Treatment of acquired aplastic anemia in children and adolescents

INTRODUCTION — Acquired aplastic anemia (AA) is a bone marrow (BM) failure syndrome characterized by multilineage cytopenias and severely hypocellular BM, without infiltrative disease of the bone marrow [1,2]. AA is caused by immune-mediated destruction of pluripotent hematopoietic stem and progenitor cells. Acquired AA must be distinguished from other causes of pancytopenia, including inherited BM failure syndromes (IBMFS), which are caused by inherited gene variants and may be associated with extra-hematopoietic manifestations and high risk for development of acute myeloid leukemia, myelodysplastic syndromes, and other cancers. Management of AA is distinctly different from IBMFS and other causes of hypoplastic/aplastic marrow.

This topic discusses pretreatment evaluation, management, and outcomes with acquired AA in children and adolescents.

Evaluation, diagnosis, and differential diagnosis of AA are discussed separately. (See "Aplastic anemia: Pathogenesis, clinical manifestations, and diagnosis".)

Treatment of AA in adults is discussed separately. (See "Treatment of aplastic anemia in adults".)

PRETREATMENT

Diagnosis — Acquired AA should be suspected in children who present with clinical findings associated with pancytopenia, such as pallor, fatigue, easy bruisability or bleeding, and/or infections.

The diagnosis of AA is based on pancytopenia and absolute reticulocytopenia (frequently <40,000/microL) together with bone marrow hypocellularity (picture 1), normal cellular morphology, and no abnormal infiltrates, as discussed in greater detail separately. (See "Aplastic anemia: Pathogenesis, clinical manifestations, and diagnosis", section on 'Diagnostic criteria'.)

Exclude other causes — It is important to exclude other causes of cytopenias/marrow aplasia in children and adolescents. Treatment of AA requires potentially toxic treatments that are distinct from those used to manage other causes of cytopenias. However, severe AA can cause life-threatening complications, and treatment should not be delayed while awaiting all results from testing for other causes of cytopenias.

Examples of conditions that should be excluded in children and adolescents with AA include:

●Inherited bone marrow failure syndromes (IBMFS) – Inherited disorders should be excluded at the time of diagnosis of AA. Initial evaluation for IBMFS by clinical evaluation and functional screening tests is followed by genetic testing, as described below. (See 'Clinical/laboratory evaluation' below.)

IBMFS may be associated with hematopoietic disorders and extra-hematopoietic findings. The cytopenias may be mild or of gradual onset, but they may also be associated with a hematologic malignancy. Other findings may include skeletal or urogenital anomalies, cutaneous findings (eg, premature hair graying), unexplained liver or lung dysfunction/fibrosis, and other findings.

Management of children with IBMFS is distinct from that for AA, as children are at risk for complications from organ dysfunction, excessive sensitivity to chemotherapy and other treatments, and they require specific surveillance for cancer and other effects. (See "Hematopoietic cell transplantation (HCT) for inherited bone marrow failure syndromes (IBMFS)".)

Examples of clinical presentations of IBMFS follow; note that some findings may be absent from the child who presents with AA, but also from their seemingly unaffected potential transplant donors:

•Short stature, café-au-lait spots, skeletal or urogenital anomalies suggestive of Fanconi anemia (FA). (See "Clinical manifestations and diagnosis of Fanconi anemia", section on 'Clinical features'.)

•Early graying of hair, unexplained pulmonary fibrosis or hepatic cirrhosis suggestive of a telomere biology disorder (eg, dyskeratosis congenita). (See "Dyskeratosis congenita and other telomere biology disorders", section on 'Clinical features'.)

•Persistent warts and infections suggestive of GATA2 deficiency. (See "Familial disorders of acute leukemia and myelodysplastic syndromes", section on 'Familial MDS/AML with mutated GATA2'.)

•Intestinal disease, lymphadenopathy, infections, autoimmunity suggestive of CTLA4 deficiency. (See "Autoimmune lymphoproliferative syndrome (ALPS): Clinical features and diagnosis", section on 'CTLA-4 haploinsufficiency with autoimmune infiltration (CHAI) disease'.)

•Other conditions that may cause pancytopenia and marrow aplasia include Shwachman-Diamond syndrome, Diamond-Blackfan anemia, and syndromes associated with inherited pathogenic variants of SAMD9/SAMD9L and MPL. (See "Clinical manifestations and diagnosis of Fanconi anemia" and "Dyskeratosis congenita and other telomere biology disorders" and "Familial disorders of acute leukemia and myelodysplastic syndromes".)

●Hematopoietic malignancies – Hematopoietic malignancies may present with clinical manifestations of cytopenias, but they are generally associated with findings in the blood smear, bone marrow, or physical examination.

Examples include:

•Acute lymphoblastic leukemia/lymphoma (ALL) – (See "Overview of the clinical presentation and diagnosis of acute lymphoblastic leukemia/lymphoma in children".)

•Acute myeloid leukemia (AML) – (See "Acute myeloid leukemia in children and adolescents".)

•Myelodysplastic syndromes (MDS) – (See "Clinical manifestations, diagnosis, and classification of myelodysplastic syndromes (MDS)".)

●Hemophagocytic lymphohistiocytosis (HLH) – HLH presents with cytopenias and systemic illness and may be caused by inherited mutations or in association with infection, malignancy, and other causes. (See "Clinical features and diagnosis of hemophagocytic lymphohistiocytosis".)

Clinical/laboratory evaluation — Pretreatment clinical and laboratory evaluation is needed to exclude other causes of cytopenias in children (above) and to classify the severity of AA (below) (see 'Classification of severity' below):

●History and physical examination – Personal or family history of cytopenias, cancers, somatic anomalies, unexplained organ dysfunction.

Medications (eg, chemotherapeutic agents, anti-seizure medications, nonsteroidal anti-inflammatory drugs [NSAID], anti-thyroid medications), ionizing radiation, chemical exposure (eg, benzene, solvents), hepatitis, HIV, other viruses, and other causes (table 1) should be documented.

●Laboratory testing

•Complete blood count with differential

•Reticulocyte count

•Liver function tests, including serum aminotransferases, lactate dehydrogenase

•Serum folate and vitamin B12

•Viral serologies – HIV, cytomegalovirus, Epstein-Barr virus, herpes simplex, hepatitis screen

●Blood smear – Microscopic examination usually reveals normocytic red blood cells (RBCs) and reduced numbers of white blood cells and platelets that are morphologically normal. RBCs may have macrocytic features.

●Flow cytometry of peripheral blood to evaluate for a paroxysmal nocturnal hemoglobinuria (PNH) clone (ie, CD55/59 deficient leukocytes)

●Bone marrow

•Microscopy – Bone marrow examination to confirm the diagnosis of AA and to exclude other disorders should include microscopic assessment of bone marrow cellularity, evidence of dysplasia, and presence/absence of excess blasts, abnormal infiltrates, or fibrosis.

•Cytogenetic and molecular evaluation – Karyotype (chromosomal analysis), fluorescence in situ hybridization (FISH), single nucleotide polymorphism (SNP) array, and next generation sequencing (NGS) panel are performed to assess clonal evolution that may be associated with AA. The SNP array can evaluate for copy neutral loss of heterozygosity at chromosome 6p (6p CN-LOH) and NGS panel should evaluate acquired gene mutations that are commonly associated with AA (eg, BCOR/BCORL1, ASXL1, RUNX1) [3,4].

●Functional screening for other causes of AA should include (table 2):

•Chromosome stress (breakage) test with diepoxybutane (DEB)/mitomycin C (MMC) to rule out Fanconi anemia.

•Telomere length analysis of leukocyte subsets to rule out dyskeratosis congenita and other telomere biology disorders.

●Genetic testing – We perform germline genetic testing to exclude an IBMFS in all children with new onset of multilineage cytopenias and bone marrow hypocellularity who do not have a PNH clone (ie, CD55/59 deficient leukocytes by flow cytometry) or 6p CN-LOH. Detection of a PNH clone or 6p CN-LOH was shown to have 100 percent specificity for acquired AA [4], germline genetic testing is not required for patients with these findings of clonality.

•Test non-hematopoietic tissue, rather than blood – Germline genetic testing should be performed using cultured skin fibroblasts, because testing for germline mutations using blood or marrow risks misidentifying an acquired somatic mutation as a germline variant.

•Gene panel or whole exome sequencing – Next generation sequencing of a targeted gene panel (for known causes of IBMFS) or whole exome sequencing can identify a germline cause for an IBMFS. Because there are >50 inherited conditions that may cause cytopenias, we do not delay treatment of patients with severe AA while awaiting completion of germline genetic testing.

Genetic testing is particularly valuable for children with moderate and/or chronic pancytopenia and for those in whom immunosuppressive therapy has failed to improve the cytopenias. Genetic testing is less likely to be positive in patients with severe pancytopenia who do not have a suggestive family history, clinical features, or extra-hematopoietic anomalies suggestive of an IBMFS.

Transplant eligibility — Eligibility for hematopoietic cell transplantation (HCT) requires adequate medical fitness, including cardiac, liver, and lung function and other considerations. We refer patients with AA for consultation with transplant specialists soon after the diagnosis is established, regardless of the severity of disease.

Eligibility for allogeneic HCT is discussed separately. (See "Determining eligibility for allogeneic hematopoietic cell transplantation".)

Donor search — Human leukocyte antigen (HLA) testing of the child and siblings should be performed promptly after the diagnosis of AA is established. Considerations for donor selection are discussed separately. (See "Donor selection for hematopoietic cell transplantation".)

●Matched related/sibling donor – Only one-quarter of full biological siblings are expected to be a full HLA-match with the patient and there is little likelihood of other relatives being a full match (in the absence of consanguinity or parents sharing common HLA haplotypes).

If multiple HLA-matched siblings are identified, selection should favor a donor in good health, of similar weight and age as the patient (for an adequate graft dose), of the same sex (to lessen immunologic mismatch) and cytomegalovirus (CMV) serologic status of donor and recipient.

●Unrelated donor (UD) – If there is no suitable HLA-matched sibling, we promptly begin a search for a UD in a donor registry, regardless of the severity of AA. The likelihood of finding an optimal donor varies among racial and ethnic groups [5]. (See 'Classification of severity' below.)

CLASSIFICATION OF SEVERITY — All patients with a diagnosis of AA require classification of disease severity to guide treatment and assess prognosis.

Severity of AA is determined by the level of cytopenias and the extent of hypocellularity, as follows:

●Moderate AA (MAA) is defined by fulfilling both of the following findings:

•Bone marrow that is hypocellular for age (typically <50 percent cellularity).

•Stable or worsening decrease in at least one cell line for several weeks, if accompanied by bone marrow hypocellularity in multiple cell lineages, documented by:

-Absolute neutrophil count (ANC) <1500/microL (calculator 1)

-Platelet count <150,000/microL

-Anemia with absolute reticulocyte count (ARC) <60,000/microL

•Does not meet criteria for severe AA (SA), as described below.

The outcome for patients with MAA is highly variable. Many patients may exhibit spontaneous recovery, some maintain a diagnosis of MAA for months to years, while others progress to SAA. In one series, 16 of 24 children with MAA progressed to SAA at a median follow-up of 66 months [6].

Management of MAA is described below. (See 'Moderate AA (MAA)' below.)

●Severe AA (SAA) – Criteria for SAA are [7]:

•Severe multilineage cytopenias, defined by ≥2 of the following:

-ANC <500/microL

-Platelet count <20,000/microL

-ARC <60,000/microL

Plus one of the following:

•Severe bone marrow hypocellularity (defined by one of the following):

-Marrow cellularity <25 percent of normal for age

or

-Marrow cellularity <50 percent of normal for age with non-lymphocytic hematopoietic cells accounting for <30 percent of cellularity

Spontaneous recovery from SAA is rare and unless it is successfully treated; >70 percent will not survive beyond one year from diagnosis [8-10].

Management of SAA is described below. (See 'Severe AA and very severe AA' below.)

●Very severe AA (vSAA) – Diagnosis of vSAA is based on:

•Fulfills the criteria for SAA (above)

Plus

•ANC <200/microL (<0.2 x 10⁹/L)

Distinguishing SAA and vSAA has prognostic importance, but they can be considered together for treatment decisions. However, diagnosing vSAA increases the urgency for prompt initiation of treatment, as described below.

MANAGEMENT — Management of a child or adolescent with AA requires urgent evaluation and care by a clinician experienced with this disorder [11].

Medications or other agents that may cause or exacerbate cytopenias should be withdrawn; however, treatment for severe AA (SAA) or very severe AA (vSAA) should not be delayed while awaiting a response. Potentially offending agents are described above. (See 'Clinical/laboratory evaluation' above.)

Management of children who are not suitable for hematopoietic cell transplantation (HCT; eg, poor medical fitness) is described below. (See 'Immune suppression therapy (IST)' below.)

Severe AA and very severe AA — Management of transplant-eligible children and adolescents with SAA or vSAA depends on the availability of a human leukocyte antigen (HLA)-matched sibling donor (MSD; ie, 8 of 8, or 10 of 10 HLA match).

Matched sibling donor available — For children and adolescents with an available MSD graft, we recommend allogeneic HCT rather than immunosuppressive therapy (IST); both approaches yield similar long-term survival, but compared with IST, MSD transplantation is associated with more favorable disease control and acceptable toxicity.

Compared with IST, allogeneic HCT with an MSD graft provides faster restoration of blood counts, less early treatment failure, fewer relapses, and prevents clonal evolution to hemolytic paroxysmal nocturnal hemoglobinuria (PNH), acute myeloid leukemia (AML), and myelodysplastic syndromes (MDS) [12]. Informative studies include:

●Outcomes with MSD HCT – MSD HCT is associated with 90 percent two-year overall survival (OS) and 95 percent engraftment, according to a summary of 11 studies of transplantation for SAA in children [13]. MSD HCT is associated with acceptable rates of adverse effects, including acute graft-versus-host disease (GVHD; 10 to 20 percent), chronic GVHD (10 to 30 percent), and late effects of therapy, including endocrine dysfunction, avascular necrosis, and rare cancers [14-17]. Further details of outcomes with MSD HCT are presented below. (See 'MSD HCT' below.)

●Outcomes with IST – A large consortium study reported that IST was associated with 93 percent five-year OS and 60 percent complete response (CR), but 7 percent of children developed new clonal chromosomal abnormalities and five-year event-free survival (EFS) was only 64 percent [18]. Furthermore, salvage HCT is generally required for children who are refractory to IST or who later relapse, but outcomes with salvage HCT are inferior to those with front-line transplantation. Details of outcomes and adverse effects with IST (eg, early serum sickness, infections, possible progression to PNH, AML, or MDS) are described below. (See 'IST outcomes' below and 'Adverse effects' below.)

●Comparisons of MSD HCT with IST for SAA in children and adolescents – Few studies have directly compared MSD HCT with IST:

•A registry study reported superior outcomes with MSD HCT compared with IST for children <12 years who were treated from 2000 to 2009 [19]. This comparative study was based on "genetic randomization," in which 487 children with an available MSD graft underwent HCT, while 167 children without an MSD were treated with IST, using anti-thymocyte globulin (ATG) plus cyclosporine A (CsA). Three-year EFS was superior with HCT (87 versus 33 percent), but there was no difference in three-year OS (91 versus 87 percent), because HCT using an unrelated donor (UD) was given to the 55 percent of IST-treated children who did not respond to IST. For those children who underwent rescue HCT, OS was inferior (83 percent) compared with those who received front-line HCT (91 percent) and children who responded to initial IST (97 percent). The rate of graft rejection was 2 percent for all transplants, but compared with front-line HCT, rescue HCT was associated with more grade ≥2 acute GVHD (25 versus 8 percent) and more chronic GVHD (20 versus 6 percent).

•We consider that studies from before 2000 do not reflect contemporary outcomes and are not valid for comparison of HCT versus IST. HCT outcomes have improved with advances in transplantation techniques and management, as well as improved supportive care; outcomes with IST have improved with greater recognition of inherited bone marrow failure syndromes (IBMFS). Examples include an inconclusive meta-analysis of three studies that compared HCT versus IST from the 1970s to the 1990s [20]. A study of 100 consecutive children with moderate AA or SAA (1980s and 1990s) reported superior outcomes with front-line MSD HCT (37 patients) compared with IST (63 patients); 11 children who did not respond to IST were salvaged with UD HCT [21]. However, one-fifth of children who received IST developed MDS, which likely reflected a high incidence of IBMFS (eg, inherited GATA2 abnormalities).

MSD not available — For transplant-eligible children and adolescents with SAA or vSAA who do not have an available MSD graft, we consider either upfront IST or allogeneic HCT using an alternative donor (AD) graft to be acceptable approaches. Both IST and HCT using a well-matched AD are associated with excellent OS, but transplantation is associated with superior EFS and more early toxicity.

The choice of HCT versus IST must be individualized and should consider the degree of immunologic mismatch between the potential AD and the recipient and preferences of the patient and family. Our approach is consistent with a current multi-institutional clinical trial that randomly assigns patients to upfront IST versus transplantation using an AD graft (NCT02845596). For those who choose upfront transplantation, we prefer a matched unrelated donor (MUD; ie, matched at 10 of 10, or 8 of 8 HLA alleles). (See 'MUD HCT' below.)

Treatment with either IST or HCT using an AD graft achieves excellent long-term OS in children with SAA, but they are associated with different toxicities and adverse outcomes [18,22-25]. Studies that directly compared IST versus HCT with an HLA-MUD graft include:

●One study compared outcomes with front-line MUD HCT in 29 children (≤19 years) with data from historical controls, including patients who received front-line MSD HCT (87 patients), IST (58 patients), and rescue HCT (ie, after failure to respond to IST; 24 patients) [26]. Two-year OS for upfront MUD HCT was 96 percent, which was similar to MSD HCT (91 percent) and IST-responsive patients (94 percent); however, two-year OS was inferior for children who required rescue HCT (74 percent). Corresponding rates of two-year EFS were 92, 87, 40, and 74 percent, respectively.

●A pilot study compared outcomes of patients ≤25 years with SAA who were randomly assigned to HCT using a MUD graft (12 patients) versus IST (11 patients); median follow-up was 18 months [27]. Transplantation of 10 children began a median of 36 days after randomization; one child experienced secondary graft failure and required a second transplant. Six of the children who were treated with IST had a complete response (CR), while the five with IST-refractory disease underwent successful salvage HCT; one child who had a CR to IST relapsed after discontinuation of IST and later died of infection after salvage HCT.

●A retrospective study of 42 children <18 years with SAA reported superior outcomes with upfront HCT using an AD graft compared with IST [28]. However, interpretation of the findings is limited because the IST regimens differed from most studies (ie, mostly rabbit ATG, rather than horse ATG and a shorter course of CsA). For the 23 children who underwent HCT, the 91 percent rate of failure-free survival (FFS) was superior to the 31 percent FFS in children who received IST. Five-year OS did not differ (91 versus 71 percent, respectively) because 6 of 11 children who were IST-refractory received salvage HCT; EFS for salvage HCT was inferior to that with upfront HCT (51 versus 91 percent). Grade ≥3 acute GVHD was 4 percent and chronic GVHD developed in 61 percent of transplanted children; GVHD rates did not differ for the front-line HCT versus salvage HCT groups.

Additional studies of outcomes with HCT and IST for children with SAA are presented below. (See 'HCT outcomes' below and 'IST outcomes' below.)

Moderate AA (MAA) — For children and adolescents with MAA, close observation with supportive care is offered as initial management, rather than proceeding to HCT or IST, as long as an underlying IBMFS has been excluded.

The choice of approach must be individualized and may be influenced by the severity of cytopenias and associated complications, frequency of transfusions, and patient/family preferences. Some may choose IST to lessen the need for transfusions and complications, while reserving transplantation in the event of progression to SAA or vSAA. Others may choose allogeneic HCT because they weigh the potential for cure and prevention of progression more than its toxicity. With progressive cytopenias, particularly severe neutropenia and/or transfusion dependence, treatment with HCT or immunosuppression should be considered. (See 'Supportive care' below.)

Children with MAA and symptomatic levels of hemolysis from PNH should begin treatment with a complement inhibitor, as described separately. (See "Treatment and prognosis of paroxysmal nocturnal hemoglobinuria", section on 'Symptomatic hemolytic PNH'.)

Monitoring of the patient with MAA is described below. (See 'Response assessment/monitoring' below.)

ALLOGENEIC HCT

HCT techniques — Techniques for allogeneic hematopoietic cell transplantation (HCT) in children with severe AA (SAA) differ among institutions. Choices for graft donors, conditioning regimens, and graft manipulation (eg, T cell depletion) are described in the following sections.

Graft donor — The preference for transplanting children and adolescents using an available human leukocyte antigen (HLA)-matched sibling donor (MSD) graft is described above. (See 'Matched sibling donor available' above.)

Alternative donor (AD) grafts are acceptable for children and adolescents who do not have available MSD. The preferred AD is a matched unrelated donor (MUD; ie, 10 of 10, or 8 of 8 HLA-match), because outcomes and toxicity are more favorable and there is more extensive experience using MUD grafts than other AD grafts. (See 'MUD HCT' below.)

T cell depletion and graft source — T cell depletion (TCD) can reduce graft-versus-host disease (GVHD) in the transplant recipient, but the choice and methods vary with the donor source, degree of immunologic match, and institutional practice.

●Graft manipulation – There is no consensus about the role for T cell depletion nor the preferred method.

Partial TCD by ex vivo graft processing or by post-transplant cyclophosphamide (PTCy; ie, in vivo TCD) may reduce GVHD after mismatched unrelated donor (mMUD) or haploidentical donor grafts [29,30].

Compared to T replete MUD grafts, TCD of MUD or mMUD grafts are associated with less GVHD and similar high rates of engraftment; this effectively expands the donor pool, especially for ethnic groups that are less well-represented in donor registries [5].

●Graft source – The choice of bone marrow (BM) versus peripheral blood stem/progenitor cells (PBSPC) as the graft source varies:

•Bone marrow – BM is generally preferred for unmanipulated grafts (ie, no TCD) and when PTCy is used, because it is associated with lower rates of GVHD.

In a study using unmanipulated MSD grafts for HCT in children ≤20 years, compared with recipients of BM (558 patients), recipients of PBSPC (134 patients) had more chronic GVHD (RR 2.82 [95% CI 1.46-5.44]), but there was no difference in grade ≥2 acute GVHD (14 versus 10 percent) [15].

•Peripheral blood stem/progenitor cells – For AD grafts, PBSPC with partial TCD is an acceptable choice.

Conditioning therapy — There is general agreement that myeloablative conditioning (MAC) is not required for HCT in children and adolescents with SAA, but there is no consensus conditioning therapy and no regimen has emerged as superior to others.

The preferred approach varies by country, study group, and institution. Nearly all conditioning regimens include cyclophosphamide and serotherapy (ie, either alemtuzumab or anti-thymocyte globulin [ATG]), but the doses and schedules vary. Regimens also differ regarding addition of low dose radiation and/or fludarabine. MAC can be functionally defined as regimens that incorporate busulfan and/or ≥1000 centigray (cGy) total body irradiation (TBI).

A phase 3 trial that compared conditioning regimens for children undergoing allogeneic HCT for SAA was inconclusive. In this trial, there was no difference in survival for 134 patients with SAA who were randomly assigned to conditioning with cyclophosphamide versus cyclophosphamide plus ATG, alemtuzumab, or other agents; transplantation used T cell-replete BM grafts from an MSD [31]. With a median follow-up of six years, five-year overall survival (OS) was 74 percent for cyclophosphamide conditioning versus 80 percent for cyclophosphamide plus ATG. There were no differences in the rates of graft failure or GVHD. Beyond this trial, it is difficult to compare outcomes with conditioning regimens across different studies.

HCT outcomes — Transplant outcomes vary with the degree of HLA-match graft donor.

MSD HCT — Transplantation with an MSD is associated with 90 percent two-year OS and 95 percent engraftment, according to a summary of 11 studies of transplantation for SAA in children [13]. MSD HCT is associated with acceptable rates of adverse effects, including acute GVHD (10 to 20 percent), chronic GVHD (10 to 30 percent), and late effects of therapy, including endocrine dysfunction, avascular necrosis, and rare cancers [14-17].

Studies that demonstrated superior outcomes with MSD HCT versus immunosuppressive therapy (IST) are presented above. (See 'Matched sibling donor available' above.)

MUD HCT — Outcomes of transplantation using an MUD (ie, 10 of 10, or 8 of 8 HLA match) have improved considerably over recent decades. Availability of suitable MUD grafts in the United States is higher in White individuals than in Black individuals and other minorities. (See "Donor selection for hematopoietic cell transplantation", section on 'Unrelated donors'.)

MUD HCT is associated with >90 percent long-term OS and acceptable levels of toxicity [22-24]. Outcomes with MUD HCT now approach those with MSD grafts [26,28]. As an example, five-year failure-free survival (FFS) was 95 percent in 44 consecutive children who received an MUD graft; 40 of the children had previously failed IST [32].

Improved outcomes with MUD HCT are associated with allele-specific HLA matching, better GVHD management, improved antiviral prophylaxis, and advances in supportive care [33-37]. The importance of allele-specific matching to identify a proper MUD source is illustrated by a study in which only 55 percent of donor and recipient pairs who were judged to be serologically identical were highly-matched by molecular typing [33].

Studies that compared MUD HCT with IST are presented above. (See 'MSD not available' above.)

Other alternative donor HCT — HCT using MUD grafts is generally superior to other AD sources, such as mMUD (≤9 of 10, or ≤7 of 8 HLA match), incompletely HLA-typed unrelated donors, umbilical cord blood (UCB), and haploidentical (haplo) grafts. Beyond transplantation using an MUD graft, the preference for other AD sources varies among institutions and there is no consensus regarding a preferred approach.

There is less experience using non-MUD AD grafts, but compared with MUD HCT, they are generally associated with higher rates of graft rejection, GVHD, and infectious complications.

●Mismatched MUD – A registry study of 195 patients (≤20 years) transplanted with AD grafts (1989 to 2003) reported that 10-year OS was inferior with mMUD compared with MUD HCT (39 versus 57 percent; HR 0.25 [95% CI 0.09-0.70]) [38]. Chronic GVHD was reported in 35 percent of the patients in this study, with increased relative risk (RR 2.05) in mMUD recipients.

However, ex vivo TCD or post-transplant cyclophosphamide (PTCy) reduces the risk of GVHD using mismatched donors, thereby allowing outcomes similar to those with MUD donors [29,30].

●Haploidentical HCT (haplo) – A systematic review and meta-analysis of haplo HCT for SAA, which included 577 patients from 15 studies (mostly children, adolescents, and young adults), reported 7 percent treatment-related mortality (TRM) per year, grade ≥2 acute GVHD in 27 percent, chronic GVHD in 25 percent, and successful engraftment in 97 percent [39].

A multicenter study in both children and adults reported no differences between haplo HCT versus MSD HCT in three-year OS (86 and 91 percent, respectively), FFS (85 and 90 percent), or engraftment (>97 percent for both), but haplo HCT was associated with more grade ≥2 acute GVHD (30 versus 2 percent) and more chronic GVHD (31 versus 4 percent). Similar outcomes with haplo HCT were reported in other multicenter studies [40,41].

PTCy was given to 13 patients who received haplo grafts and three who received unrelated donor grafts in a single-center study of HCT for SAA [29]. All 16 patients (median age 30 years) engrafted without evidence of clonality and were alive with excellent performance status after median 21-month follow-up. There was limited GVHD; two patients had grade 2 acute GVHD and two had skin or oral chronic GVHD.

●Umbilical cord blood – Few patients have a UCB unit matched at the antigen level at HLA-A and HLA-B loci and matched at high resolution at HLA-DRB1, but units with one or two HLA loci mismatches are acceptable and available for almost all patients <20 years [5]. The dose should be >3 x 10⁷ total nucleated cells/kg recipient weight, although the threshold dose has not been rigorously tested; if the UCB cell dose is inadequate, supplemental bone marrow may be added. UCB HCT is associated with substantial risk of TRM and engraftment failure.  

A multicenter study reported outcomes with UCB grafts in 25 patients (median age 16 years) with refractory SAA [42]. With 39-month median follow-up, one-year OS was 88 percent, engraftment occurred in 88 percent, and adverse effects included 46 percent grade ≥2 acute GVHD and 36 percent chronic GVHD.

IMMUNE SUPPRESSION THERAPY (IST) — IST is an acceptable treatment for front-line therapy of severe AA (SAA) in patients who do not have a human leukocyte antigen (HLA)-matched sibling donor (MSD), as discussed above. (See 'MSD not available' above.)

IST is the preferred approach for children who are not medically-fit for allogeneic hematopoietic cell transplantation (HCT).

Administration — The preferred regimen for IST varies among institutions.

We currently treat as follows:

●Horse anti-thymocyte globulin (ATG) – 40 mg/kg per day for four days.

We provide Pneumocystis prophylaxis with pentamidine or atovaquone for children within three months of receiving ATG. (See "Treatment and prevention of Pneumocystis pneumonia in patients without HIV".)

●Cyclosporine A (CsA) – CsA is begun on day 1 to maintain a target blood trough of 150 to 300 ng/mL. Provided there is no nephrotoxicity, we continue therapy for one year.

The starting dose of CsA is as follows:

•No voriconazole – Not currently taking antifungal prophylaxis with voriconazole:

-Age ≤12 years – 7.5 mg/kg every 12 hours

-Age >12 years – 5 mg/kg every 12 hours for patients not receiving antifungal prophylaxis

•Currently taking voriconazole – For children taking voriconazole the starting dose is 2.5 mg/kg every 12 hours.

If there is evidence of stable trilineage recovery, we attempt a slow wean by tapering the daily dose by 25 percent every three months over the ensuing 12 months in children who maintain a complete response (CR) or good partial response (PR).

Some patients cannot be successfully weaned and require prolonged CsA therapy. Children who do not exhibit trilineage recovery by six to eight months or who relapse following weaning of CsA should proceed to evaluation for HCT [43,44]. Management of children with IST-refractory or relapsed SAA is described below. (See 'Relapse' below.)

●Prednisone – To prevent and control serum sickness, treat with prednisone 1 mg/kg/day in divided doses every 12 hours (maximum dose of 50 mg/day) for 10 days. This is followed by a slow taper over the two weeks, with the aim to discontinue corticosteroid therapy before 30 days.

●Other components – There is no consensus regarding addition of the thrombopoietin (TPO) receptor agonist, eltrombopag, or granulocyte colony stimulating factor (G-CSF) to the IST regimen for children with SAA.

•Eltrombopag – We add eltrombopag for patients ≥18 years. Subgroup analysis of a US National Institutes for Health (NIH) study reported that, compared with historical controls treated with IST alone, there was no response or survival advantage to IST plus eltrombopag for children <18 years, but median time to relapse or high-risk clonal evolution was shorter in the eltrombopag group [25]. Eltrombopag does improve responses in adults with SAA (discussed separately), but prospective trials are needed to determine if eltrombopag can further improve responses to ATG/CsA in children and adolescents. (See "Treatment of aplastic anemia in adults", section on 'Triple IST (hATG, CsA, EPAG)'.)

•G-CSF – We do not routinely add G-CSF to the IST regimen, as its benefit is controversial. Prolonged use of G-CSF appears to be a risk factor for clonal evolution toward myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) in children with AA [45].

A phase 3 trial reported no benefit with G-CSF added to IST among 192 patients randomly assigned to ATG plus CsA, with or without G-CSF [46]. There was no difference in 15-year overall survival (OS; 57 percent versus 63 percent, respectively) or event-free survival (EFS; 24 and 23 percent). MDS or AML arose in 9 patients, 10 had an isolated clonal cytogenetic abnormality (ie, without MDS or AML), 7 had a solid cancer, 18 developed paroxysmal nocturnal hemoglobinuria (PNH), 8 had osteonecrosis, and 12 developed chronic kidney disease; there was no difference in these outcomes between patients treated with or without G-CSF. The cumulative incidence of any late event (MDS, AML, isolated cytogenetic abnormalities, solid cancer, clinical PNH, aseptic osteonecrosis, chronic kidney disease, and relapse) was 50 percent for the G-CSF group and 49 percent for the non-G-CSF group.

The role of G-CSF in supportive care is discussed below. (See 'Supportive care' below.)

Adverse effects — Adverse effects of IST in children include:

●Serum sickness occurs 7 to 10 days after ATG therapy and is common, but it may be prevented with prednisone treatment during the first two weeks after initiating ATG. Fever typically develops on the first day of treatment with ATG and lasts from one to six days.

In one study of 39 such patients, positive blood cultures were obtained from only four patients, and the organism in each case was a coagulase-negative Staphylococcus [47]. Therefore, discontinuing antibiotics in febrile patients who are stable and have no evidence of sepsis is a reasonable option.

●Opportunistic infections – Intense immunosuppression puts the patient at increased risk for opportunistic infections, particularly fungal infection. We recommend antifungal prophylaxis for patients with prolonged severe neutropenia (absolute neutrophil count [ANC] <500 for >1 week). IST may also impair immune responses to vaccines, and thus vaccines should not be given while patients are receiving IST. Inactivated vaccines may be given after IST has been discontinued for six months, while live-attenuated virus vaccines may be given after IST has been discontinued for one year.

●Drug toxicity – Cyclosporine has frequent adverse effects, including hypertension, nephrotoxicity, hirsutism, and gingival hypertrophy. In severe cases of gingival hypertrophy, treatment with azithromycin and even replacement of cyclosporine to tacrolimus can be considered. (See "Cyclosporine and tacrolimus nephrotoxicity".)

●Aplasia – In some series, the cumulative risk of relapse of aplasia increases over time, from 9 percent at five years to 35 percent at 14 years [44,48,49].

●Paroxysmal nocturnal hemoglobinuria – PNH clones are detectable in 40 to 50 percent of pediatric patients at the time of diagnosis of SAA or very severe AA (vSAA), but most clones are only 1 to 10 percent and not symptomatic [18].

Transformation of subclinical PNH clones to hemolytic PNH is poorly defined for pediatric patients with SAA. The probability of developing clinical PNH was 10 percent in a series of 57 children [50]. The rate is considerably higher in adults; 15 to 33 percent of adults treated with immunotherapy recovered with clonal evidence of PNH and clinical manifestations that may be mild, transient, or progress to full-blown disease [51,52].

We recommend screening at least yearly for PNH evolution in all pediatric SAA patients treated with IST or those treated with HCT who lack 100 percent donor myeloid chimerism. (See "Clinical manifestations and diagnosis of paroxysmal nocturnal hemoglobinuria", section on 'Clinical manifestations'.)

●MDS/AML – Evolution to MDS/AML is infrequent in children, even though many have clonal hematopoiesis when evaluated using sensitive genetic tests.

Among 314 children with SAA in North America, only 2 percent developed MDS or leukemia after median follow-up of five years [18]. The most common chromosomal abnormalities in the cases of MDS that do arise from pediatric SAA are losses of part or all of chromosome 7 (eg, -7, del(7q)).

Next generation sequencing and/or single nucleotide polymorphism (SNP) array genotyping can detect clonal hematopoiesis in two-thirds of children with AA [53]. However, most reflect clonality associated with PNH, clonal loss of HLA allele expression through allele-specific inactivating somatic mutations, clonal copy-neutral loss of heterozygosity of chromosome 6p, or tolerated copy number changes including del(13q) [54,55].

Response assessment/monitoring

●Response – The response to IST may be slow. Typically, granulocyte recovery occurs first, followed by stabilization of hemoglobin and a decline in transfusion requirements; platelet recovery may take months to years. Long-term survivors may show persistent thrombocytopenia, red blood cell macrocytosis, and elevated hemoglobin F concentrations [56].

●Monitoring – The schedule for follow-up is individualized and should reflect the response to therapy, evidence of clonal evolution, and concerns of the clinician and patient/family.

We screen for evolution to PNH by flow cytometry of peripheral blood at least annually in all children with SAA who are treated with IST or who have <100 percent donor myeloid chimerism after HCT.

We generally perform a bone marrow examination one year after beginning IST treatment to assess relapse of AA and to assess clonal progression to MDS or AML (eg, development of low-frequency clones with chromosome 7 abnormalities or acquired somatic mutations predictive of MDS/AML, such as AXL1, or TP53).

For children with a partial response to IST and for those who have a cytogenetic abnormality to follow, we repeat the bone marrow annually, or as clinically indicated. Otherwise, we do no perform routine schedule bone marrow evaluation.

IST outcomes — Treatment with IST in children with SAA is associated with approximately 90 percent long-term survival, but only one-third to one-half of children achieve CR and many require salvage HCT for relapse or refractory disease. Studies that compared IST with allogeneic HCT for treatment of SAA/vSAA in children are discussed above. (See 'Severe AA and very severe AA' above.)

●Survival and response rates – A multi-institutional study of >300 children with SAA reported 92 percent five-year OS [18], while 84 percent OS was reported from a series of studies at the NIH [25]. Outcomes with IST are similar in young children compared with adolescents [19,57]. However, only one-third to one-half of children achieve CR with IST. Five-year event-free survival (EFS) for children treated with IST was only 62 percent, and there was no apparent plateau, even after five years. Other large retrospective studies of IST for SAA in children report similar outcomes [19,48].

●Toxicity – Adverse effects of IST include serum sickness, increased risk for infections, and high-risk clonal evolution (eg, chromosome 7 abnormality, complex karyotype, or overt myeloid malignancy). However, most pediatric patients progress normally through puberty, have low incidence of organ toxicity, and have high rates of preserved fertility [58]. Further description of adverse effects of IST is presented above. (See 'Adverse effects' above.)

After IST, 20 to 30 percent of responders relapse, have CsA dependence (20 to 30 percent), or have clonal evolution to PNH, MDS, or AML [19,59]. Most children who are IST-refractory or relapse after IST later require transplantation using an alternative donor, but outcomes with rescue transplantation are inferior to those with front-line transplantation.

SUPPORTIVE CARE — All children with AA should receive supportive care to ameliorate the cytopenias/marrow aplasia [60].

●Transfusions – Red blood cell (RBC) transfusions should be used only to treat definite symptoms/signs rather than maintain an arbitrary level of hemoglobin. Blood and platelet transfusions should be used selectively in candidates for hematopoietic cell transplantation (HCT) to avoid sensitization/alloimmunization. All blood products should be irradiated, leukocyte-reduced, and cytomegalovirus (CMV) safe, and should not be from family members, to avoid human leukocyte antigen (HLA) alloimmunization against potential graft donors for HCT. (See "Red blood cell transfusion in infants and children: Selection of blood products".)

The transplant donor should not be used as a source for transfusion support. In the past, graft rejection was often associated with allosensitization to minor histocompatibility antigens from RBC and platelet transfusions [61]. Although the use of anti-thymocyte globulin (ATG) has reduced the risk of graft failure and consequently reduced earlier concerns about avoiding transfusions, the transplant donor should not be used for transfusion support.

●Neutropenia – Patients with severe neutropenia are at risk for serious bacterial infections (table 3). (See "Management of children with non-chemotherapy-induced neutropenia and fever", section on 'Risk-based management approach'.)

•Fever – Affected children with fever require immediate evaluation, blood cultures, and empiric treatment with broad-spectrum antibiotics.

•Prophylaxis – We provide the following:

-Antifungal – For all children with prolonged severe neutropenia (ie, absolute neutrophil count [ANC] <500 white blood cells/microL for >1 week) we provide antifungal prophylaxis with an azole or echinocandin class antifungal agent.

There is no well-demonstrated benefit for the following:

-Antibacterial prophylaxis – The use of prophylactic antibacterial agents is controversial, with some studies demonstrating no benefit [62].

-Neutropenic diet – There is no benefit to the "neutropenic diet," which is not microbially superior, and may be nutritionally inferior, to a diet following standard United States Department of Agriculture (USDA) food safety guidelines [63,64].

-Myeloid cytokines – We reserve treatment with granulocyte-colony stimulating factor (G-CSF) to patients who have ANC <200 and have a documented infection or are suspected to have bacterial or fungal infection. We treat with G-CSF 5 mcg/kg/day subcutaneously or intravenously if a severe bacterial or fungal infection is suspected or confirmed.

RELAPSE — For children who relapse after immunosuppressive therapy (IST), we suggest alternative donor allogeneic hematopoietic cell transplant (HCT) rather than repeated treatment with IST, based on superior long-term survival and acceptable toxicity with transplantation and low rates of response to repeated IST.

For children who relapse, but have no satisfactory alternative donors (including no haploidentical related donors), we would attempt a second course of IST, but substitute rabbit anti-thymocyte globulin (ATG; instead of horse ATG) in the IST regimen.

The following studies illustrate outcomes with treatment of relapsed AA:

●Salvage HCT versus repeated IST – In a prospective multicenter study of 60 children who failed initial IST for severe AA (SAA), five-year failure-free survival (FFS) in 21 patients who received a second course of IST was 10 percent, compared with an 84 percent FFS rate for patients who received HCT as therapy following failure of initial IST [65].

In a retrospective study of pediatric patients with SAA, of 52 patients requiring a second therapy after failed IST with horse ATG and cyclosporine (CsA), only 37 percent had not required additional therapy at time of data analysis [18]. Compared with second IST, HCT had superior event-free survival (EFS; hazard ratio [HR] 0.19 [95% CI 0.08-0.47]).

●Salvage HCT – In a study of SAA, 29 percent of 205 children failed to respond to IST after six months; HCT from an alternative donor was associated with 84 percent five-year FFS [65]. Other studies using matched unrelated donor or another alternative donor graft HCT are presented above. (See 'MUD HCT' above and 'Other alternative donor HCT' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Bone marrow failure syndromes".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Aplastic anemia (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Acquired aplastic anemia (AA) is an immune-mediated loss of hematopoietic stem and progenitor cells that is associated with pancytopenia and bone marrow aplasia.

●Exclude other disorders – It is important to exclude an inherited bone marrow failure syndrome (IBMFS), hematologic malignancy, and hemophagocytic lymphohistiocytosis in children and adolescents with apparent AA, as described above. (See 'Exclude other causes' above.)

●Transplant eligibility and donors – All patients should be referred for consultation with transplant specialists to determine eligibility for hematopoietic cell transplantation (HCT). (See 'Transplant eligibility' above.)

Human leukocyte antigen (HLA) testing of the child and siblings should be performed promptly. If there is no suitable HLA-matched sibling, we promptly begin a search for an unrelated donor (UD) in a donor registry. (See 'Donor search' above.)

●Disease severity – AA is classified according to level of cytopenias and extent of hypocellularity (details above) (see 'Classification of severity' above):

•Moderate AA (MAA)

•Severe AA (SAA)

•Very severe AA (vSAA)

●Management – Treatment varies with disease severity and available transplant donors:

•SAA/vSAA – Management varies according to availability of a matched sibling donor (MSD; ie, 10 of 10, or 8 of 8 HLA match) graft:

-Available MSD graft – For transplant-eligible children and adolescents with an available MSD graft, we recommend allogeneic HCT rather than immunosuppressive therapy (IST) (Grade 1B). (See 'Matched sibling donor available' above.)

-No MSD graft available – For patients who do not have an available MSD graft, we consider either upfront IST or allogeneic HCT using an alternative donor (AD) graft to be acceptable approaches; treatment choice must be individualized. (See 'MSD not available' above.)

•MAA – Close observation with supportive care is given, rather than HCT or IST. (See 'Moderate AA (MAA)' above.)

●Transplantation – HCT techniques differ among institutions; the following aspects are discussed above:

•Graft donor – Preference for an MSD graft and selection of an AD (if an MSD is not available) are discussed above. (See 'Graft donor' above.)

•T cell depletion (TCD) – There is no consensus about TCD and choice of bone marrow versus peripheral blood stem and progenitor cells as the graft source varies with HCT technique, donor, and TCD. (See 'T cell depletion and graft source' above.)

•Conditioning regimen – Myeloablative conditioning (MAC) is not required for HCT in children and adolescents with SAA, but there is no consensus conditioning therapy, and no regimen has emerged as superior to others. (See 'Conditioning therapy' above.)

•Outcomes with HCT for SAA are presented above. (See 'HCT outcomes' above.)

●Immunosuppressive therapy – IST generally includes anti-thymocyte globulin (ATG), cyclosporine A (CsA), and prednisone. Administration of IST, adverse effects, response monitoring, and outcomes are discussed above. (See 'Immune suppression therapy (IST)' above.)

●Relapse – For children who relapse after IST, we suggest AD HCT, rather than repeated treatment with IST (Grade 2C). (See 'Relapse' above.)

ACKNOWLEDGMENTS — The editorial staff at UpToDate acknowledges Alison A Bertuch, MD, PhD, and Stanley L Schrier, MD, who contributed to earlier versions of this topic review.