Advancing Allogeneic Stem Cell Transplantation in Adult Acute Myeloid Leukemia: Current Perspectives and Strategies

What is the current landscape of allo-HSCT in adult AML?

The role of allogeneic hematopoietic stem cell transplantation (allo-HSCT) in adult acute myeloid leukemia (AML) continues to evolve as new evidence emerges regarding its efficacy across different patient populations. Recent research has significantly expanded our understanding of transplantation outcomes in diverse age groups, donor types, and disease states. This systematic review synthesizes findings from ten studies published between 2020 and 2025, offering valuable insights into the current landscape of allo-HSCT in adult AML patients. The studies encompass a broad range of clinical scenarios, from elderly patients receiving reduced-intensity conditioning to relapsed acute promyelocytic leukemia cases, providing a comprehensive overview of transplantation strategies and their associated outcomes. The evidence suggests that while allo-HSCT remains a potentially curative option, its success is heavily influenced by factors such as donor selection, conditioning intensity, disease status at transplantation, and patient-specific characteristics including age and genetic risk profile. Understanding these nuances is crucial for optimizing transplantation strategies and improving survival outcomes in this challenging disease.

Can novel transplant strategies enhance outcomes for older AML patients?

Haploidentical transplantation has emerged as a promising alternative for elderly AML patients without matched sibling donors (MSD). Li and colleagues demonstrated significantly improved one-year overall survival (OS) and leukemia-free survival (LFS) with haploidentical HSCT compared to chemotherapy in patients aged 55 years or older (OS: 77.9% vs. 62.0%; LFS: 74.1% vs. 42.0%). This survival advantage was primarily driven by a dramatic reduction in relapse rates (16.5% vs. 56.6%), with an acceptable non-relapse mortality (NRM) of 9.4%. These findings are particularly significant as they challenge the traditional view that transplantation carries prohibitive risks in older adults. The study's propensity score matching methodology strengthens the validity of these results, suggesting that age alone should not be a barrier to transplantation when appropriate donors are available. However, it's worth noting that the follow-up period was limited to one year, leaving questions about long-term survival and late complications. As haploidentical donors are available for nearly all patients, these results could significantly expand transplantation access for elderly AML patients who previously had limited options beyond chemotherapy alone. Could this data influence future trial designs to focus more on haploidentical approaches in older populations previously excluded from transplantation studies?

The impact of conditioning intensity and patient age on transplantation outcomes was further elucidated by Vo et al., who examined 495 AML patients aged 60 years and older. Their findings revealed that while reduced-intensity conditioning (RIC) was associated with lower relapse rates in patients aged 60-69 years, it led to significantly worse outcomes in those 70 years and older, with higher non-relapse mortality (HR=3.88, p=0.001) and shorter overall survival (HR=3.46, p<0.001). In contrast, nonmyeloablative (NMA) conditioning appeared to mitigate age-related risks, suggesting that conditioning intensity should be carefully tailored to individual patient characteristics. These observations align with the growing recognition that chronological age alone is an insufficient predictor of transplantation tolerance, and that functional status, comorbidities, and disease biology must be considered collectively. The study's stratification of patients into discrete age cohorts (60-64, 65-69, and ≥70 years) provides valuable granularity that can guide clinical decision-making. The results emphasize the importance of personalized conditioning approaches that balance disease control against treatment-related toxicity, particularly in the most vulnerable patients. As transplantation continues to extend into older populations, the development of refined geriatric assessment tools specific to HSCT candidates may further improve patient selection and outcomes.

Donor type emerged as a critical determinant of transplantation success, with consistent evidence supporting the superiority of matched sibling donors (MSD) over other options. Maffini and colleagues reported substantially better two-year overall survival with MSD (62.4%) compared to matched unrelated donors (MUD, 43%) and haploidentical donors (25.9%) in patients aged 70 years and older with active disease. This advantage was primarily attributed to lower non-relapse mortality with MSD (17.5%) versus MUD (32.2%) and haploidentical donors (43.9%), while relapse rates remained similar across donor types. These findings reinforce the enduring value of MSD as the preferred donor source when available. However, in a separate analysis focusing on patients in complete remission, the same investigators found that haploidentical and matched unrelated donors were associated with lower relapse rates (HR 0.46 and 0.44, respectively) compared to MSD, suggesting that alternative donors may offer enhanced graft-versus-leukemia effects in certain contexts. This apparent dichotomy highlights the complex interplay between disease control and transplantation-related complications, and suggests that donor selection should be guided by individual patient risk profiles rather than universal hierarchies. The expanding use of post-transplantation cyclophosphamide for graft-versus-host disease prophylaxis has improved outcomes with haploidentical donors, potentially narrowing the historical gap between donor types.

How can disease control be balanced with treatment-related risks?

Disease status at transplantation consistently emerged as one of the strongest predictors of outcomes. Yanada et al. demonstrated that patients with relapsed acute promyelocytic leukemia (APL) transplanted in complete remission achieved a five-year OS of 58%, whereas those with active disease at transplantation fared significantly worse. Similarly, prior autologous HSCT was identified as an adverse prognostic factor, with extremely poor outcomes (OS: 6%) observed in patients who underwent allo-HSCT with active disease after failed autologous transplantation. These findings emphasize the critical importance of disease control prior to allo-HSCT and suggest that salvage transplantation for refractory disease remains challenging despite advances in supportive care and conditioning regimens. The timing of transplantation also influenced outcomes, with Sanz et al. reporting that a shorter interval from diagnosis to transplantation was associated with inferior survival in APL patients. This observation may reflect insufficient time for recovery from prior therapy or selection bias toward more aggressive disease biology in patients requiring early transplantation. The collective evidence underscores the value of achieving deep remission before proceeding to transplantation, though this must be balanced against the risk of disease progression during prolonged preparation.

The only randomized controlled trial included in this review, conducted by Niederwieser and colleagues, provided high-quality evidence regarding the trade-offs between disease control and treatment-related mortality in older AML patients. Comparing allo-HSCT to non-transplantation approaches in patients aged 60-75 years in first complete remission, the investigators found that transplantation significantly reduced relapse rates (37.8% vs. 91.1%, p<0.0001) and improved leukemia-free survival (24.5 vs. 15.6 months, p=0.022). However, these benefits were offset by higher non-relapse mortality (33.4% vs. 0%), resulting in no significant difference in overall survival between the two approaches (27.8 vs. 28.6 months). This apparent paradox illustrates the fundamental challenge in allo-HSCT: balancing relapse reduction against treatment-related mortality. The study's rigorous methodology, including intention-to-treat analysis and restricted mean survival time calculations, strengthens the validity of these findings. While the lack of overall survival benefit may seem disappointing, the significant improvement in leukemia-free survival represents an important quality-of-life consideration for many patients. These results highlight the need for novel strategies to reduce non-relapse mortality while preserving anti-leukemic efficacy, such as improved GVHD prophylaxis regimens and enhanced supportive care approaches. What regulatory challenges might arise in implementing post-transplantation maintenance strategies that could further improve outcomes in these high-risk patients?

For relapsed acute promyelocytic leukemia, a distinct pattern emerged from the studies by Min et al. and Sanz et al. While allo-HSCT appeared to be effective as salvage therapy in second relapse (with 6 of 7 patients surviving in Min's cohort), autologous HSCT demonstrated superior outcomes in second complete remission (two-year OS: 82.4% vs. 64.3% with allo-HSCT). This advantage was primarily driven by lower non-relapse mortality with autologous transplantation (2.7% vs. 17.3%), while relapse rates were comparable between the two approaches. These findings align with current guidelines recommending autologous HSCT for relapsed APL in molecular remission, with allogeneic transplantation reserved for high-risk features or multiple relapses. The distinct biology of APL, including its exquisite sensitivity to arsenic trioxide and all-trans retinoic acid, contributes to these differential outcomes compared to other AML subtypes. Kulkarni et al. further demonstrated excellent results with autologous transplantation in molecular remission following bortezomib-enhanced arsenic trioxide salvage therapy, with 95.5% overall survival at a median follow-up of 48 months. These studies collectively suggest a tailored approach to transplantation in APL, with treatment decisions guided by molecular response, prior therapy, and individual risk assessment rather than universal algorithms.

Can risk stratification guide transplantation decisions effectively?

Risk stratification based on European LeukemiaNet (ELN) criteria emerged as another important factor in transplantation decision-making. Socie et al. demonstrated that allo-HSCT improved overall survival specifically in intermediate and poor-risk AML patients (HR=0.51, p=0.004), while the benefit was less clear in those with good-risk disease. This finding aligns with current consensus guidelines that recommend risk-adapted approaches to post-remission therapy. Patients with favorable genetic profiles may achieve durable remissions with chemotherapy alone, whereas those with adverse features derive greater benefit from the graft-versus-leukemia effect of allogeneic transplantation. The study also highlighted significant barriers to transplantation access, including comorbidities, remission status, and donor availability, with only 21.9% of patients ultimately undergoing the procedure. This underscores the continued challenge of extending transplantation benefits to broader patient populations, particularly in resource-limited settings. The expansion of alternative donor strategies, including haploidentical approaches, may help address these access disparities by increasing the pool of potential donors for patients lacking matched siblings.

Is the evidence base free from bias?

The risk of bias assessment revealed generally robust methodological quality among the included studies, with the randomized controlled trial by Niederwieser et al. and the propensity-matched analysis by Li et al. demonstrating minimal bias. However, limitations were identified in several retrospective studies, including potential selection bias, variability in outcome definitions, and incomplete reporting of conditioning regimens. The predominance of retrospective designs and registry analyses introduces inherent limitations, as these approaches are susceptible to unmeasured confounders and reporting biases. Additionally, the heterogeneity in patient populations, conditioning protocols, and follow-up durations precluded formal meta-analysis, limiting the ability to generate pooled effect estimates. Future research would benefit from prospective, multicenter studies with standardized definitions and comprehensive reporting of both efficacy and safety outcomes. The field would also benefit from longer follow-up periods to capture late effects and quality-of-life outcomes, which are increasingly recognized as important considerations in transplantation decision-making. Could standardized patient-reported outcome measures specific to transplantation recipients help address this gap in future studies?

How might these insights shape integrated AML care pathways?

This systematic review offers several key insights for clinical practice. First, it confirms that allo-HSCT remains a potentially curative option for adult AML, particularly for intermediate and poor-risk disease. Second, it highlights the expanding role of alternative donors, especially haploidentical transplantation, in extending transplantation access to patients without matched siblings. Third, it emphasizes the importance of tailoring conditioning intensity to patient characteristics, with nonmyeloablative approaches showing promise in the oldest patients. Fourth, it reinforces the critical importance of achieving remission prior to transplantation, as outcomes for patients with active disease remain poor despite advances in supportive care. Finally, it demonstrates that age alone should not be a barrier to transplantation consideration, though careful assessment of functional status and comorbidities remains essential. The integration of these findings into clinical decision-making requires a multidisciplinary approach, considering disease biology, patient preferences, donor availability, and institutional expertise. As transplantation strategies continue to evolve, the balance between disease control and treatment-related toxicity remains the central challenge in optimizing outcomes for adults with acute myeloid leukemia. How might these findings influence the development of integrated care pathways that incorporate both transplantation and non-transplantation approaches in a risk-adapted manner?

Summary

This comprehensive review analyzes the current state of allogeneic hematopoietic stem cell transplantation (allo-HSCT) in adult acute myeloid leukemia (AML) treatment. The analysis reveals that transplantation outcomes are significantly influenced by donor type, conditioning intensity, disease status, and patient characteristics. Haploidentical transplantation has shown promising results in elderly patients, while matched sibling donors generally yield superior outcomes. The review emphasizes that age alone should not exclude patients from transplantation consideration, though careful assessment of individual risk factors is crucial. Disease control prior to transplantation remains a critical success factor, and risk stratification based on European LeukemiaNet criteria helps guide treatment decisions. The findings highlight the need for personalized approaches in transplantation strategies while balancing disease control against treatment-related risks.