Immunologic recovery and basis for infections in the pediatric hematopoietic stem cell transplant recipient

Pre-engraftment period

The pre-engraftment period is often considered to correspond to days 0 to 30 after HSCT; however, this period can include days before neutrophil engraftment as well as days soon after. The infection risk in this early period is attributed to many factors, including neutropenia, the breakdown of mucosal barriers leading to subsequent microbial invasion, and acute graft-versus-host disease (GVHD) leading to further barrier breakdown and immunosuppression. Patients receiving myeloablative conditioning are most vulnerable in this early period owing to longer durations of neutropenia and mucosal barrier injury associated with myeloablative regimens. The majority of infections are monomicrobial and are most often the result of bacterial pathogens. The most common isolated organisms include coagulase-negative staphylococci, Enterococcus species, Staphylococcus aureus , or enteric gram-negative bacilli. The portal of entry for many of these organisms is either via a central venous catheter or translocation of a compromised mucosal barrier, often secondary to myeloablative conditioning regimens. Oral mucosal barrier injury can predispose specifically to viridans group streptococci; however, mucosal barrier injury can include any part of the alimentary tract from the oral cavity to the anus. Clostridium difficile infection is one of the most common causes of infectious diarrhea after HSCT given frequent exposure to precipitating factors such as chemotherapy and antibiotics.

Invasive fungal disease, either yeast or molds, tends to occur in a biphasic pattern after HSCT with the first risk period presenting in this pre-engraftment phase. The most common fungal pathogens are Candida or Aspergillus species. More recent use of supportive care such as antifungal prophylaxis has led to a reduction of these events in the pre-engraftment period; however, breakthrough invasive fungal disease from genera other than Candida or Aspergillus may occur and, more rarely, resistance may occur.

The next most frequently involved pathogens are viruses, which commonly include cytomegalovirus (CMV), herpesviruses, adenovirus, BK virus, respiratory viruses, and gastrointestinal viruses. Antiviral prophylaxis for some of these viral infections may delay or postpone the presentation into the early post-engraftment or late post-engraftment periods. CMV is the most common viral infection encountered in the posttransplant period. Antiviral prophylaxis has limited utility in the immediate posttransplant period owing to associated toxicities with current options. This differs from varicella zoster virus and herpes simplex virus, members of the herpesvirus family frequently encountered after transplant, for which institutions often use antiviral prophylaxis with acyclovir for the first year to prevent reactivation. ^, Other herpesvirus infections (Epstein-Barr virus, human herpes virus type 6, and, less commonly, human herpes virus type 7 and human herpes virus type 8) can also be identified in this period but far less frequently. Although adenovirus and BK virus are not members of the herpesvirus family, these viruses can maintain a persistent asymptomatic state before transplantation and reactivation can occur in the early and late posttransplant periods. Lastly, hospital- and community-acquired respiratory (i.e., respiratory syncytial virus, rhinovirus, influenza) and gastrointestinal (i.e., norovirus, astrovirus) viral infections can be encountered in this early posttransplant period, most without any adequate treatment or prophylaxis, and can be devastating.

Early post-engraftment period

In the early post-engraftment period, neutropenia has resolved, which is an important milestone for reduced vulnerability to bacterial and fungal pathogens. However, the clinician should still be alert to opportunistic infections from these pathogens owing to risk factors, such as continued need for central venous access, residual mucositis, intermittent neutropenia secondary to medication toxicity, or more rarely, graft loss. Persistent lymphopenia and slow T-cell reconstitution is primarily responsible for vulnerability to infections during this period. Poor T-cell reconstitution can be further delayed by the need for immune suppressive agents to manage HSCT complications such as acute GVHD. This combination of poor T-cell function and need for additional immune suppression predisposes children to an increased risk of latent viral reactivations, poor outcomes from typically self-limiting primary viral infections, and invasive mold disease. Additionally, this period is an important window of risk for Pneumocystis jirovecii pneumonia; therefore all patients continue prophylaxis through this period and often until T-cell reconstitution.

Late post-engraftment period

The late post-engraftment period starts 100 days after HSCT but varies in duration owing to the individual patient requirement for ongoing immunosuppression and delayed immune reconstitution. Patients may remain at high risk for infections because of prolonged immunosuppression secondary to treatment for chronic GVHD or autoimmune cytopenias. This results in delayed reconstitution of both cellular and humoral immunity. Bacteremia, sinusitis, upper respiratory tract infections, pneumonia, and meningitis are not infrequently caused by encapsulated bacteria ( Streptococcus pneumoniae , Haemophilus influenzae , Neisseria meningitidis ) during this period. Patients with chronic GVHD are particularly susceptible to these infections owing to poor opsonophagocytosis and hyposplenism, referred to as functional asplenia. Therefore some institutions initiate antibiotic prophylaxis for functional asplenia to prevent overwhelming bacterial sepsis. In addition to encapsulated organisms, bacteremia during this period may also result from Staphylococcus species or gram-negative bacteria. Factors that may predispose patients to bacteremia from these pathogens include the continued presence of central venous access or persistent mucosal barrier dysfunction. Although the peak of reactivation of latent viruses is in the early post-engraftment period, the risk persists through this late phase. For Epstein-Barr virus, reactivation can lead to the development of posttransplant lymphoproliferative disease, which typically presents between 3 and 5 months after transplant. Other atypical late post-engraftment infections may be due to Nocardia species, Listeria species, Cryptococcus species, and nontuberculous mycobacteria. Finally, the risk for P. jirovecii pneumonia can remain well after 100 days from HSCT, particularly if continued immune suppression is required.

Approaches to prophylaxis relative to infectious risk periods

In general, there are three methods to provide posttransplant antimicrobial prophylaxis: pharmacologic prophylaxis, immunoprophylaxis (immunoglobulin [IG] replacement therapy), and vaccinations. For each method, there are two approaches to guide initiation or duration of prophylaxis: a uniform time at risk approach or an individualized approach that takes into account ongoing immunosuppression and immune reconstitution. With the uniform time-at-risk approach, antimicrobial prophylaxis is continued until a designated time period elapses. An example would be the use of antifungal prophylaxis until day 100 (which may be extended for patients requiring immunosuppressive treatment for GVHD). Much of the data on the effectiveness for prophylaxis have evolved from trials designed with this simpler standardized time-at-risk approach. Although it has the advantage of consistency and ease of use, this approach likely results in overtreatment of some patients and undertreatment of others.

The second approach is individualization of the duration of time during which prophylaxis is provided or of the timing at which vaccinations are administered based on an assessment of an individual patient’s cellular and humoral immunity. This system is much more cumbersome and has less evidence to support its use; however, it should theoretically result in earlier discontinuation of prophylaxis for some patients with adequate immune reconstitution and appropriately prolong prophylaxis for patients with immune defects that persist beyond an estimated risk period duration from transplantation. Examples of this approach include continuation of antifungal or antiviral prophylaxis until patients achieve functional cluster of differentiation (CD)4 ⁺ T-cell reconstitution or discontinuation of IG replacement therapy when patients have adequate CD19 ⁺ B cells, switched memory B cells, and evidence of IgM and IgA production. Whether a center uses the fixed time-at-risk approach or a more individualized approach will depend on the infrastructure of the transplant center and its ability to consistently apply a more nuanced approach to prophylaxis.

After HSCT and Ig replacement therapy is discontinued, recipients must be revaccinated. There are limited data on vaccine efficacy and ideal timing of vaccinations in HSCT recipients; however, it is accepted that there must be at least partial recovery of T and B cells before administration. Vaccination with polysaccharide antigen vaccines elicits T-cell–independent antibody responses and therefore typically fails to produce protective immunity in most allogeneic HSCT recipients within the first year after transplantation. However, conjugate vaccines evoke T-cell–dependent antibody responses and produce protective antibody responses within the first year after allogeneic HSCT even with patients receiving immunosuppression. Therefore most revaccination guidelines are based on timing from transplantation, and HSCT recipients could undergo early revaccination with conjugate vaccines analogous to newborn vaccination schedules and achieve protective long term immunity. Vaccination with inactivated or toxoid-containing vaccines is recommended as early as 3 to 6 months after HSCT, whereas administration of live-attenuated vaccines is recommended at 24 months after HSCT. The delayed use of live-attenuated vaccines is based on concerns about transmission of vaccine-mediated disease and limited clinical data on safety or immunogenicity of earlier vaccination. The Advisory Committee on Immunization Practices to the Center for Disease Control and Prevention and the Infectious Diseases Society of America publish time-based guidelines on vaccination after transplantation, but vaccine schedules vary greatly among institutions.

Innate immune recovery after transplantation

The innate immune system can be divided into nonhematopoietic and hematopoietic compartments. The nonhematopoietic compartment includes physical barriers, such as the skin and mucosal surfaces, which can be damaged during transplantation by the pretransplant conditioning regimen. However, the damage is typically restored soon after transplantation. Repair can be inhibited by GVHD of the skin or mucous membranes. The skin and mucosal barriers can also be compromised by the presence of foreign material, such as a central venous catheter or a gastrostomy tube, for prolonged periods after transplantation.

Hematopoietic innate immune cells include neutrophils, macrophages, as well as natural killer (NK) cells. After myeloablative conditioning, patients undergo an aplastic phase, which is identified by severe neutropenia, anemia, and thrombocytopenia. The first laboratory sign of hematologic recovery is typically neutrophil recovery. Engraftment, classically defined as absolute neutrophil count greater than 500/μL, is typically achieved between 10 and 42 days and transplant, depending on the stem cell source (see Table 2.1 ). Hematopoietic growth factors, such as granulocyte colony-stimulating factor (G-CSF), can be used to accelerate recovery of granulocyte counts, minimize the duration of neutropenia, and decrease the risk for severe infections. The use of G-CSF after HCT is universal in the autologous setting but is more controversial in allogeneic graft recipients owing to a lack of benefit in reducing mortality. Most centers use G-CSF for recipients of umbilical cord blood (UCB) transplants, whereas its use in others is more variable as there are concerns that G-CSF may increase rates of GVHD or malignant relapse. However, administration of certain post-HSCT medications (such as ganciclovir or valganciclovir) may result in secondary neutropenia. Furthermore, the neutrophils may have abnormal function for up to 2 months after transplant.

Monocytes are leukocytes that circulate peripherally until they eventually migrate into tissues where they develop into macrophages and dendritic cells. Monocytes, macrophages, and dendritic cells function through phagocytosis, a process that is particularly important for pathogen killing and tissue repair. However, mononuclear phagocytes also activate the adaptive immune system via antigen presentation and cytokine production. Posttransplant neutrophil recovery is occasionally preceded by the detection of peripheral monocytes; however, monocyte function may remain suboptimal for up to 1 year after transplant. Although monocyte function is difficult to measure in clinical laboratories, based on data extrapolated from animal models, it is thought that tissue macrophages and dendritic cells are not significantly depleted as a result of transplant conditioning, and natural turnover results in their being gradually replaced by donor-derived cells for up to a year after HSCT. Peripheral dendritic cells can also be detected at the time of neutrophil count recovery and a low dendritic cell count at engraftment may predict relapse, death, and acute GVHD.

NK cells are also key members of the innate immune system that influence adaptive function. Derived from the lymphoid lineage, NK cells have a unique function in the prevention of viral infections and antitumor immunity. NK cell reconstitution varies by graft source. NK cells recover in number and function in the first few weeks after allogeneic HSCT, much earlier than T- and B-cell reconstitution, which is likely related to the high IL-2 levels during the early posttransplant period. Not only do NK cells appear early, they also acquire functional competence much earlier than other lymphocytes. Furthermore, NK cell activity after HSCT remains normal even in the presence of severe GVHD.

Delay in neutrophil engraftment greatly increases morbidity, and failure of sustained neutrophil engraftment after a myeloablative-conditioning regimen requires urgent retransplantation. Engraftment failure can occur from inadequate hematopoietic stem cell quantity from poor collection or loss in postcollection processing, inadequate host support of infused cells, posttransplantation events such as infection or medications, or from host-versus-graft (immunologic) rejection. Engraftment failure is a very rare complication of autologous transplantation and is likely only present in the setting of poor cryopreservation of stem cells. For allogeneic HSCT, the risk of engraftment failure is dependent on a number of variables, including baseline host immunity, HLA disparity, type of conditioning regimen and stem cell source used, low stem cell dose, ex vivo T-cell depletion (TCD), ABO incompatibility, and disease status at transplantation. Patients with hematologic malignancies have a rejection risk of approximately 5%, whereas the risk may be greater than 10% in patients with nonmalignant conditions. Graft rejection is more common in haploidentical related and mismatched unrelated donor (URD) transplants and much less frequent in matched sibling donor transplantation. Generally, UCB transplant recipients have the highest risk of graft failure, whereas peripheral blood stem cell grafts have the lowest risk of graft rejection. The incidence of graft failure also varies considerably among institutions owing to differing approaches to conditioning.

Adaptive cellular immune recovery after transplantation

Often clinicians are reassured about a patient’s infection risk once neutrophil engraftment is achieved. Although neutrophil recovery is an important milestone, patients continue to be vulnerable to opportunistic infection because of persistent cellular immunodeficiencies involving the adaptive immune system. The recovery of the adaptive immune system is much more nuanced, involving refinement and adjustment of T and B lymphocytes over the lifetime of an individual. After HSCT, T and B lymphocytes reconstitute slowly and develop both a cellular and humoral response. The cellular immune response to pathogens is initiated by antigen-presenting cells (e.g., macrophages and dendritic cells) but also requires the presence of functional T cells for activation. HSCT results in impairment of the adaptive immune response through loss of naïve T cells and reduced function of existing T cells. The recovery of the T-cell compartment initially relies on peripheral expansion of infused donor memory T cells, which leads to a narrow T-cell receptor repertoire. This process is driven by cytokines, such as IL-7 and IL-15, as well as by antigen stimulation and T-cell receptor (TCR) engagement. Peripheral T-cell expansion is eventually followed by the production of naïve T cells in the thymus leading to a population of memory T cells with a diverse TCR repertoire. In patients receiving T-cell–replete grafts, peripheral expansion of infused memory T cells with a limited repertoire occurs initially until hematopoietic progenitors seed the thymus and produce T cells with a more diverse repertoire. In T-cell–depleted transplants, seeding of the thymus by hematopoietic progenitors is the primary route to T-cell reconstitution.

In either route, effective long-term and sustained T-cell lymphopoiesis is dependent on the presence of a functional thymus. Thymic dependence in generating a diverse T-cell repertoire after HSCT is a critical hurdle in patients without a thymus or with a poorly functioning thymus. Aging, recurrent or chronic infections, chemotherapy, radiation exposure, and GVHD can all lead to thymic atrophy and subsequent difficulty with T-cell reconstitution. Detection of recent thymic immigrants and T-cell polyclonality are typical methods used to determine thymic function. TCR excision circles (TRECs) are small circularized portions of DNA created through T-cell maturation in the thymus, which can be used as a surrogate marker for reconstitution of thymus-derived CD4 ⁺ CD45RA ⁺ naïve T cells. However, these are just markers for thymic output and, in general, it is difficult to completely assess the function of the thymus after transplantation.

Naïve T-cell populations are usually reduced for long periods after HSCT. The inability to reconstitute the naïve T-cell compartment for several years after HSCT, in the absence of GVHD, is likely a consequence of both thymic dysfunction and impaired peripheral naïve T-cell homeostatic mechanisms and survival. CD4 ⁺ lymphocytes require a functional thymus for generation of CD4 ⁺ CD45RA ⁺ naïve T cells, whereas CD8 ⁺ lymphocytes are predominantly derived by clonal expansion outside the thymus. Therefore CD4 ⁺ T lymphocytes appear later than CD8 ⁺ T lymphocytes leading to the inversion of the CD4/CD8 ratio found after transplant. Inversion of the CD4/CD8 ratio is one of the earliest features of T-cell reconstitution after autologous or allogeneic transplantation from any graft source and can persist for up to several years after HSCT. . CD4 ⁺ T-cell reconstitution to a level of approximately 200/μL typically occurs around 3 months after HSCT but can vary considerably depending on the use of TCD methods, graft source (UCB), receipt of total body irradiation, or development of GVHD.

The development of regulatory T cells (Tregs) may be important in determining outcomes after allogeneic HSCT. Tregs suppress the activity of effector T cells, thus reducing inflammation and promoting immune homeostasis after allogeneic HSCT. The presence of donor Tregs enhances immune reconstitution and improves TCR diversity after transplantation. Increased donor Tregs are associated with a decreased risk of GVHD, and many studies have shown that Tregs are significantly reduced in HSCT recipients with GVHD. The relative predominance of effector T cells, compared with Tregs, leads to a proinflammatory milieu of cytokines. IL-6, characterized as both proinflammatory and antiinflammatory, is of particular interest in GVHD and moderates the differentiation of naïve T cells into either Tregs or effector T cells. IL-6 blockade promotes differentiation into Tregs and may mitigate the severity of GVHD. Tumor necrosis factor alpha (TNF-α) is typically classified as a proinflammatory cytokine; however, it may also have antiinflammatory properties mediated through its effects on Tregs. TNF-α has been shown to increase expansion, stability, and possibly the function of Tregs, and may therefore have conflicting effects on both GVHD incidence (high is bad early after HSCT) and severity (low is bad later after HSCT).

Adaptive humoral immune recovery after transplantation

The adaptive humoral immune response is mediated by antibodies and requires both functional T and B cells. In addition to the delayed recovery of T cells after HSCT, there is also impaired reconstitution of B cells. Impairment of B-lymphocyte number and function leads to absent Ig production and susceptibility to infections with encapsulated bacteria such as S. pneumoniae and H. influenzae . The B-cell compartment is the slowest to reconstitute. B-cell reconstitution depends on the intensity of conditioning. Typically, when myeloablative conditioning is administered, all B cells are genetically donor in origin but produced de novo from the bone marrow, and therefore do not retain immunologic memory from the donor. In reduced intensity conditioning, B cells may be of mixed host and donor origin, although data are lacking on whether the persistence of host B cells provides a bridge of immunologic memory. Typically, B-cell reconstitution occurs within 12 months but may take several years for complete development of memory B cells after allogeneic HSCT. Hematopoietic stem cells (HSCs) within the bone marrow undergo multiple stages of B-cell differentiation. Pro-B cells develop into pre-B cells and finally immature/transitional B cells. Transitional (CD19 ⁺ CD21 ^low CD38 ^high ) B cells are the first B cells emigrating from the bone marrow and are elevated in the peripheral blood in the first months after HSCT before progressively decreasing. Transitional B cells emigrate to the spleen where they differentiate into IgM ⁺ memory or mature B cells. Mature B cells migrate to the primary follicle of the lymph node and spleen for antigen exposure and differentiation into switched memory B cells or plasma cells. Reconstitution of switched memory B cells occurs upon antigen exposure from pathogens, the environment, or vaccines and requires CD4 ⁺ T-cell help for isotype switching. Therefore although naïve B cells reach normal levels by approximately 6 months after allogeneic HSCT, levels of IgM ⁺ memory B cells can remain low for up to 2 years. Much like the TCR repertoire, B-cell antibody diversity is severely diminished and suffers prolonged recovery, which is worsened by GVHD and by the medications used to treat it. Recovery of the B-cell count or specific antibody production is primarily of donor origin but can vary among types of allogeneic stem cell grafts, CD34 ⁺ cell doses, donor ages, or recipient ages.

After myeloablative conditioning, B cells are typically entirely of donor origin; however, plasma cells remain primarily of host origin in the first several months after transplant. Given the long-lived nature of plasma cells, it takes months to years to replace host plasma cells by newly produced donor plasma cells. Therefore institutions may consider continuation of Ig replacement therapy in HSCT recipients until there is adequate evidence of B-cell and plasma cell function as opposed to a predetermined period. Some considerations affecting discontinuation of Ig replacement may include absolute B-lymphocyte count, B-lymphocyte phenotyping, including percent of switched memory B cells (CD27 ⁺ IgM ⁻ IgD ⁻ ), IgM and IgA production, and isohemagglutinin production.