Introduction
Allogeneic hematopoietic cell transplantation (HCT) is a curative treatment for severe aplastic anemia (SAA), but specific transplant guidelines have been lacking. A panel of experts developed consensus recommendations using GRADE methodology, emphasizing the preferential use of bone marrow as a graft source, rabbit ATG for conditioning, and fludarabine-based regimens for high-risk patients and those with unrelated or haploidentical donors. They removed the historical 40-year age limit, recognizing that fit older adults may also benefit from HCT. The panel advocates prioritizing unrelated or haploidentical donor HCT over immunosuppressive therapy when a matched related donor is unavailable. For graft-versus-host disease (GVHD) prophylaxis, they recommend either a calcineurin inhibitor with methotrexate or post-transplant cyclophosphamide. These guidelines represent significant advancements in transplant strategies for SAA and highlight the need for continued research in donor selection, conditioning, and post-transplant care.
Disease and Patient Assessment Prior to HCT
Question 1: How Should Patients With Suspicion of SAA Be Assessed Prior to HCT?
Recommendation
The panel recommends that all patients undergo evaluation to confirm the SAA diagnosis and exclude an underlying inherited bone marrow failure (BMF) syndrome. All newly diagnosed patients should be assessed for HCT and early donor search initiated (Good practice statement).
Implementation Considerations
SAA diagnosis requires thorough evaluation to exclude inherited BMF or secondary causes. Key tests include bone marrow cellularity, cytogenetics, PNH screening, and exclusion of hypoplastic MDS. Patients under 50 should undergo telomere length and chromosome breakage analysis to rule out genetic disorders. Early HLA typing is essential for donor identification, and DSA testing is needed for haplo-HCT. Delays beyond three months and high transfusion burden worsen outcomes, so timely donor selection and transplant preparation are crucial.
Question 2: Should An Age Cutoff of 40 yr Be Used for Adult Patients Receiving HCT for SAA?
Recommendation
The guideline panel recommends against using age 40 yr as the cutoff. Due to improvement in supportive care and tolerability of current conditioning regimens, our panel suggests that upfront HCT may be considered selectively in biologically fit patients up to 50 yr or even beyond. This is an especially important consideration in older adults with a low likelihood of response to immunosuppressive therapy (absence of PNH clone, very severe aplastic anemia, presence of myeloid mutations)
Summary of Evidence:
Traditionally, HCT was recommended for patients ≤40 years with a matched related donor (MRD), while IST was preferred for those >40 years without an MRD. However, recent studies show comparable survival outcomes for patients up to 50 years using MRD or matched unrelated donor (MUD) transplants.
GVHD remains a concern in older adults, with higher chronic GVHD rates but similar acute GVHD rates compared to younger patients. The introduction of PTCy-based GVHD prophylaxis has significantly improved GVHD-free survival.
Implementation Considerations
Upfront HCT should be considered for patients age >40 yr who have a life-expectancy that fits with the HCT goal of providing long-term benefits of disease control plus mitigation of clonal evolution, while outweighing HCT risks by selecting patients with adequate performance status and suitable organ function.
Decision for HCT
Question 3: Should MRD or IST Be Used for Newly Diagnosed Children and Young Adult Patients With SAA?
Recommendation
We recommend upfront MRD-HCT for children and young adults with SAA as it offers high cure rates with a lower risk of disease transformation (Strength of recommendation, Strong; Certainty of evidence; moderate).
Summary of Evidence:
HCT is potentially curative demonstrating >90% OS and failure free survival (FFS) in children with minimal long-term effects on growth, development, and fertility, and risk of progression to MDS or AML is reduced significantly. The European Group for Blood and Marrow Transplantation (EBMT) reported lower 3-year event-free survival (EFS) when IST was compared to MRD-HCT in younger children (33% IST vs. 87% HCT) as well as adolescents (64% IST vs. 83% HCT). Another study examined outcomes for 1488 patients and reported that age <20 yr was among the best predictors of survival.The guideline panel conducted a 1374 patient meta-analysis comparing MRD-HCT to IST and showed significantly better FFS with HCT in comparison to IST; RR 0.30 (95% CI 0.18 to 0.50), I2 83%. OS was assessed for 1973 patients and was similar in both arms; RR 0.91 (95% CI 0.62 to 1.34). (supplemental figures 2A and 2B). In contrast to adults where addition of eltrombopag to IST improved ORR, similar improvements were not reported (p = .836) in children.Implementation Considerations
MRD-HCT is the recommended first-line therapy for children and young adults as it offers high OS and FFS with minimal risk of GVHD. Despite improvement in IST with the incorporation of eltrombopag, complete responses remain low, relapses are common and risk of progression to MDS and AML persists, especially in children with potentially more decades of life ahead.
Question 4: Should MRD-HCT Be Prioritized Over IST for Newly Diagnosed Adults With SAA?
Recommendation
We recommend HCT as a preferred first-line treatment for patients with a MRD
Summary of Evidence:
A Meta-analysis of 15 studies by Zhu et al. showed a superior OS and FFS for patients receiving first line HCT. Superiority of HCT is retained in the era of triple IST, as shown by a recently published prospective study comparing triple IST with MRD-HCT. A systematic review and meta-analysis comparing MRD-HCT with IST subsequent to 2000 showed superior OS and FFS with HCT (Supplemental Figure 3A and 3B). Conversely, patients receiving IST are at increased risk of relapse and clonal evolution usually within 2 to 4 yr of IST. Certainty in the evidence was judged moderate because randomized controlled trials were lacking and data from studies examining OS and FFS were heterozygous.
Implementation Considerations
MRD HCT is a preferred treatment for adults with SAA (Figure 2) and is available at most centers specializing in the treatment of SAA. Centers not offering HCT should refer their patients to centers with expertise
Donor Selection
Question 5: Should HCT be prioritized over IST for children and young adults who lack a MRD?
Recommendation
The panel suggests upfront HCT (either MUD or Haplo-HCT) for children and young adults lacking a MRD
For patients failing the first course of IST, the panel recommends HCT (either MUD or Haplo-HCT) over the second course of IST
Summary of Evidence:
About 70-80% of children and young adults with SAA lack an MRD and require IST or an alternative donor transplant (ADT) like MUD or haplo-HCT. While IST has lower toxicity and 60-70% response rates, it has low complete response, a 30% relapse risk, and a 10-15% chance of clonal evolution to myeloid malignancy. Survival after upfront MUD-HCT has improved, making it comparable to MRD-HCT and superior to IST, leading to its recommendation in UK guidelines. Studies also show that haplo-HCT provides high OS and EFS, with GVHD rates below 10%. ADT is preferred for patients with severe infections, significant bleeding, transfusion dependence, or MDS-related cytogenetic abnormalities. For IST failure or relapse, HCT remains the superior option. Evidence suggests better OS (RR 0.56) and FFS (RR 0.25) in the ADT group.
Implementation Considerations
Despite the absence of randomized data, retrospective and prospective data have shown superior survival, fewer relapses and lower risk of clonal evolution in children and young adults receiving ADT. If a donor can be identified early and patients are treated in centers with expertise in ADT, it is suggested to proceed with upfront MUD or Haplo-HCT. For children and young adults with only a highly mismatched unrelated donor or CBT available, upfront IST is preferred. The panel acknowledges that long term data on fertility using fludarabine, 4 Gy TBI for children and young adults is lacking.
Question 6: Should HCT Remain a Priority for Adults With SAA Who Lack a MRD?
Recommendation
The panel suggests either MUD or haplo-HCT for patients without a MRD
Summary of Evidence
In the US, the probability of finding a MRD is 30% for older adults, but it is 1.5 times lower for those aged 18 to 44. Recent data show improved outcomes with MUD or Haplo-HCT. A meta-analysis of 5 studies (343 patients) found a 5-year OS favoring upfront ADT over IST (OR 0.44) and also superior to salvage ADT (OR 0.31). An EBMT study showed MUD to be non-inferior to MRD HCT. Haploidentical donors are increasingly used in upfront settings with promising results; a BMT CTN trial showed a 1-year OS of 81% for Haplo-HCT. A meta-analysis by Zhao et al. found no difference in OS, FFS, or engraftment outcomes between haploidentical and MRD-HCT, but haplo-HCT showed better 3-year FFS compared to IST. Long-term outcomes with IST are less durable, with FFS generally below 40%. A recent meta-analysis favored upfront ADT over IST for 5-year OS. However, concerns remain around GVHD, graft failure, and infertility. A study from Johns Hopkins showed 92% OS with Haplo-HCT and low risk of GVHD. A meta-analysis by the guideline panel comparing ADT (MUD or Haplo) to IST showed no difference in OS (RR 1.05), but FFS was significantly better with ADT (RR 2.02). The certainty of evidence for OS was low due to data heterogeneity, while FFS evidence was rated moderate due to imprecision.
Implementation Considerations
Our panel recommends the use of ADT (MUD or Haplo) over IST in SAA because of the long-term risk for disease relapse and secondary MDS/AML after IST. Implementation of ADT as a first-line treatment strategy provides the best chance of long-term disease-free survival.
Question 7: Should MUD-HCT Be Prioritized Over Haplo-HCT for Patients Lacking a MRD?
Recommendation
The panel suggests either a MUD or haplo-HCT for patients lacking a MRD with insufficient evidence for prioritization.
Summary of Evidence
Outcomes of MUD and Haplo-HCT have improved over last decade and are comparable to MRD while superior to IST. A recent meta-analysis reported similar OS, FFS and engraftment between haplo-HCT and MRD.
Implementation Considerations
For patients lacking a MRD, either MUD or haplo-HCT may be used, but there is insufficient evidence to prioritize one approach over the other. Practical considerations include the presence of donor specific antibodies (DSA) to the related donor or timing of available donors.
HCT Procedures
Question 8: Should Rabbit ATG or Horse ATG Be Used in Conditioning Regimens for SAA Patients?
Recommendation
The guideline panel recommends rabbit ATG over horse ATG as part of the HCT conditioning regimen for SAA patients receiving HCT.
Summary of Evidence
For the initial IST treatment of SAA, horse ATG (h-ATG) is superior to rabbit ATG (r-ATG). However, for HCT conditioning, r-ATG is preferred based on the results of an 833-patient CIBMTR study where acute and chronic GVHD rates were lower (p < .001) after MRD-HCT for patients who received r-ATG versus h-ATG. Following MUD-HCT there was no difference in cGVHD between the two groups, however, aGVHD was higher (p < .001) in the h-ATG group while OS was lower (p = .02). Rabbit ATG is associated with more effective lymphocyte depletion and enhances the number and function of regulatory T cells likely facilitating tolerance induction and reduction in GVHD. An EBMT study reported improvement in survival and reduction in GVHD with use of ATG regardless of the source of stem cells. The survival benefit of ATG extended to URD as well; 5 year survival of unrelated donor grafts with ATG was 70% vs 52% without ATG.Implementation Considerations
Allo-HCT conditioning for SAA should include ATG regardless of donor type, stem cell source, patient age and gender. When used for HCT preparative regimens, the rabbit ATG formulation is preferred over the horse formulation. Some centers have successfully replaced ATG with alemtuzumab in both MRD and MUD-HCT.
Question 9: Should Fludarabine Containing HCT Conditioning be Prioritized Over Other Regimens for Aplastic Anemia?
Recommendation
The panel recommends cyclophosphamide-ATG conditioning for children and young adults receiving MRD-HCT. Fludarabine containing conditioning is recommended for adults or those with a high-risk of graft failure
Summary of Evidence
There are no randomized controlled trials available comparing conditioning regimens in aplastic anemia. CIBMTR analysis by Bejanyan et al reported similar survival with Cy-ATG and Flu-Cy-ATG. In patients receiving MUD-HCT in the same study, the optimal regimens for HCT were Flu-Cy-ATG and Flu-Cy-ATG+TBI.
Retrospective single-center and multicenter studies have shown the superiority of fludarabine (FLU)-containing conditioning in patients undergoing MRD HCT with 1 or more high-risk factors. A pooled analysis of 3 studies including 1189 patients reported the superiority of fludarabine-containing regimens (OR 0.62; 95% CI 0.41 to 0.92). Overall certainty of evidence was rated as low due to the absence of randomized trials and no large effect. British guidelines recommend fludarabine-containing conditioning for patients >30 yr of age undergoing MRD HCT and for all patients undergoing MUD HCT.
Implementation Considerations
Choice of conditioning regimen depends on factors such as recipient and donor age, donor type, risk factors of graft rejection, comorbidities, center expertise and resource availability. Fludarabine-containing conditioning is suggested for adults >30 yr undergoing MRD-HCT and for any patient receiving MUD or haplo-HCT irrespective of age.
Question 10: Should Bone Marrow Be Prioritized Over Peripheral Blood as Stem Cell Source in Patients With SAA?
Recommendation
For SAA patients undergoing MRD, MUD or Haplo-HCT, our panel recommend using bone marrow as the preferred stem cell source.
Summary of Evidence
Recently published meta-analysis and studies published over last 2 decades have reported lower cumulative incidence of GVHD and better survival outcomes using bone marrow (BM) stem cells. If we consider effects on quality of life (QoL), the peripheral blood (PB) graft source can have large undesirable effects on target population.
For selected patients with serious ongoing infections before HCT, those where the donor only consents for PB collection, or when there are clinical concerns about large donor-recipient weight disparity and the ability to harvest adequate TNC/kg, clinicians may use PB as graft source while accepting a higher risk of GVHD. Use of post-transplant cyclophosphamide may be considered in this situation to lower rates of cGVHD. Evidence shows that BM results in better overall survival (OS) and lower rates of both acute and chronic GVHD compared to PB.
Implementation Considerations
BM is the preferred stem cell source for HCT in SAA irrespective of donor type and results in superior OS, FFS and lower cumulative incidence of GVHD. PB stem cells may be used in selected scenarios and will require enhanced GVHD prophylaxis.
Question 11: What Should be the Preferred Regimen for GVHD Prophylaxis?
Recommendation
The panel suggests either calcineurin inhibitor (CNI) plus methotrexate (MTX) or PTCy based prophylaxis for patients undergoing MRD or MUD HCT.
For patients undergoing Haploidentical HCT, our panel recommends PTCy-based prophylaxis.
Best Practice Statement
MRD and MUD: For CY-ATG/Flu-Cy-ATG conditioning, CNI+MTX has been the standard GVHD prophylaxis. Cyclosporine (CSA) trough levels are recommended to be maintained 200 to 300 ng/mL and tacrolimus (Tac) levels 8 to 12 ng/mL. CSA/Tac should be continued for 6 to 9 mo followed by a linear taper over 3 to 6 mo. CNI-alone GVHD prophylaxis is recommended for protocols incorporating alemtuzumab. Recently uniform conditioning has been proposed employing PTCy regardless of donor type.
Haploidentical: PTCy 50 mg/kg on d +3 and +4, mycophenolate mofetil 15 mg/kg 3 times a day (max dose 3000 mg daily) from d 5 to d 35 and CNI given orally or IV from d 5 to 9 mo, maintaining a trough serum level 8 to 12 ng/mL for Tac and 200 to 300 ng/mL for CSA until a linear taper over 3 to 6 mo may begin in the absence of chronic GVHD.
Cord blood: CSA alone if APCORD regimen used targeting trough concentrations 200 to 400 ng/mL for 3 mo before progressive tapering to stop at 1 year
Implementation Considerations
CNI + MTX remains the standard regimen for GVHD prophylaxis in MRD and MUD-HCT. For Haplo-HCT, PTCy based prophylaxis is preferred. In recent years, PTCy based GVHD prophylaxis for MRD and MUD-HCT has been employed successfully; prospective studies will be needed to determine whether a uniform conditioning and GVHD prophylaxis is efficacious for all SAA patients receiving HCT.
Question 12: How Should Graft Failure and Poor Graft Function Be Managed Post HCT in SAA Patients?
Recommendation
Chimerism assessment using split chimerism on flow-sorted peripheral blood leukocytes is preferred for monitoring. Mixed T-cell chimerism may occur in patients receiving ATG-based conditioning regimens and doesn’t require intervention if blood counts are normal and donor myeloid engraftment is complete. In cases of poor graft function that don’t respond to immunosuppression, a stem cell boost may be necessary. Graft failure typically requires conditioning to remove residual recipient T-cells and may not respond to a boost alone.
Good Practice Statement
Split chimerism is recommended for post-transplant monitoring, and any donor chimerism below 95% should be considered mixed. In aplastic anemia, graft function is defined by the correction of marrow failure, even in the presence of mixed chimerism. Stable mixed T-cell chimerism with normal counts requires only monitoring. Decreasing blood counts with full donor chimerism suggests poor graft function, possibly due to low stem cell viability, which may be improved with increased immunosuppression, erythropoietin, or eltrombopag. If immunosuppression escalation fails, a stem cell boost may be needed. For graft failure or rejection, conditioning is required to eliminate recipient T-cells, and in cases of early rejection, a second HCT with a different donor is recommended.
Limitations of the Guidelines
A major limitation of these guidelines is the lack of randomized trials for many addressed questions. Most evidence comes from non-randomized prospective or retrospective studies, with some questions having low or very low certainty due to imprecision, indirectness, and biases, as assessed by the ROBINS-I tool. The panel’s recommendations are based on current evidence but may change with new findings. Ongoing trials, such as the TransIT and CUREAA trials, are expected to provide more evidence on upfront treatment choices for SAA patients lacking MRD.
The ASTCT guidelines are globally followed by transplant physicians, but regional differences in disease demographics, HCT expertise, donor and drug availability, and financial support exist. These guidelines can be adopted, adapted, or developed anew to create local guidelines based on available evidence, which can be regularly updated.
Conclusion
Improvement in multiple aspects of HCT care have expanded applicability of the procedure to older fit patients with superior outcomes compared to IST. Since the results of MUD and Haplo-HCT have improved over the last decade and outcomes at experienced centers are now comparable to MRD-HCT, alternative donor HCT may be offered in preference to IST for patients lacking a MRD. With attempts to improve outcomes utilizing uniform and novel conditioning and optimal GVHD prophylaxis, along with better understanding of approaches to treat and prevent graft failure, transplant outcomes are likely to improve in patients with SAA. Ongoing prospective trials are likely to further clarify treatment algorithms for pediatric and adult patients when a MRD is not available.
References
1. SB Killick, N Bown, J Cavenagh, et al.
Guidelines for the diagnosis and management of adult aplastic anaemia
Br J Haematol, 172 (2) (2016), pp. 187-207
2. A Kulasekararaj, J Cavenagh, I Dokal, et al.
Guidelines for the diagnosis and management of adult aplastic anaemia: a British Society for Haematology Guideline
Br J Haematol, 204 (3) (2024), pp. 784-804
3. SL Norris, LJA Bero
GRADE methods for guideline development: time to evolve?
Am Coll Phys (2016), pp. 810-811
4. Schünemann HJ, Al-Ansary LA, Forland F, et al. Guidelines International Network: principles for disclosure of interests and management of conflicts in guidelines. 2015;163(7):548-53.
5. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. 2008;336(7650):924-6.
6. P Alonso-Coello, HJ Schünemann, J Moberg, et al.
GRADE evidence to decision (EtD) frameworks: a systematic and transparent approach to making well informed healthcare choices. 1: Introduction
BMJ, 353 (2016), p. i2089
7. Aguayo-Albasini JL, Flores-Pastor B, Soria-Aledo VJCE. GRADE system: classification of quality of evidence and strength of recommendation. 2014;92(2):82-8.
8. P Ahmed, QuN Chaudhry, TM Satti, et al.
Epidemiology of aplastic anemia: a study of 1324 cases
Hematology, 25 (1) (2020), pp. 48-54
9. AE DeZern, JEJBA. Churpek
Approach to the diagnosis of aplastic anemia
Blood Adv, 5 (12) (2021), pp. 2660-2671
10. M. Park
Overview of inherited bone marrow failure syndromes
Blood Res, 57 (S1) (2022), pp. 49-54
11. GE Georges, K Doney, R. Storb
Severe aplastic anemia: allogeneic bone marrow transplantation as first-line treatment
Blood Adv, 2 (15) (2018), pp. 2020-2028
12. A Bacigalupo, S. Giammarco
Haploidentical donor transplants for severe aplastic anemia
Semin Hematol, 56 (3) (2019), pp. 190-193
13. A Bacigalupo, G Socié, RM Hamladji, et al.
Current outcome of HLA identical sibling versus unrelated donor transplants in severe aplastic anemia: an EBMT analysis
haematologica, 100 (5) (2015), p. 696
14. QUN Chaudhry, R Iftikhar, TM Satti, et al.
Outcome of Fludarabine-based conditioning in high-risk aplastic anemia patients undergoing matched related donor transplantation: a single-center study from Pakistan
Biol Blood Marrow Transplant, 25 (12) (2019), pp. 2375-2382
15. JP Maciejewski, AM. Risitano
Aplastic anemia: management of adult patients
Hematol Am Soc Hematol Educ Program (1) (2005), pp. 110-117
