Introduction

Inherited bone marrow failure (BMF) syndromes (IBMFSs) are a heterogeneous group of genetic disorders characterized by BMF, disease-specific symptoms, and predisposition to malignancies. IBMFSs account for 10–20% of all childhood BMFs, and differential diagnosis between IBMFSs and acquired BMFs is critical for determining optimal treatment approaches including hematopoietic cell transplantation (HCT) [1,2,3]. Classically, the clinical diagnosis of IBMFSs is based on physical findings and disease-specific tests. The recently implemented comprehensive genetic analyses are efficacious not only for the definitive diagnosis of IBMFSs, but also for the differential diagnosis between IBMFSs and acquired BMFs [4,5,6]. Furthermore, advances in molecular genetics have led to the identification of novel disorders such as AMeD (aplastic anemia, mental retardation, and dwarfism) and MIRAGE (myelodysplasia, infection, restriction of growth, adrenal hypoplasia, genital phenotypes, and enteropathy) syndromes while improving our understanding of the pathogenesis underlying IBMFSs (Fig. 1) [7, 8]. A summary of the diagnostic approaches utilized for classical IBMFSs is presented in Fig. 2. Albeit a promising treatment option to overcome BMF in patients with IBMFS HCT does not correct nonhematological defects and may enhance the risk of secondary malignancies due to the inherited predisposition to cancer. Disease-specific HCT should be proposed, given that IBMFSs differ in underlying defects and susceptibility to toxicities of drugs or irradiation [9, 10]. In this review, we summarize current practices in HCT for IBMFSs, including indications for transplantation, selection of the donor source and conditioning regimens, and management during and after HCT, based on latest data from recent studies.

Fig. 1
figure 1

Schematic summary of the pathogenesis underlying inherited bone marrow failure syndromes. Due to the genetic defects in DNA repair mechanisms associated with Fanconi anemia (FA), cells are unable to properly repair DNA damage, which leads to genomic instability and increased susceptibility to cytotoxic agents, and predisposition to malignancies. Defects in telomere maintenance mechanisms in dyskeratosis congenita (DC) cause telomere shortening and induce apoptosis/senescence and carcinogenesis. Aberrant ribosomal protein biosynthesis in Diamond–Blackfan anemia (DBA) initiates apoptosis of erythroid progenitor cells via the activation and stabilization of p53, resulting in pure red cell aplasia. Shwachman-Diamond syndrome (SDS) is caused by mutations in SBDS, which encodes a protein involved in normal ribosome function and centrosome amplification. Severe congenital neutropenia (SCN) is caused by impaired maturation of granulocytes due to mutations in ELANE, HAX1, GFI1, or other genes associated with the expression, folding, processing, secretion, or degradation of neutrophil elastase in myeloid cells. Dysfunction of the thrombopoietin (TPO) receptor, encoded by c-MPL, causes defects in TPO-driven signaling that is essential for thrombopoiesis as well as hematopoietic stem cell homeostasis, resulting in amegakaryocytosis and pancytopenia in congenital amegakaryocytic thrombocytopenia (CAMT). MIRAGE syndrome is caused by gain-of-function mutations in SAMD9/SAMD9L encoding an endosome facilitator protein downregulating cellular growth, resulting in severe growth restriction. Chr chromosome, ER endoplasmic reticulum, GB Golgi body, Lys lysosome, Mit mitochondria

Full size image
Fig. 2
figure 2

Approach for the evaluation of classical inherited bone marrow failure syndromes. Evaluation for inherited bone marrow failure syndromes (IBMFSs) should be considered for all pediatric, adolescent, and young adult patients with bone marrow failure (BMF) and myelodysplastic syndrome. The pattern of cytopenia in IBMFSs is varied, and specific physical malformations and family history can be clues for the diagnosis. Each IBMFS requires specific screening tests. In Fanconi anemia (FA), increased chromosomal breakage is observed using the chromosome fragility test. Evaluation of telomere length in peripheral blood lymphocytes by flow fluorescence in situ hybridization is useful as an initial screening for dyskeratosis congenita (DC). Elevated activity of erythrocyte adenosine deaminase (eADA) and erythrocyte reduced glutathione (eGSH) are utilized as biomarkers for Diamond–Blackfan anemia (DBA), although a history of blood transfusion may affect the test results. Bone marrow (BM) examination usually shows normocellular and nearly absent erythroid progenitors. Screening for trypsinogen and pancreatic isoamylase can reveal pancreatic defects, and imaging studies may show pancreatic lipomatosis, in patients with Shwachman–Diamond syndrome (SDS). BM examination for severe congenital neutropenia (SCN) typically shows granulocyte maturation arrest at the level of promyelocytes. Elevated serum levels of thrombopoietin (TPO) and the absence of megakaryocytes in BM are specific for congenital amegakaryocytic thrombocytopenia (CAMT). Genetic tests using target and whole exome sequencing have recently become available for the definitive diagnosis of IBMFS. In parallel, evaluation for the indications for hematopoietic cell transplantation (HCT) and preparation for HCT, including human leukocyte antigen (HLA) testing and donor search, are necessary

Full size image

Fanconi anemia

Fanconi anemia (FA), which is the most common IBMFS, is often accompanied by a variety of congenital abnormalities, including short stature, microcephaly, skin pigmentation, heart defects, abnormalities in radius and thumb, and genital defects. The characteristic FA phenotype is progressive pancytopenia, which may transform to myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML), and predisposition to other malignancies [11]. Chromosome fragility test, which detects genomic instability due to genetic defects in DNA repair mechanisms, is useful as a simple screening tool for the diagnosis of FA and should be adopted for all patients with BMF including children, adolescents, and young adults.

Indications for HCT in FA include BMF due to severe aplastic anemia and MDS/AML. All patients should undergo annual marrow evaluation to detect early signs of clonal evolution and are recommended to receive HCT before the development of MDS/AML. In patients with FA, special considerations regarding conditioning regimens, such as dose reduction of alkylating agents and irradiation, are necessary due to the increased susceptibility to associated toxicities. The outcomes of HCT for FA have dramatically improved following the introduction of fludarabine-based reduced-intensity conditioning (RIC). Although irradiation is generally avoided in patients with FA because of the underlying cancer predisposition, low-dose irradiation is acceptable in cases of high risk for graft failure such as frequent transfusion history, unrelated donors, and cord blood transplantation (CBT). Currently used conditioning regimens usually include fludarabine and reduced doses of cyclophosphamide (total dose up to 50 mg/kg) with or without low-dose total body irradiation (TBI; 1–3 Gy) and/or antithymocyte globulin (ATG) [12,13,14,15,16]. In a study including 795 patients with FA, the European Society for Blood and Marrow Transplantation (EBMT) demonstrated that age < 10 years at the time HCT, transplantation before clonal evolution, transplantation from a matched related donor (MRD), and fludarabine-based RIC were associated with favorable outcomes [13]. Although unaffected MRDs are preferred, bone marrow transplantation (BMT) from a matched unrelated donor (MUD) following fludarabine-based RIC is a good alternative with an overall survival (OS) rate of 60–80% (Table 1) [14,15,16]. MacMillan et al. reported that patients without a history of opportunistic infections or frequent transfusions who underwent HCT from an MUD following fludarabine-based RIC with ATG had an excellent 5-year OS of 94% [14].

Table 1 Long-term outcomes after hematopoietic cell transplantation for inherited bone marrow failure syndromes
Full size table

Bone marrow is the best stem cell source, and the use of peripheral blood stem cells is not recommended due to the increased risk of chronic graft-versus-host disease (GVHD) and secondary malignancies. Since the report of the first successful human leukocyte antigen (HLA)-identical sibling CBT for a 5-year-old boy with FA by Gluckman et al., CBT has been widely adopted as a treatment option for a variety of diseases [17]. Several retrospective studies showed that unrelated CBT in FA was associated with a high rate of graft failure, whereas a recent study from University of Minnesota described that the rate of engraftment with unrelated CBT using fludarabine-based RIC with 3-Gy TBI was comparable to that with MRD/MUD-BMT [14, 18]. As another potential stem cell source for patients with FA, the EBMT-Severe Aplastic Anemia Working Party (SAAWP) reported promising outcomes in patients who underwent haploidentical HCT with in vivo T-cell depletion, with a 2-year event-free survival (EFS) of 86% [19]. A study from Italy reported that haploidentical HCT with ex vivo depletion of T-cell receptor α/β+ and CD19+ cells was successfully performed in patients with FA, with a 5-year EFS of 86% [20]. These results suggest unrelated CBT and haploidentical HCT as promising strategies; however, these procedures should be performed in experienced centers [14,15,16, 21]. In contrast, outcomes have been poor in patients with FA undergoing HCT after progression to MDS/AML. A retrospective analysis by the EBMT revealed that the 5-year OS after HCT was only 42% in patients with FA that transformed to MDS/AML [22]. A recent study from the Japanese Society of Transplantation and Cellular Therapy has also reported that these patients had poor prognosis: the 5-year OS rates, which reached 89% in patients with aplastic anemia, were only 71 and 44% in those with MDS and acute leukemia, respectively [16].

Several studies have shown that chronic GVHD is highly associated with increased risk of secondary malignancies and late mortality in patients with FA, highlighting the need for the optimization of GVHD prophylaxis for HCT in these patients [9, 13, 16]. A recent study from the UK reported that fludarabine-based RIC with alemtuzumab provided excellent OS and chronic GVHD-free EFS in patients with FA undergoing HCT, suggesting the specific benefit of alemtuzumab for these patients [23]. Frequently observed malignancies after HCT in FA include head and neck squamous cell carcinoma (HNSCC) and gynecological cancers [13, 16, 24]. Yabe et al. investigated the long-term incidence of secondary malignancies in 137 patients with FA who survived more than 1 year after HCT [16]. The authors reported that secondary malignancies were observed in 15 patients at a median of 8 (range 1–15) years after HCT and that 12 of the 15 patients developed HNSCC; irradiation-containing regimens and older age at HCT were found as risk factors for HNSCC. Long-term follow-up by multidisciplinary teams with careful monitoring for head and neck malignancies, gynecological checkups, and screening for breast cancer are essential given that more patients with FA survive into adulthood [9].

AMeD syndrome

AMeD syndrome, defined by the presence of aplastic anemia, mental retardation, and dwarfism, is a novel FA-like IBMFS caused by digenic mutations in ALDH2 and ADH5, which lead to an increase in cellular formaldehyde sensitivity and multisystem abnormalities including impaired hematopoiesis [8]. A study by Oka et al., which provided a detailed description of the clinical characteristics of ten patients with AMeD syndrome, reported that chromosome fragility was not observed in any of the patients with AMeD syndrome despite the close similarity in physical appearance between the patients with AMeD syndrome and those with FA [8]. In that study, six and one patient developed MDS and AML, respectively, three patients with MDS and one patient with AML received allogeneic HCT, and three of the four patients achieved long-term survival. In general, patients with AMeD syndrome, unlike those with FA, can successfully undergo HCT using the RIC regimen without special considerations such as dose reduction of alkylating agents. Further studies are warranted to optimize the HCT conditioning regimen in patients with AMeD syndrome.

Dyskeratosis congenita

Dyskeratosis congenita (DC) is a rare multisystemic disorder and IBMFS characterized by the classical triad of reticular skin pigmentation, nail dystrophy, and oral leukoplakia; patients with DC can also exhibit a variety of other clinical manifestations including BMF, pulmonary and hepatic fibrosis, gastrointestinal abnormalities, and predisposition to cancers, especially HNSCC and MDS/AML [25, 26]. The phenotypic spectrum of DC is wide, ranging from the mild cryptic type to the severe Hoyeraal–Hreidarsson and Revesz syndromes. Defects in telomere maintenance, which results in very short telomeres, is the main pathogenic mechanism underlying DC [27, 28]. Therefore, all pediatric, adolescent, and young adult patients with BMF should be screened using lymphocyte telomere length measurement to rule out telomere biological disorders.

In patients with DC, BMF is the main cause of mortality and allogeneic HCT is a curative treatment option for patients with unmanageable BMF. HCT cannot improve systemic organ dysfunction associated with telomeropathies and can even enhance the risk of organ dysfunction and secondary malignancies; therefore, the indications for HCT in patients with DC require careful evaluation. Regarding donor selection, an unaffected MRD is the best option, although telomere length evaluation is necessary to eliminate the possibility of cryptic DC. MUD-HCT is also acceptable, but mismatched unrelated donor-HCT, haploidentical-HCT, and unrelated CBT are experimental options [29]. Given the presence of fragile tissue that is hypersensitive to irradiation and chemotherapy, myeloablative conditioning (MAC) regimens are not suitable for patients with DC. The increasingly used fludarabine-based RIC regimens provide better short-term results but do not contribute to long-term survival [29,30,31,32,33,34]. Fioredda et al. reported the transplantation outcomes of 94 patients with DC who received RIC regimens (42% with fludarabine-based RIC and 83% with serotherapy including ATG or alemtuzumab) [29]. The 5-year OS rate was 59%, and the OS curve showed a continuous decline over time, with a 10-year OS estimate of less than 30%. Notably, organ damage including the lungs, liver, and the gastrointestinal system as well as secondary malignancies were observed even a decade after HCT. Based on these data, HCT is not always the best option for DC and long-term surveillance should be carefully performed in patients with DC undergoing HCT (Fig. 3).

Fig. 3
figure 3

Secondary malignancies and organ dysfunction after hematopoietic cell transplantation in patients with Fanconi anemia and dyskeratosis congenital. In addition to bone marrow failure, patients with Fanconi anemia (FA) and dyskeratosis congenita (DC) present with characteristic malformations and harbor unique cancer predispositions. Complications reflecting these characteristics can also occur after hematopoietic cell transplantation and should be carefully followed-up. AL acute leukemia, HNSCC head and neck squamous cell carcinomas, MDS myelodysplastic syndrome

Full size image

Danazol, a synthetic sex hormone, is converted to estrogen in cells and increases telomerase activity [35, 36]. The results of a phase I/II trial on danazol for patients with DC conducted by the National Institutes of Health demonstrated the presence of hematologic response in 79% and 83% of the patients at 3 and 24 months, respectively, and suggested that danazol could reduce the risk of pulmonary fibrosis [37]. A phase II trial on danazol for short telomere-related pulmonary fibrosis is currently underway (NCT04638517) [38].

Diamond–Blackfan anemia

Diamond–Blackfan anemia (DBA) is an almost autosomal dominant IBMFS, characterized by neonatal or infantile anemia with erythroblastopenia, possible congenital malformations (e.g., short stature, cleft palate, heart defects, and urogenital anomalies), and predisposition to MDS/AML or solid tumors, which are caused by aberrant ribosome biogenesis [39,40,41].

Red blood cell transfusions are the main treatment for patients with DBA during the first year of life, and steroid therapy is recommended for later ages; responses to steroids are observed in approximately 60% of treated patients [39, 42]. Allogeneic HCT represents the only cure for the hematopoietic impairment and should be proposed to patients with transfusion-dependent, no response to steroids, steroid dependence (above 0.3 mg/kg/day), progression to severe pancytopenia, or MDS/AML [41,42,43]. Transfusion-related iron overload can cause multiple organ damages and increase the risk of transplant-related mortality [43]. To avoid transfusion-related iron overload, adopting HCT before the age of 10 years, preferably between 2 and 5 years, seems better.

Recent reports have described improved outcomes of HCT for DBA with long-term survival rates of more than 90% (Table 1) [44, 45]. An analysis conducted by the EBMT-SAAWP involved the largest cohort of 106 patients with DBA who underwent HCT and showed similar outcomes between MRD and MUD transplants [46]. Bone marrow remains the cell source of choice, and related CBT is also a good option. The Eurocord and EBMT-SAAWP reported an excellent outcome of related CBT with a 3-year OS of 95% in 20 patients with IBMFS, including 13 patients with DBA [30]. Additionally, a more recent Japanese study has suggested promising results of unrelated CBT for DBA; the 5-year OS and EFS rates of 11 patients with DBA who received unrelated CBT were 100 and 82%, respectively [21]. Most patients have received HCT with MAC regimens for this disorder; however, recent data have shown comparable outcomes between MAC and RIC regimens, suggesting that HCT using RIC is a promising option for young patients with DBA [45]. Long-term follow-up for patients with DBA who received HCT should focus on iron overload and the development of secondary malignancies, particularly osteosarcoma and colorectal cancer.

Severe congenital neutropenia

Severe congenital neutropenia (SCN) is a group of inherited disorders characterized by profound neutropenia caused by impaired maturation of granulocytes in bone marrow due to mutations of the ELANE, HAX1, GFI1, or other genes [47]. Although the use of granulocyte colony-stimulating factor (G-CSF) improved the prognosis of this disorder, close attention should be paid to malignant transformation if high doses of G-CSF are used; the cumulative incidence of MDS/AML is 11–22% after 15–20 years of G-CSF treatment [48, 49].

Indications for HCT in SCN include absent or poor (requiring more than 10 μg/kg/day) response to G-CSF and MDS/AML. Recently, a report from the French SCN registry has suggested that early HCT reduces the risk of clonal evolution in patients requiring high doses of G-CSF [50]. The largest study from the EBMT reported the outcomes of 136 patients with SCN (87 with MDS/AML) who underwent HCT between 1990 and 2012 (Table 1) [51]. The 3-year OS rate was 82%, and HCT at a younger age (< 10 years old), in recent years, and from MRD or MUD was associated with significantly better OS. Focusing on stem cell sources, BMT and CBT provided favorable outcomes; the 3-year OS and EFS rates were 85 and 76%, respectively, after BMT and 92 and 70%, respectively, after CBT. In contrast, the use of peripheral blood stem cells was significantly associated with an increased incidence of chronic GVHD, suggesting that it is not recommended for HCT in SCN [51]. Thus far, MAC regimens have been preferably used for this disorder; however, no difference in outcomes has been demonstrated between MAC and RIC regimens. Less toxic fludarabine-based RIC regimens might reduce early and late morbidity and mortality; however, more experience is needed to draw any conclusion.

Shwachman–Diamond syndrome

Shwachman–Diamond syndrome (SDS) is an autosomal recessive disorder characterized by neutropenia with 10–30% cumulative incidence of progression to MDS/AML, accompanied with exocrine pancreatic dysfunction, metaphyseal dysplasia, mental retardation, cardiomyopathy, and immune dysfunction [52,53,54]. The pathogenic mutations in SBDS cause abnormal ribosome synthesis and inadequate maintenance of the stromal microenvironment [54, 55].

Indications for HCT in SDS include progressive cytopenia and MDS/AML. HCT before the development of MDS/AML is highly recommended because of poor prognosis in patients with malignant transformation. A retrospective analysis of 74 patients with SDS from the EBMT found that the 5-year OS after HCT was only 29% in patients who developed AML but was 71% in those with BMF [56]. Similar results were also reported by Myers et al., who analyzed a cohort of 52 patients with SDS and found that only 2 of 13 patients (15%) who underwent HCT for MDS/AML were alive at 5 years compared to a 5-year OS of 72% for patients who underwent HCT during BMF [57]. In both studies, the main causes of death were graft failure and GVHD in patients with SDS-related BMF and relapse in those with MDS/AML-transformed SDS. The type of conditioning regimen for SDS remains an open question that should be addressed, although RIC is more commonly indicated and a difference in outcomes between RIC and MAC has not been observed in patients with BMF.

Congenital amegakaryocytic thrombocytopenia

Congenital amegakaryocytic thrombocytopenia (CAMT) is a rare IBMFS characterized by markedly decreased or absent megakaryocytes in bone marrow caused by a mutation in c-MPL, which encodes the thrombopoietin receptor. Patients with CAMT usually present with thrombocytopenia at birth or within the first year of life and subsequently develop pancytopenia [58, 59]. HCT represents the only curative treatment for CAMT, and the indications include transfusion-dependent BMF and clonal evolution. To date, the EBMT and the Center for International Blood and Marrow Transplant Research reported two large retrospective studies on HCT for patients with CAMT (Table 1), in which the 5-year OS rates were 77 and 86%, respectively; in both studies, MAC regimens were frequently used [60, 61]. The EBMT study observed no differences in transplantation outcomes based on age, sex, year of transplantation, HLA disparity, and stem cell source, whereas the study from the Center for International Blood and Marrow Transplant Research demonstrated that HLA disparity was a poor prognostic factor [60, 61].

BMF associated with SAMD9/SAMD9L mutations including MIRAGE syndrome

Comprehensive genetic analyses have identified novel germline mutations in SAMD9/SAMD9L as causative aberrations of MIRAGE syndrome, which includes myelodysplasia, infection, restriction of growth, adrenal hypoplasia, genital phenotypes, and enteropathy. The mutations were detected in 8%–19% of patients with a suspicious IBMFS or childhood MDS [4, 7, 62,63,64]. Loss of the mutated allele via monosomy 7 or deletion 7q was reported in patients with SAMD9/SAMD9L mutations who progressed to MDS/AML [7]. HCT is generally indicated for patients who harbor these mutations and develop BMF or clonal evolution; these patients are considered to be at high risk of complications after HCT [65, 66]. Ahmed et al. reported the outcomes of HCT in 12 patients with hematologic disorders associated with mutations in SAMD9/SAMD9L, including 4 patients with MIRAGE syndrome. Although 10 of the 12 patients survived with a median follow-up of 3.1 (range 0.1–14.7) years, most of the patients experienced serious transplantation-related complications such as cardiorespiratory distress, severe infections, thrombotic microangiopathy, and sinusoidal obstruction syndrome [65]. In addition, the risk of serious infections due to the lack of polyclonal T-cell reconstitution even after HCT has been reported; thus, the indication for HCT should be carefully evaluated in patients with severe MIRAGE syndrome [66].

Conclusion

IBMFSs include a wide spectrum of disorders, with varied pathogenic mechanisms and extramedullary abnormalities resulting in differing levels of risk for progression to MDS/AML and development of acute or late complications after HCT (Fig. 3). HCT can only correct impaired hematopoiesis; therefore, nonhematological as well as hematological findings should be carefully evaluated during the pretransplantation decision-making process. General indications for HCT in IBMFSs include transfusion dependence and severe neutropenia. For patients at risk of clonal evolution, frequent marrow evaluation should be performed with consideration of HCT before the development of MDS/AML. The most commonly preferred donor is a healthy MRD, and bone marrow is the best source of stem cells. Accumulating evidence shows that MUD-BMT or related CBT is a good option in patients without suitable MRDs. Peripheral blood stem cell transplantation is associated with increased risk of chronic GVHD and secondary malignancies in patients with IBMFSs; therefore, its use should not be encouraged. Unrelated CBT and haploidentical HCT are promising strategies, although these should be performed in experienced centers. Conditioning regimens must be adopted depending on the type of IBMFS, and fludarabine-based RIC provides better long-term outcomes, especially in patients with FA or DC. Irradiation should be avoided because of the known cancer predisposition in patients with IBMFSs. Long-term follow-up is essential to detect complications related to the IBMFS or HCT. In addition, age-appropriate patient education and transition from pediatrics to adult medicine are important issues that should be addressed in future clinical and studies.