mutational profiling was performed at diagnosis using a 54-myeloid gene next-generation sequencing panel as previously described. [5] OS was calculated from the date of transplant and estimated by the Kaplan–Meier method with log-rank comparison; the composite score was assigned exactly as specified by Warraich et al. [2] (one point each for age < 65 years, STAG2 wild-type and DNMT3A wild-type). The study was approved by the institutional review board.
Cohort characteristics are summarized in Table 1. Median age was 63.5 years (interquartile range 51.8–69.4); 27 patients (58.7%) were younger than 65 years, 18 (39.1%) were 65–75 years, and only one (2.2%) was older than 75 years. European LeukemiaNet (ELN) 2022 risk was adverse in 28 patients (60.9%). At transplant, 28 patients (60.9%) were in first complete remission (CR1), and 18 (39.1%) had refractory disease. Conditioning was fludarabine/busulfan/cyclophosphamide in 31 patients (67.4%), and 19 patients (41.3%) received a haploidentical graft. STAG2MUT and DNMT3AMUT were each present in 6 patients (13.0%), frequencies closely comparable to the 16% and 13% reported in the index cohort.
After a median follow-up of 34.7 months, STAG2MUT was confirmed as a favorable predictor of posttransplant survival: no deaths occurred among the six STAG2MUT patients, giving 2-year OS of 100% vs. 39.2% in STAG2 wild-type patients (median OS 11.7 months; p = 0.028; Fig. 1A). Because no events occurred in the mutant group, the hazard ratio is not estimable by standard Cox regression; using Firth’s penalized likelihood, [6] the estimated hazard ratio was 0.10 (95% CI 0.001–0.72, p = 0.015). In addition, the post-HSCT relapse rate was numeric higher in STAG2WT than in STAG2MUT patients (37.5% vs. 0%, p = 0.16), while the median RFS was not reached vs. 8.7 months, respectively (p = 0.014).
Kaplan–Meier estimates of overall survival from the date of allogeneic stem cell transplantation among 46 patients with newly diagnosed acute myeloid leukemia treated with low-intensity induction, stratified by (A) STAG2 mutation status, (B) age group ( < 65, 65–75, > 75 years), (C) DNMT3A mutation status, and (D) the composite risk score of Warraich et al. (one point each for age < 65 years, STAG2 wild-type and DNMT3A wild-type; low risk 0–1 point, intermediate risk 2 points, high risk 3 points).
In contrast, neither of the other two components of the model was reproduced. Survival did not differ across age groups (p = 0.17), and the curves for patients aged < 65 and 65–75 years were essentially superimposed throughout follow-up (Fig. 1B); the single patient older than 75 years died early. DNMT3AMUT likewise showed no association with OS (p = 0.75), with curves that crossed at approximately 29 months (Fig. 1C). Consequently, the composite score failed to stratify our cohort (p = 0.42): although the low-risk group had the highest survival, the intermediate- and high-risk groups were indistinguishable (Fig. 1D). The score also distributed our patients very differently, assigning 22 (47.8%) to high risk and only 7 (15.2%) to low risk, compared with 25 (22.5%) and 19 (17.1%) of 111 patients, respectively, in the index cohort, a difference that follows directly from the age term in a younger population.
The reproducibility of the STAG2 signal is the most informative finding. It aligns with a large real-world United Kingdom cohort in which STAG2MUT predicted superior OS and high CR rates with Ven-HMA (n = 574; 5.5% underwent allogeneic transplantation in CR1), [7] and with a relapsed/refractory series showing numerically higher response rates in STAG2MUT disease. [8] Our cohort extends this observation to a different low-intensity backbone: whereas the index cohort was treated uniformly with Ven-HMA, five of our six STAG2MUT patients received venetoclax plus LDAC, and the favorable posttransplant outcome was preserved. This suggests that the benefit tracks venetoclax exposure rather than the HMA specifically. Because only one STAG2MUT patient received venetoclax-free therapy, our data cannot distinguish whether the favorable prognosis of STAG2MUT reflects true venetoclax sensitivity or intrinsically more indolent, transplant-curable disease, a distinction with practical consequences that can be resolved only in cohorts with substantial venetoclax-free exposure.
The failure of the age term is, in our view, methodologically instructive rather than contradictory. Warraich et al. [2] attributed the age effect to a markedly higher cumulative incidence of relapse in younger patients, who more often required bridging therapy (23% vs. 7%) and were more often measurable residual disease–positive before transplant (38% vs. 18%). Because all of their patients were in CR/CRi at transplant, remission status could not capture how readily that remission had been achieved, and age may therefore have partly encoded refractoriness to frontline Ven-HMA. Our cohort differs on precisely this axis, since 39% were transplanted with refractory disease and pretransplant disease control is directly recorded rather than uniform, and in this setting age showed no prognostic association. The age structure of the two cohorts also differs substantially: 59% of our patients were younger than 65 years compared with 23% in the index cohort, and patients older than 75 years are effectively absent from our data because allogeneic transplantation is rarely offered beyond that age at our institution. Since the > 75-year stratum contributed the highest survival in the index cohort, our cohort lacks the group in which the reported benefit was most evident. For DNMT3A, all six patients harboring mutations in this study were classified in the adverse-risk ELN-2022 category (one with GATA2/MECOM and five with MR gene mutations), and the small sample size provided very limited statistical power. This contrasts with the previous report, in which DNMT3A-mutated patients had a substantially lower frequency of adverse karyotype (11% vs. 54%), which may partly explain the discrepant prognostic impact observed between the two studies.
Our study has clear limitations. The sample is small and single-center, with only six STAG2MUT patients and zero events among them, so the effect estimate is unstable and multivariable adjustment was not feasible; the observed benefit should be regarded as directionally confirmatory rather than quantitatively precise. Induction was predominantly LDAC-based rather than HMA-based, more than a third of patients were transplanted without remission, conditioning was largely busulfan-based and 42% of grafts were haploidentical, all substantial departures from the index cohort, in which 51% received fludarabine/melphalan and 75% received matched unrelated grafts. We are therefore underpowered to exclude modest effects of age or DNMT3AMUT, and our findings should be read as demonstrating that the composite model does not transport across these differences, not that its components are prognostically inert.
In summary, our independent Asian cohort provides directional validation of the favorable posttransplant outcome associated with STAG2MUT after low-intensity induction, including a predominantly venetoclax-LDAC backbone. These findings support further evaluation of STAG2 as a potential prognostic biomarker while highlighting the need for external validation of the age- and DNMT3A-based components before broader clinical application.
Data availability
The analyzed in this study are available from the corresponding author upon reasonable request, subject to institutional data-sharing policies and ethics approval. The datasets analyzed during this study are available from the corresponding author upon reasonable request.
References
Pratz KW, DiNardo CD, Arellano M, Thirman MJ, Pullarkat V, Becker PS, et al. Long-Term Outcomes of Stem Cell Transplant in Older Patients With Acute Myeloid Leukemia Treated With Venetoclax-Based Therapies. Am J Hematol. 2026;101:926–9.
Warraich M, Kumar S, Fatima M, McCullough K, Al-Kali A, Alkhateeb HB et al. Older Age, STAG2, and DNMT3A Mutations Predict Superior Posttransplant Survival in Newly Diagnosed Acute Myeloid Leukemia Treated With Venetoclax Plus Hypomethylating Agents. Am J Hematol 2026. https://doi.org/10.1002/ajh.70480.
Lo M-Y, Tsai XC-H, Lin C-C, Tien F-M, Kuo Y-Y, Lee W-H, et al. Validation of the prognostic significance of the 2022 European LeukemiaNet risk stratification system in intensive chemotherapy treated aged 18 to 65 years patients with de novo acute myeloid leukemia. Am J Hematol. 2023;98:760–9.
Wei AH, Montesinos P, Ivanov V, DiNardo CD, Novak J, Laribi K, et al. Venetoclax plus LDAC for newly diagnosed AML ineligible for intensive chemotherapy: a phase 3 randomized placebo-controlled trial. Blood. 2020;135:2137–45.
Chang Y-S, Tien F-M, Wang Y-W, Yao C-Y, Tsai XC-H, Ni S-C, et al. Prognostic impact of co-occurring FLT3 mutations across molecular subgroups in intensively treated acute myeloid leukemia: insights from real-world genomic data. Blood Cancer J. 2026;16:3.
Heinze G, Schemper M. A Solution to the Problem of Monotone Likelihood in Cox Regression. Biometrics. 2001;57:114–9.
Othman J, Lam HPJ, Leong S, Basheer F, Abdallah I, Fleming K, et al. Real-world outcomes of newly diagnosed AML treated with venetoclax and azacitidine or low-dose cytarabine in the UK NHS. Blood Neoplasia. 2024;1:100017.
Shahswar R, Gabdoulline R, Krueger K, Wichmann M, Götze KS, Braitsch K, et al. A novel prognostic risk model for patients with refractory/relapsed acute myeloid leukemia receiving venetoclax plus hypomethylating agents. Leukemia. 2025;39:614–22.
Acknowledgements
We sincerely thank Tai-Chung Tseng, Sao-Chih Ni, Min-Yen Lo, Wan-Hsuan Lee, Feng-Ming Tien, Xavier Cheng-Hong Tsai, Chien-Chin Lin, Bor-Sheng Ko, Hwei-Fang Tien, and Wen-Chien Chou (National Taiwan University Hospital, Taipei, Taiwan) for their valuable contributions in providing patient data.
Funding
This study was funded by the Ministry of Science and Technology, Taiwan (MOST 111-2314-B-002-279 and 114-2314-B-002-256-MY3) and the Ministry of Health and Welfare, Taiwan (MOHW 111-TDU-B-221-014001, and 115-TDU-B-221–114004). The funding agencies had no role in the study design, data acquisition, data analysis, manuscript preparation, or the decision to submit the manuscript for publication.
Author information
Authors and Affiliations
Department of Internal Medicine, National Taiwan University Hospital, Hsin-Chu Branch, Hsin-Chu, Taiwan
Yu-Sung Chang
Graduate Institute of Clinical Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan
Yu-Sung Chang & Hsin-An Hou
Division of Hematology, Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Yu-Sung Chang & Hsin-An Hou
Division of General Medicine, Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan
Hsin-An Hou
Authors
- Yu-Sung Chang
- Hsin-An Hou
Contributions
YS Chang: Conceptualization, Formal Analysis, Data Curation, Writing – Original Draft, Visualization; HA Hou: Design, Conceptualization, Supervision, Funding Acquisition, Methodology, Writing – Review & Editing, Project Administration.
Corresponding author
Correspondence to Hsin-An Hou.
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
This study was conducted in accordance with the Declaration of Helsinki and was approved by the Research Ethics Committee of National Taiwan University Hospital (approval numbers 202109078RINB, 202207050RINB and 202511035RINE).
Informed consent
All patients provided written informed consent for the collection, banking and research use of their clinical data and biological samples at the time of enrollment.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Chang, YS., Hou, HA. STAG2 mutation predicts superior posttransplant survival in acute myeloid leukemia treated with low-intensity induction. Blood Cancer J. 16, 194 (2026). https://doi.org/10.1038/s41408-026-01661-1
Received: 04 September 2026
Accepted: 15 October 2026
Published: 31 October 2026
Version of record: 31 October 2026
DOI: https://doi.org/10.1038/s41408-026-01661-1
来源:Nature 健康护理主题 · nature.com