Hematopoietic cell transplantation for acute myeloid leukemia and myelodysplastic syndromes in children and adolescents

INTRODUCTION — Allogeneic hematopoietic cell transplantation (HCT) can cure acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) in children, but it is associated with substantial toxicity and possible transplant-related mortality (TRM). The transplanted hematopoietic cells may recognize the recipient's cells as "foreign" and cause graft-versus-host disease (GVHD), which is a major cause of morbidity and mortality associated with allogeneic HCT. However, the allogeneic cells also establish a graft-versus-leukemia (GVL) effect, which contributes importantly to eradication of the malignant cells. Limiting the toxicity of GVHD, while enabling the benefit of GVL, is key to successful outcomes with allogeneic HCT.

Autologous transplantation is not used to treat AML or MDS in children.

For children with AML or MDS, it is especially important to exclude an underlying inherited bone marrow failure syndrome (IBMFS). Children with an IBMFS require special considerations for donor selection (to avoid reconstituting the bone marrow with the same condition) and choice of conditioning regimen, they may have non-hematologic manifestations of the syndrome that can affect transplantation, and they have increased risk for TRM, as discussed separately. (See "Hematopoietic cell transplantation (HCT) for inherited bone marrow failure syndromes (IBMFS)".)

This topic discusses allogeneic HCT in children with AML or MDS who do not have an IBMFS.

Management of AML and MDS are discussed separately. (See "Acute myeloid leukemia in children and adolescents" and "Treatment of high or very high risk myelodysplastic syndromes".)

GENERAL CONCERNS — Outcomes with transplantation for AML and MDS in children have improved dramatically over recent decades [1]. Nevertheless, the potential for HCT to cure myeloid malignancies must be balanced against available alternative treatments and the short-term and long-term consequences of transplantation.

The decision to pursue transplantation must be individualized, with consideration of the child's age, current clinical status, alternative treatment options, transplant-related risks and complications, and the patient's/family's wishes. Alternative treatments for AML and MDS and indications for allogeneic HCT are discussed separately. (See "Acute myeloid leukemia in children and adolescents" and "Treatment of high or very high risk myelodysplastic syndromes".)

It is important to consider whether AML or MDS in a child is related to an underlying inherited bone marrow failure syndrome (IBMFS). (See 'Exclude inherited disorders' below.)

Transplantation in children — Children require evaluation, management, and long-term surveillance that differ from transplantation in adults.

For children, ongoing growth and the potential for decades of post-transplant survival require special attention. The conditioning regimen and other aspects of transplantation can impair growth and development, while graft-versus-host disease (GVHD) and other transplant-related adverse effects can be devastating in children. In addition to monitoring for disease relapse, post-transplant surveillance must focus on short- and long-term complications, as some consequences of transplantation in children may not emerge for decades. (See 'Post-transplantation' below.)

Transplantation must also consider the age of the child. Outcomes and toxicity with HCT differ for infants (<2 years), young children (≥2 to <12 years), and adolescents (≥12 to 18 years); note that these age categories are differences of degree, rather than distinct age thresholds. Outcomes according to the age at transplantation are discussed below. (See 'Patient-related factors' below.)

Exclude inherited disorders — It is especially important in children to exclude the possibility that AML or MDS arose from an underlying IBMFS.

Children with Fanconi anemia, dyskeratosis congenita, Diamond-Blackfan anemia, Shwachman-Diamond syndrome, and other IBMFS are at increased risk for myeloid malignancies; in some cases, a myeloid malignancy is the first manifestation of that disorder. HCT in a child with an underlying IBMFS is associated with increased risk for transplant-related toxicity and late effects, including a higher incidence of second cancers. Graft selection must carefully exclude as potential donors those relatives whose hematopoietic grafts might reconstitute the syndrome in the transplanted recipient. Selection of conditioning regimens and other treatments must be tailored for the underlying IBMFS. Special considerations for HCT in patients with an underlying IBMFS are discussed separately. (See "Hematopoietic cell transplantation (HCT) for inherited bone marrow failure syndromes (IBMFS)".)

GVHD and GVL — Successful transplantation in children requires proper management of GVHD and the graft-versus-leukemia (GVL) effect.

Immunologic mismatches (even seemingly minor differences) between the transplanted hematopoietic cells ("graft") and the transplant recipient ("host") cause the graft to recognize host cells as "foreign." These immunologic differences can cause GVL (which contributes importantly to the eradication of malignant cells), but they also cause GVHD (a major cause of transplant-related morbidity and mortality). Enabling GVL while limiting GVHD is important for successful long-term survival after allogeneic HCT for AML or MDS.

Factors that influence the incidence and degree of GVHD and GVL include:

●Donor selection – The preference for a related donor graft and selection of an unrelated donor (if a related donor graft is not available) are discussed below. (See 'Graft' below.)

●Graft source – The choice of bone marrow versus peripheral blood stem/progenitor cell sources is discussed below. (See 'Graft source' below.)

●Conditioning regimen – The intensive chemotherapy-based treatment that is administered immediately prior to transplantation is referred to as conditioning therapy (also called preparative therapy). The choice of myeloablative conditioning (MAC) versus reduced intensity conditioning (RIC) is discussed below. (See 'Conditioning therapy' below.)

PRETRANSPLANT EVALUATION — Pretransplant evaluation includes assessment of medical fitness, disease status, and the individual child's risk for organ dysfunction.

●Transplant eligibility – Eligibility for HCT requires adequate medical fitness, including cardiac, liver, and lung function. Details of the pretransplant clinical and laboratory assessment are described separately. (See "Determining eligibility for allogeneic hematopoietic cell transplantation".)

●Other studies are guided by the child's clinical condition, and may include:

•Cardiac function – Electrocardiogram and echocardiogram or radionuclide ventriculogram to assess ejection fraction.

•Pulmonary function tests – Testing according to symptoms, imaging, and age.

•Kidney function – Assessment of glomerular filtration rate (GFR) with nuclear GFR test and/or creatine and cystatin C. (See "Assessment of kidney function".)

•Imaging – Computed tomography (CT) of the chest and/or abdomen may be useful for identifying a cause for organomegaly or fever. Positron emission tomography (PET)/CT may detect extramedullary tumors, and magnetic resonance imaging (MRI) can be useful for assessment of neurologic findings.

•Lumbar puncture as clinically indicated.

●Specialty consultation – We routinely obtain consultation prior to HCT with specialists in cardiology, pulmonary, nephrology, and/or endocrinology to consider risks to organ systems with transplantation.

●Fertility preservation – Children undergoing HCT should be offered fertility consultation and fertility preservation. (See "Fertility and reproductive hormone preservation: Overview of care prior to gonadotoxic therapy or surgery".)

●Disease status – Documentation of disease status (ie, remission versus active disease) requires bone marrow examination, including morphology, flow cytometry, and molecular studies. (See "Clinical manifestations, diagnosis, and classification of myelodysplastic syndromes (MDS)", section on 'Bone marrow examination' and "Acute myeloid leukemia: Induction therapy in medically fit adults", section on 'Introduction'.)

TRANSPLANTATION — Choices of the optimal donor, graft source, manipulation (eg, T cell depletion), and management of graft-versus-host disease (GVHD) influence outcomes with transplantation for myeloid malignancies in children.

Graft — The degree of immunologic match between host and graft influences the likelihood of GVHD and the graft-versus-leukemia (GVL) effect. A suitable transplant donor can be identified for nearly all children who require HCT. (See 'GVHD and GVL' above.)

Donor search — The search for a suitable donor should begin as soon as possible in a child with AML or MDS.

●Related donors – The donor search should begin with human leukocyte antigen (HLA) typing of the patient and available siblings and parents. HLA typing should utilize molecular techniques for HLA class I and II antigens; a well-matched donor corresponds to ≥9 of 10 alleles at HLA-A, -B, -C, -DRB1, and -DQB1. (See 'Matched sibling donor' below.)

●Alternative donors – If a suitable HLA-matched sibling donor (MSD) is not identified, the search for an alternative donor should be initiated promptly. (See 'Alternative donors' below.)

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), and of the same sex (to lessen immunologic mismatch); other considerations include cytomegalovirus (CMV) serologic status of donor and recipient and NK alloreactivity. Factors that affect donor selection are discussed separately. (See "Donor selection for hematopoietic cell transplantation".)

Matched sibling donor — A MSD graft is preferred for transplantation in children because this has long been associated with superior outcomes. However, most studies that have compared MSD grafts with alternative donor grafts should be interpreted with caution, because they used less precise methods of matching than contemporary standards.

Most published studies evaluated recipients and potential donors at only 8 HLA loci, rather than the 10 HLA loci that are assessed with contemporary high-resolution typing. Importantly, lower-resolution matching is less likely to detect mismatches with unrelated donors (UD). As a result, such studies may have overstated the relative benefits of MSD grafts because they may have inadvertently compared MSD grafts with imperfectly-matched UD grafts.

Examples of studies that compared transplant outcomes with MSD versus UD grafts include:

●Children who received MSD grafts had superior overall survival (OS) and lower transplant-related mortality (TRM) compared with recipients of alternative donor grafts, in a registry study of children (≤18 years) transplanted for a hematologic malignancy from 2008 to 2014 by the Center for International Blood and Marrow Transplant Research (CIBMTR) [2]. Importantly, this study matched donors at 8 HLA loci, rather than 10 loci. Two-year OS for the entire 978 patient cohort was 63 percent and included 653 children with AML and 150 with MDS.

•Compared with 539 children who received an MSD graft, the relative risk (RR) for mortality among 310 children with a matched unrelated donor (MUD) graft (8 of 8 HLA loci) was 1.3 (95% CI 1.1-1.7) [2]. For 129 children who received a mismatched unrelated donor (mMUD) graft (7 of 8 HLA match) the RR for mortality was 1.6 (95% CI 1.2-2.2) [2].

•Alternative donor grafts were also associated with increased TRM and treatment failure (ie, death plus relapse) [2]. Compared with an MSD graft, the RR for TRM with MUD was 1.85 (95% CI 1.2-2.8) and for mMUD was 2.7 (95% CI 1.7-4.5); the RR for treatment failure was 1.2 (95% CI 1.0-1.5) with MUD and 1.5 (95% CI 1.1-2.0) with mMUD. The study did not distinguish rates of grade ≥3 chronic GVHD (cGVHD; 24 percent in the overall population), hepatic sinusoidal obstruction syndrome (SOS; 15 percent overall), or engraftment failure according to donor type.

●In an earlier CIBMTR study of 180 children transplanted for AML (2000 to 2009), multivariate analysis reported that rates of OS, leukemia-free survival (LFS), and relapse did not differ between recipients of grafts from a UD versus an MSD graft [3]. However, recipients of UD grafts had more grade ≥2 acute GVHD (aGVHD; 2.5 RR [95% CI 1.2-4.9]) and more cGVHD (2.3 RR [95% CI 1.1-4.7]) than MSD recipients.

●In a multicenter study of transplantation in children with advanced MDS, recipients of mMUD grafts that were incompletely HLA-typed had inferior outcomes compared with recipients of MSD or MUD grafts (≥7/8 HLA match) [4]. Five-year OS was 63 percent for the entire cohort, which included 39 MSD grafts, 31 MUD grafts, and 17 incompletely HLA-typed mMUD grafts [4]. Recipients of MSD and MUD grafts had similar event-free survival (EFS; ie, alive and free of disease; 67 and 53 percent, respectively), cumulative risk of relapse (15 and 24 percent), and TRM (18 and 23 percent). However, compared with MSD and MUD, recipients of mMUD grafts had inferior EFS (35 percent; 3.6 RR compared with MSD; 95% CI 1.4-9.5) and more relapses (29 percent; 5.6 RR; 95% CI 1.4-22.7); TRM did not differ significantly according to graft type.

Outcomes with allogeneic HCT for AML and MDS in children are discussed below. (See 'Outcomes' below.)

Alternative donors — If a suitable HLA-matched related/sibling donor is not identified, a search for a UD should begin promptly. Patients from certain racial or ethnic minority populations may be less likely to find a well-matched UD, but use of haploidentical donors (eg, parents) has considerably increased the likelihood of identifying a satisfactory donor for all children who will be transplanted.

Options for alternative donor selection follow:

●Matched unrelated donor (MUD) – For transplant candidates who do not have a suitable MSD/sibling donor, we suggest an HLA-MUD (ie, ≥9 of 10 HLA alleles), rather than a mMUD, umbilical cord blood (UCB) graft, or haploidentical donor.

No randomized studies have directly compared various alternative graft sources, but outcomes using MUD grafts are generally superior to those using mMUD or incompletely HLA-typed unrelated donors [3,4]. Outcomes with MUD grafts approach those with MSD grafts in some studies [2,4], as discussed above. (See 'Matched sibling donor' above.)

●Other alternative donors

•Mismatched unrelated donor (mMUD) – These grafts are generally defined as <9 of 10 HLA alleles (or <7 of 8 HLA alleles in older studies). Outcomes using mMUD versus MSD and MUD grafts are discussed above. (See 'Matched sibling donor' above.)

•Umbilical cord blood donor – UCB is an acceptable alternative graft source for children who do not have a suitable MUD donor, but it is associated with substantial risk of TRM and engraftment failure.

The dose for UCB grafts should be >3 x 10⁷ total nucleated cells/kg, but the threshold dose has not been rigorously tested. If the UCB cell dose is inadequate, supplemental bone marrow may be added. The UCB match should be ≥6 of 8 HLA alleles.

Among 95 UCB transplants for AML in children, TRM at day 100 was 20 percent; the primary causes of non-leukemic death were infection (in 18 cases) and GVHD (in three patients) [5]. In a single-institution study of children undergoing HCT for various disorders, primary graft failure (PGF) occurred in one-third of 51 children transplanted with UCB grafts, compared with none among 49 recipients of MSD or MUD grafts; most PGF in this study occurred in children who had non-malignant diseases [6].

•Haploidentical donor – There is limited experience with haploidentical grafts for HCT in children. Most studies include few, if any, haploidentical grafts, but a study in which haploidentical grafts constituted <10 percent of HCT reported that outcomes did not differ from those of children receiving an MSD graft [7].

Evidence-based guidelines for selection of unrelated donors and UCB donors are available from the National Marrow Donor Program (NMDP) and the CIBMTR [8].

Graft source — For HCT in children, we suggest bone marrow (BM) as the graft source, rather than peripheral blood stem/progenitor cells (PBSPC). The two graft sources are associated with comparable survival, but we consider the rates of cGVHD, TRM, and late deaths using PBSPC grafts to be unacceptably high for transplantation in children. PBSPC are acceptable as the graft source if use of BM is precluded by donor size or cell dose.

BM and PBSPC graft sources are associated with different patterns of adverse effects and causes of death. Compared with BM, PBSPC grafts are associated with more cGVHD and TRM, but lower rates of relapse. PBSPC grafts have approximately 10-fold more lymphocytes than BM grafts [9] and this may affect immune reconstitution, GVHD, and GVL. PBSPC grafts are collected from donors after treatment with granulocyte colony-stimulating factor (G-CSF).

Retrospective studies have compared outcomes with BM and PBSPC grafts:

●Survival was similar for recipients of BM versus PBSPC grafts among 650 children transplanted for acute leukemia between 2000 and 2012 in CIBMTR registry data; half of the children had AML, while the others had acute lymphoblastic leukemia (ALL) [10]. Eight-year OS did not differ for BM versus PBSPC grafts (47 versus 42 percent, respectively) and LFS was 40 percent for both groups. However, compared with BM, PBSPC was associated with higher rates of late (ie, >6 month) overall mortality (HR 1.3 [95% CI 1.0-1.7]), late TRM (1.9 [95% CI 1.3-2.9]), grade ≥2 aGVHD (28 versus 43 percent, respectively; HR 1.5 [95% CI 1.2-1.8]), and cGVHD (25 versus 13 percent; HR 1.9 [95% CI 1.6-2.4]). These adverse effects of PBSPC were offset by a lower risk for relapse (HR 0.8 [95% CI 0.6-1.0]). BM and PBSPC did not differ regarding early (ie, ≤6 months) TRM, early overall mortality, treatment failure, or graft failure (11 percent with BM after one year versus 6 percent with PBSPC).

●Survival was superior using BM grafts compared with PBSPC grafts in a registry study of children (8 to 20 years) who received BM (630 patients) versus PBSPC (143 patients) grafts for AML or ALL from 1995 to 2000 [11]. After adjustment for other significant factors, three-year OS with BM grafts was 58 percent versus 48 percent with PBSPC. Compared with BM, PBSPC grafts were associated with higher RR for mortality (1.4 [95% CI 1.1-1.8]), TRM (1.9 [95% CI 1.3-2.8]), and treatment failure (1.3 [95% CI 1.0-1.7]); relapse rates were similar (33 percent with BM versus 38 percent with PBSPC). Grade ≥3 cGVHD at three years was higher with PBSPC (33 versus 19 percent; RR 1.8 [95% CI 1.3-2.6]), but rates of grade ≥2 aGVHD were similar.

●A single-institution study reported similar outcomes between children who received BM (39 patients) versus PBSPC grafts (35 children) for grafts from UDs [12]. OS, TRM, and relapse-free survival did not differ between the two groups, but relapses were more common with BM than PBSPC (48 versus 24 percent).

T cell depletion — T cell depletion can reduce the incidence of GVHD with all donor types, but it is typically avoided for UCB grafts and when the patient has developed MDS or leukemia prior to HCT. Preferred methods of T cell depletion vary among institutions and may differ according to the graft or donor source.

T cell depletion can be achieved by in vivo methods or ex vivo techniques:

●Ex vivo T cell depletion – Ex vivo CD34+ enrichment (ie, positive selection for CD34+ nucleated cells) depletes most T cells, while permitting limited add-back of T cells (eg, ≤1 x 10⁵ CD3+ cells/kg recipient body weight), if desired. TCR alpha/beta depletion retains TCR-gamma-delta cells, which may enhance engraftment and retain GVL effects. CD34+ enrichment has replaced earlier techniques, such as sheep erythrocyte resetting, soybean lectin agglutination, and counterflow centrifugal elutriation.

●In vivo T cell depletion – Post-transplant cyclophosphamide (PTCy) can selectively target rapidly dividing alloreactive T cells and has been employed for in vivo T cell depletion. Other in vivo T cell depletion strategies include anti-thymocyte globulin (ATG) and alemtuzumab (CAMPATH; anti-CD52 monoclonal antibody). These agents can induce cytokine release from the targeted T cells and might slow engraftment or immune recovery.

Conditioning therapy — Conditioning therapy refers to intensive chemotherapy-based treatment that kills malignant cells and enables engraftment of the donor cells.

Note that transplantation in children with certain inherited bone marrow failure syndrome (IBMFS) requires special considerations, as discussed separately. (See "Hematopoietic cell transplantation (HCT) for inherited bone marrow failure syndromes (IBMFS)".)

Types of conditioning — Conditioning regimens can be considered broadly as one of the following:

●Myeloablative conditioning (MAC) – MAC regimens use very intensive chemotherapy that causes complete bone marrow aplasia. MAC regimens exert their antileukemic effect through a combination of the intensive conditioning chemotherapy plus the GVL effect. MAC has been defined as administration of 8 mg/kg of busulfan, melphalan >140 mg/m², or thiotepa >10 mg/kg or total body irradiation (TBI) dose >8 gray (Gy) [13].

There is no preferred MAC regimen, but many include busulfan (Bu) plus cyclophosphamide (Cy), with or without melphalan (Mel). Therapeutic drug monitoring (TDM) of Bu should be used to adjust the dose according to age and serum levels [14]. Some institutions, especially in Europe, prefer treosulfan (Treo) rather than Bu, based on a favorable adverse event profile, less endothelial toxicity, and no need for TDM; Treo is available only as an investigational agent in the United States. TBI is generally avoided because radiation is not a more effective antileukemic agent than chemotherapy, and TBI is associated with increased short-term and long-term toxicity in children.

●Reduced intensity conditioning (RIC) – RIC regimens are less toxic, but they are used less often for transplantation in children. RIC may be considered for children with significant comorbidities who may not be able to tolerate MAC. RIC regimens largely depend on a GVL effect and less on the intensity of the conditioning therapy.

RIC was originally developed for transplantation in older adults (for whom MAC can cause unacceptable toxicity and TRM), but children can generally tolerate MAC. RIC only partially eliminates normal and malignant host cells in the BM. Administration of the allogeneic graft after RIC initially establishes mixed chimerism (ie, co-existence of both host and graft hematopoietic cells) and, over time, the graft cells eliminate both normal and malignant cells via GVL to establish full chimerism.

There is no preferred RIC regimen, but many include Flu plus Bu or Treo and may include Cy or other agents.

Studies that compared outcomes with MAC versus RIC in children are discussed below. (See 'Outcomes with MAC versus RIC' below.)

Choice of conditioning therapy — For children who undergo HCT for AML or MDS, we suggest MAC rather than RIC. Large retrospective studies reported no significant difference in OS or EFS, but MAC is associated with lower rates of cGVHD and acceptable levels of toxicity and TRM in children [2,3]. (See 'Outcomes with MAC versus RIC' below.)

No conditioning regimen has proven to have the most favorable balance of efficacy and toxicity in children with myeloid malignancies. Although MAC is generally preferred in children, the choice of a conditioning regimen may be influenced by comorbid illnesses, prior treatments, age at transplantation, and HLA disparity. As an example, RIC might be favored in a young child who was previously treated with gemtuzumab ozogamicin, both of which are associated with increased risk for SOS [15-17].

An international pediatric AML protocol (MyeChild01) includes random assignment to MAC (BuCy) versus RIC (BuFlu) HCT.

Outcomes with MAC versus RIC — Large registry studies have compared outcomes between MAC and RIC:

●Transplant outcomes were compared after RIC (BuFlu) versus MAC (BuCy) in a CIBMTR registry study that analyzed 978 children transplanted for hematologic malignancies (653 with AML, 150 with MDS) between 2008 and 2014 [2]. RIC was associated with inferior survival (61 percent two-year OS versus 71 percent with MAC); this effect was attributed to shorter survival after relapse for children who had previously received RIC compared with those who had received MAC, according to post-hoc analysis, but this may also reflect more adverse pretransplant prognostic scores in those who received RIC. There was no difference between RIC and MAC in cumulative relapses (31 versus 28 percent, respectively), two-year TRM (13 versus 10 percent), grade ≥3 aGVHD (18 versus 11 percent), or cGVHD (28 versus 23 percent). Notably, older children had more severe adverse effects than younger children, as discussed below.

●An earlier CIBMTR analysis (2000 to 2009) reported no difference in outcomes for children with AML transplanted using RIC (39 patients) versus MAC (141 patients) [3]. Multivariate analysis showed no significant differences between conditioning regimens for five-year OS (45 versus 48 percent, respectively), relapse (39 percent each), TRM (16 percent each), or GVHD. Approximately two-thirds of patients were 11 to 18 years old (the remainder were ≤10 years) for each conditioning regimen, but this study did not report outcomes according to age.

Outcomes with MAC and RIC, according to the underlying malignancy (eg, AML versus MDS) and disease status at the time of transplantation are described below. (See 'Outcomes' below.)

GVHD prophylaxis — Treatment with a calcineurin inhibitor (CNI), with or without other immunosuppressive drugs, is generally given to all transplant recipients to reduce the incidence and/or severity of GVHD. GVHD prophylaxis is discussed separately. (See "Prevention of graft-versus-host disease".)

The choice of a GVHD prophylaxis regimen is influenced by a related versus unrelated donor, degree of HLA-mismatch, graft source (ie, BM versus PBSPC), and use of T cell depletion. Most regimens include a CNI (eg, cyclosporine or tacrolimus), which reduces T cell activation, either alone or in combination with other immunosuppressive drugs or methods for T cell depletion. There is no consensus about a superior GVHD prophylaxis regimen for children, and the preferred regimen varies among institutions.

Evaluation and management of acute GVHD and chronic GVHD are discussed separately. (See "Clinical manifestations, diagnosis, and grading of acute graft-versus-host disease" and "Clinical manifestations and diagnosis of chronic graft-versus-host disease" and "Treatment of acute graft-versus-host disease" and "Treatment of chronic graft-versus-host disease".)

POST-TRANSPLANTATION — Children who undergo HCT are at risk for short-term transplant-related toxicity, late effects of transplantation, and disease relapse. Adverse effects vary with the type of HCT and the age at transplantation, but the true incidence of complications is not well defined because of limited numbers of long-term survivors.

●Short-term adverse effects of HCT include infections, hepatic sinusoidal obstruction syndrome (SOS; also called veno-occlusive disease), transplant-associated thrombotic microangiopathy, interstitial pneumonia, and acute graft-versus-host disease (aGVHD).

●Late toxicity includes chronic graft-versus-host disease (cGVHD), endocrinologic disorders that can lead to growth failure and infertility, neurocognitive abnormalities, and secondary malignancies.

Monitoring and management — Post-HCT management requires monitoring for short-term and long-term transplant-related adverse effects and surveillance for disease relapse.

Follow-up should be individualized based on age, transplantation technique, and medical condition. There is no optimal schedule and protocol for post-transplant monitoring, but general guidelines have been published [18-21]. Transitions from pediatric to adult care and from transplant specialists to primary care providers are especially challenging and demand excellent communication and collaboration.

We monitor the child following HCT as follows:

●Short-term – Follow-up in the first 100 days after transplantation should include:

•Immune reconstitution – Monitoring recovery of blood counts, including lymphocyte subsets.

•Infections

-Screening for viral infections, including viral titers (eg, cytomegalovirus [CMV], Epstein-Barr virus [EBV], adenovirus, BK virus, human herpesvirus-6 [HHV6]). (See "Prevention of viral infections in hematopoietic cell transplant recipients".)

-Prophylaxis should include prophylactic antibiotics for encapsulated organisms and Pneumocystis jirovecii (formerly P. carinii) pneumonia and antiviral agents for rising viral titers or copy numbers. (See "Prevention of infections in hematopoietic cell transplant recipients", section on 'Antimicrobial prophylaxis or pre-emptive therapy'.)

•Acute GVHD – Evaluation and management of acute GVHD are discussed separately. (See "Clinical manifestations, diagnosis, and grading of acute graft-versus-host disease" and "Treatment of acute graft-versus-host disease".)

●Longer-term

•Immunizations per guidelines, according to age. (See "Immunizations in hematopoietic cell transplant candidates and recipients".)

●Growth and development

•Annual height, weight, body mass index (BMI), and Tanner staging, as appropriate.

•Assessment of mental health, chronic pain and fatigue, risky behaviors, and health care access.

●Endocrine

•Assess thyroid function and bone age annually.

•Assessment of pubertal development and sexual and reproductive function annually.

•Luteinizing hormone (LH), follicle stimulating hormone (FSH), and testosterone in males; consider semen analysis.

•LH, FSH, and estradiol in females.

●Cancer screening – Monitoring should include:

•Follow general population cancer screening guidelines.

•Counsel regarding tobacco avoidance or cessation.

•Regular dental exams for oral cancers for patients with cGVHD.

•Mammography to begin at ≤40 years in women or beginning at age 25 for children who received radiation therapy (RT; or 8 years after total body irradiation [TBI] or chest radiation).

•Thyroid cancer screening for children who received RT.

●Neurocognitive – Screening for neurocognitive deficits include:

•Yearly screening for educational and developmental progress.

•Neuropsychological evaluation at minimum one year post-HCT (repeated as needed) and referral for those with neurocognitive deficits.

●Dental – Regular dental examinations for those with cGVHD.

●Chronic GVHD – Evaluation and management of cGVHD are discussed separately. (See "Clinical manifestations and diagnosis of chronic graft-versus-host disease" and "Treatment of chronic graft-versus-host disease".)

Long-term effects — The potential to cure myeloid malignancies with HCT is offset by long-term adverse effects. Long-term sequelae of HCT include impaired intellectual and psychomotor functioning; neuroendocrine abnormalities; reduced fertility; cardiac, pulmonary, and renal complications; second malignancies, and impaired quality-of-life [22,23]. Late adverse effects of brain irradiation (eg, neurocognitive deficits, growth hormone deficiency, secondary central nervous system tumor) especially affect younger children.

The incidence and severity of late effects vary according to whether myeloablative conditioning (MAC) or reduced intensity conditioning (RIC) was used.

●Endocrine – Children receiving HCT are at increased risk for endocrine disorders [18,22,24]. Most of these abnormalities are due to organ damage by chemotherapy or RT, but hypothalamic/pituitary dysfunction can be seen among patients who have received cranial radiation. Primary hypothyroidism is seen in one-third to one-half of children [18]. HCT survivors are at increased risk of developing metabolic syndrome with adiposity, dyslipidemia, glucose intolerance, and hypertension.

•Gonadal – Gonadal dysfunction can lead to abnormal pubertal development, infertility, and sexual dysfunction, but it has decreased with contemporary methods [22]. Alkylating agents and RT increase the risk of ovarian failure in adult women; addition of busulfan to cyclophosphamide causes permanent ovarian failure in nearly all females [25]. Most males can have permanent infertility after MAC.

•Growth impairment – Impaired growth can be caused by cGVHD, malnutrition, treatment with glucocorticoids, and growth hormone deficiency [18]. Growth impairment has been reported in more than half of children undergoing HCT. MAC using busulfan/cyclophosphamide or cyclophosphamide/TBI conditioning regimens can cause growth impairment in children.

●Infections – Infectious complications are important causes of late morbidity and mortality after HCT, with an estimated eightfold increase compared with population norms [26]. Patients with active cGVHD are at risk of life-threatening infections with encapsulated organisms, viruses, fungi, and other opportunistic organisms, as they are considered functionally asplenic and have impaired lymphocyte function [27]. Even after GVHD has resolved, some patients continue to have hypogammaglobulinemia, secretory immunoglobulin A (IgA) deficiency, and/or recurrent chronic sino-pulmonary infections for months or years after HCT.

●Second malignancies – Second cancers can arise as therapy-related MDS/AML (t-MDS/AML), lymphoma (including post-transplant lymphoproliferative disorders), and solid tumors [18]. The risk of second malignancies is higher in children with AML who received HCT (6.7 to 11.6-fold) compared with patients given chemotherapy only [28,29].

OUTCOMES — Outcomes with allogeneic HCT for AML or MDS in children have improved over recent decades, according to data from successive cohorts in the Center for International Blood and Marrow Transplant Research (CIBMTR) registry [1].

Outcomes vary according to the clinical setting (eg, relapsed AML, refractory AML, complete remission [CR]), patient characteristics (eg, age, performance status), conditioning regimen, donor source, and other factors.

Patient-related factors

●Age – Rates of complications and transplant-related mortality (TRM) vary with the age at transplantation.

•Adolescent/young adults (AYA) – AYA patients generally have higher rates of TRM compared with younger children, especially for those receiving myeloablative conditioning (MAC) HCT.

As an example, a CIBMTR registry study that included MAC and reduced intensity conditioning (RIC) HCT reported that age 10 to 18 years was independently associated with increased TRM but lower rates of relapse [2]. Compared with children ≤9 years, for older children the relative risk (RR) for TRM was 2.5 (95% CI 1.6-3.9) and RR for relapse was 0.6 (95% CI 0.5-0.8). In a study that used MAC HCT, age ≥12 years was associated with 31 percent TRM compared with 9 percent in younger children; by contrast, TRM for children receiving RIC HCT was 15 percent and did not differ according to age at the time of transplant [30].

•Infants – Infants (eg, <2 years) have increased rates of hepatic sinusoidal obstructive syndrome (SOS); the risk is heightened for infants who receive human leukocyte antigen (HLA)-discordant grafts, receive MAC with radiation therapy (RT) in the conditioning regimen, have pre-existent liver disease, or who previously received gemtuzumab ozogamicin. (See "Hepatic sinusoidal obstruction syndrome (veno-occlusive disease) in children".)

●Performance status (PS) – Analysis of transplantation by CIBMTR reported that, compared with children who had Karnofsky PS (KPS) 90 to 100, those with KPS <90 had impaired overall survival (OS; HR 3.22 [95% CI 2.01-5.16]), relapse (HR 3.00 [95% CI 1.75-5.13]), and treatment failure (HR 2.98 [95% CI 1.86-4.70]) [3].

HCT-related factors — Adverse effects, TRM, relapse, and other outcomes vary according to transplantation factors.

Factors that influence outcomes with HCT for AML include:

●Donor type and HLA match – (See 'Matched sibling donor' above and 'Alternative donors' above.)

●Graft source – (See 'Graft source' above.)

●T cell depletion – (See 'T cell depletion' above.)

●Conditioning regimen – (See 'Conditioning therapy' above.)

AML — Outcomes vary with the clinical setting:

●First complete remission – HCT in first CR is generally limited to children with adverse prognostic features and/or persistent measurable residual disease (MRD) [31-40]. For transplantation of a child in first CR, outcomes with allogeneic HCT are superior to those with autologous HCT or consolidation chemotherapy. Indications for HCT for children in first CR are described separately. (See "Acute myeloid leukemia in children and adolescents", section on 'Prognostic factors'.)

●Relapsed/refractory AML – Allogeneic HCT for children with refractory or relapsed AML is associated with long-term survival in approximately half of patients who achieved CR after salvage chemotherapy, but outcomes are inferior for children with active disease at the time of transplantation. Outcomes have improved over successive decades; as an example, five-year OS for children with relapsed AML improved to 45 percent in 2005 to 2010 compared with 28 percent in 1999 to 2004 [41].

Transplantation when the child is in remission is more successful than HCT with active relapsed/refractory AML.

•Five-year OS was greater for children transplanted in second CR (47 percent), compared with relapsed AML (28 percent) or refractory AML (17 percent), in a National Marrow Donor Program (NMDP) analysis [42]. Rates of mortality, treatment failure, and relapse were superior for HCT in second CR compared with transplantation with active disease. TRM was higher for children 11 to 18 years, compared with younger children (HR 2.20 [95% CI 1.35-3.60]).

A retrospective cooperative group study reported that MAC HCT for 152 children in second CR was associated with 62 percent five-year OS, 35 percent relapse, and 12 percent TRM [43]. Only 10 percent received matched sibling donor (MSD) grafts, while half received matched unrelated donor (MUD) grafts and the remainder received mismatched grafts. By contrast, a Japanese registry study reported 23 percent three-year OS in 417 children with refractory AML who were not in remission at the time of transplantation [7].

MDS — HCT using MAC can cure MDS in more than half of patients, but outcomes vary with the indication for transplantation (ie, refractory cytopenias versus progression toward AML).

HCT performed for MDS with refractory cytopenias is associated with better outcomes than HCT for MDS with high risk of progression to AML. A report from the European Working Group (EWOG) MDS/EBMT group reported outcomes of MAC HCT in 89 children; 28 had refractory cytopenias and the remainder had MDS with higher-risk features [44]. Event-free survival (EFS) was better for children with refractory cytopenias (75 percent) than for the others (50 to 60 percent).

It is uncertain if intensive induction chemotherapy prior to HCT improves outcomes for children with MDS, compared with proceeding directly to HCT:

●Treatment of MDS with intensive chemotherapy before HCT did not improve OS or EFS, compared with children who proceeded directly to transplantation [45]. In multivariate analysis, three-year OS was superior in children who did not receive pretransplant chemotherapy (RR of death 0.29 [95% CI 0.04-0.78]) and in those transplanted earlier (ie, <140 days after diagnosis), but these differences may reflect a larger proportion of children transplanted for cytopenias in the non-chemotherapy group.

●In the EWOG MDS/EBMT study (described above) there was no difference in EFS for children with higher-risk MDS who received AML-type induction therapy prior to HCT versus those who did not [44]. Other studies also reported no differences based on front-line HCT versus pretransplant induction chemotherapy [4,46].

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: Acute myeloid leukemia" and "Society guideline links: Myelodysplastic syndromes".)

SUMMARY AND RECOMMENDATIONS

●Allogeneic hematopoietic cell transplantation (HCT) in children – Allogeneic HCT can achieve long-term remission/cure in children with acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS), but ongoing growth and development and decades of post-transplant survival require considerations that distinguish HCT in children. (See 'Transplantation in children' above.)

Autologous HCT is not used for children with AML/MDS.

●Exclude inherited disorders as the cause of AML/MDS – It is important to exclude an inherited bone marrow failure syndrome (IBMFS) as the cause of AML/MDS in children, as these conditions require distinct considerations for transplantation and its complications, as discussed separately. (See "Hematopoietic cell transplantation (HCT) for inherited bone marrow failure syndromes (IBMFS)".)

●Graft-versus-host disease (GVHD) and graft-versus-leukemia (GVL) – Immunologic mismatches between the transplanted hematopoietic cells ("graft") and the transplant recipient ("host") cause the graft to recognize the host as "foreign" and result in the GVL effect, which kills the malignant cells, but also causes GVHD, a major source of toxicity and transplant-related mortality (TRM). (See 'GVHD and GVL' above.)

●Graft donor

•Donor search – The donor search should begin promptly after diagnosis of AML/MDS. The search should begin with human leukocyte antigen (HLA) typing of the patient, siblings, and parents, using molecular techniques for matching 10 HLA class I and II antigens. If a suitable sibling donor is not identified, the search for an alternative donor should be initiated promptly. (See 'Donor search' above.)

•Donor choice

-Matched sibling donor (MSD) grafts are preferred for transplantation in children. (See 'Matched sibling donor' above.)

-Matched unrelated donor (MUD) – For transplant candidates without a suitable MSD, we suggest an HLA-MUD (ie, ≥9 of 10 HLA alleles) or HLA-matched related donor, if available, rather than other alternative donors (Grade 2C). (See 'Alternative donors' above.)

●Graft source – For HCT in children, we suggest bone marrow (BM) as the graft source, rather than peripheral blood stem/progenitor cells (PBSPC) (Grade 2C). However, when sufficient BM is unavailable, use of PBSPC is acceptable. (See 'Graft source' above.)

●Conditioning regimen – For children who undergo HCT, we suggest myeloablative conditioning (MAC), rather than reduced intensity conditioning (RIC) (Grade 2C). (See 'Choice of conditioning therapy' above.)

●Post-transplantation – Management must consider short-term and late effects of transplantation, and disease relapse. (See 'Post-transplantation' above.)

•Monitoring and management – (See 'Monitoring and management' above.)

-Short-term (eg, first 100 days) includes immune reconstitution, screening and prophylaxis for infections, and management of acute GVHD.

-Longer-term includes age-appropriate immunizations; assessment of growth and development, neurocognitive function, organ dysfunction; evaluation for endocrine dysfunction; management of chronic GVHD; and cancer screening.

●Outcomes – Outcomes with HCT in children vary with:

•Patient-related factors – (See 'Patient-related factors' above.)

•Transplant-related factors – (See 'HCT-related factors' above.)

•AML – (See 'AML' above.)

•MDS – (See 'MDS' above.)