Advances in immunology have led to a breathtaking expansion of recognized primary immunodeficiency diseases (PID) with over 120 disease-related genes identified. In North America alone more than 1000 children have received allogeneic blood or marrow transplant over the past 30 years, with the majority surviving long term. This review presents results and highlights challenges and notable advances, including novel less toxic conditioning regimens, to transplant the more common and severe forms of PID. HLA-matched sibling donors remain the ideal option, however, advances in living donor unrelated HSCT and banked umbilical cord blood grafts provide hope for all children with severe PID.
Hematopoietic stem cell transplantation (HSCT) has emerged over the past 50 years as a life-saving therapy for many human diseases. Most recipients of allogeneic HSCT suffer from acquired disorders: leukemia, lymphoma, or aplastic anemia. From the beginning of HSCT use, however, it became apparent that adoptive transfer of healthy marrow and the progeny of donor bone marrow–derived hematopoietic stem cells lead to full reconstitution of the immune system. This recognition led to the use of HSCT for treatment of severe forms of inherited cellular immunodeficiencies. As early as 1968, only a year after the discovery of the human major histocompatibility complex (MHC) locus, a child with severe combined immunodeficiency (SCID) and another with Wiskott-Aldrich syndrome (WAS) were transplanted successfully from their histocompatible siblings. The infant boy transplanted for SCID is the first long-term survivor of HSCT. SCID and WAS represent two common forms of the so-called primary immunodeficiency diseases (PIDs). Rapid advances in molecular immunology over the past 15 years have led to a breathtaking expansion of recognized PIDs. In most cases, PIDs are inherited as a result of monogenic defects with more than 120 disease-related genes recognized to date that manifest in more than 150 distinct clinical entities.
Currently, up to 10 new genetic defects of PID are recognized each year, requiring regular adjustment to the incidence of PIDs. A recent telephone survey of 10,000 households with nearly 27,000 household members by the United States Immunodeficiency Network suggests a population prevalence of diagnosed PID in the United States at approximately 1 in approximately 1200 persons, rendering this diagnosis more common than previously suggested.
The classical and severe forms of PID manifest with life-threatening infections during the first years of life as a result of blockage in T-lymphocyte (SCID) or B-lymphocyte (X-linked agammaglobulinemia) development. Antibody deficiencies due to abnormal B cell numbers or function can be effectively treated by replacement intravenous (IV) immunoglobulin infusions to prevent the recurrent bacterial respiratory tract infections. This is in contrast to T-lymphocyte defects that generally cannot be corrected without HSCT. This review does not discuss the promising results and associated challenges of gene therapy and thymus transplantation but rather focuses on the scientific and clinical advances in HSCT for the more common and severe forms of PID ( Table 1 ).
| Classical SCID syndromes | |
| Nonclassical “leaky” SCID/CID syndromes | BLS |
| OS | |
| CHH | |
| PNP deficiency | |
| ZAP70 deficiency | |
| Other well-defined non-SCID/CID syndromes | WAS |
| HIGM | |
| Congenital defects of phagocyte numbers/function | CGD |
| LADs | |
| MSMD | |
| Schwachman-Diamond syndrome | |
| SCN | |
| Diseases of immune dysregulation | HLH |
| Griscelli syndrome | |
| X-linked LPD | |
| CHS | |
| ALPS syndrome | |
| IPEX syndrome |
Several challenges and controversies face clinical immunologists deciding when or whether or not to refer patients for transplantation and when facing transplanters choosing between sources of grafts and conditioning regimens. Until the late 1990s, almost all HSCT was performed after myeloablative conditioning (MAC) regimens with busulfan and cyclophosphamide or no chemotherapy conditioning at all. Over the past 10 years, however, reduced intensity conditioning (RIC) protocols have been introduced to reduce transplant-related mortality (TRM) and morbidity for children already experiencing liver, lung, and gastrointestinal complications of their underlying PID but nevertheless capable of rejecting a graft with residual natural killer (NK) cells and T lymphocytes. HLA MSDs remain the gold standard to which all alternative grafts are compared, but careful selection of donor sources and preparatory regimens has improved outcome with alternative donors. Some centers of excellence have already reported data for several PID syndromes WAS, hemophagocytic lymphohistiocytosis ([HLH], and leukocyte adhesion deficiency [LAD]) demonstrating that matched unrelated donor (MUD) recipients have comparable outcomes to those receiving MSD marrow grafts. Nevertheless, significant differences will likely remain between centers based on their expertise and preferences when deciding between alternative donor grafts, in their use or avoidance of T cell depletion and even in graft-versus-host disease (GVHD) prophylaxis. Recent advances in unrelated cord blood (UCB) transplantation have made it possible for virtually all children with PID to find a suitable graft donor. The newly formed Primary Immune Deficiency Transplant Consortium recently surveyed 34 pediatric transplant centers in North America and reported that more than 1000 children have received allogeneic HSCT for one of three common PID syndromes: SCID, WAS, and chronic granulomatous disease (CGD). Significantly, approximately 750 of these children are alive.
Severe combined immunodeficiency syndromes
SCIDs constitute a heterogeneous group of genetically determined diseases impairing the T-cell differentiation program. Overall, the incidence of typical SCIDs is estimated to be 1 in 500,000. Fifteen distinct SCID conditions have been fully described by genetic and molecular analysis to date and this will likely increase. Even though all SCID patients experience an intrinsic impairment in T-cell development, there is marked heterogeneity due to variable differentiation of other hematopoietic cell lineages (ie, NK lymphocytes, B-lymphocytes, and neutrophils). In addition, in some of the diseases, developmental defects, such as microcephaly, deafness, neurologic abnormalities, and intestinal atresia, can be associated, raising specific issues in therapy. The absence of adaptive immunity is responsible for a fairly uniform clinical presentation, however, characterized by broad-spectrum susceptibility to pathogens and vulnerability to several opportunistic microorganisms that occur 3 to 12 months post birth. The exception to this paradigm is reticular dysgenesis (RD), in which the absence of neutrophils and T cells results in an earlier disease onset (ie, approximately 2 weeks after birth) characterized by septic episodes. The severity of the clinical manifestations implies that SCIDs are medical emergencies. Untreated patients typically do not live beyond the age of 6 to 12 months. The lack of T cells and the correspondingly severe outcome explains why SCID was the first condition to be successfully treated by allogeneic HSCT more than 40 years ago. Independent of the cell type used (ie, allogeneic HLA genocompatible or partially incompatible HSCT), dissection of the problems raised by HSCT in these settings has contributed to its development in the treatment of other genetic and acquired disorders of hematopoiesis.
Six Main Causes of Severe Combined Immunodeficiency and Consequences for Hematopoietic Stem Cell Transplantation
- 1.
Defective survival differentiation of T-cell and neutrophil precursor (RD).
- 2.
Premature cell death caused by the accumulation of purine metabolites (adenosine deaminase [ADA] deficiency).
- 3.
Defective cytokine-dependent survival signaling in T-cell precursors (and sometimes in NK-cell precursors). This mechanism accounts for more than approximately 50% of SCID cases. Impairment in the expression or function of the γ common (γc) cytokine receptor subunit (shared by the receptors for interleukin [IL]-2, IL-4, IL-7, IL-9, IL-15, and IL-21) causes the X-linked form of SCID, which is characterized by the complete absence of T and NK lymphocytes. Deficiency in JAK3 (which normally binds to the cytoplasmic region of γc) results in an identical phenotype that displays AR inheritance. Deficiency in IL-7Rα results in T-cell deficiency but normal NK cell development and is inherited in an AR fashion.
- 4.
Defective V(D)J rearrangements of the T-cell receptor (TCR) and B-cell receptor genes. In the Paris experience, this group accounts for 30% of SCID cases, contrasting with the Duke University series of 174 children, where it was identified in less than 5%. Deficiency in RAG1 or RAG2 (the lymphoid-specific recombination initiating elements) or Artemis (a factor involved in the nonhomologous end-joining repair pathway) leads to defective V(D)J rearrangements and thereby thymocyte and pre–B-cell death.
- 5.
Defective pre-TCR and TCR signaling. Pure T-cell deficiencies are caused by defects in a CD3 subunit (such as CD3-δ, -ϵ, or -ζ) or the CD45 tyrosine phosphatase—both of which are key proteins involved in pre-TCR or TCR signaling at the positive selection stage.
- 6.
Defective egress of thymocytes from the thymus.
The relative frequencies of the different genetic types of SCID are different in Europe from those observed in the United States. In particular, frequencies of V(D)J rearrangement defects are approximately 30% in Europe and only 4.2 in the United States. Furthermore, ADA and cytokine-dependent survival signaling defects account for 80% of all SCID forms in the United States and 60% in Europe. This difference in frequency has to be kept in mind when trying to compare HSCT clinical results in these two continents, taking into account the influence of the SCID form on the long-term outcome of these patients.
Some researchers include other T-cell immunodeficiencies in the SCID group, such as ZAP70 deficiency, CD3-γ deficiencies, HLA class II expression deficiency, purine nucleoside phosphorylase (PNP) deficiency, ligase IV and Cernunnos deficiencies, and Omenn syndrome (OS). Because these conditions are characterized by the presence of residual mature (although functionally defective) T cells, they are discussed later.
Allogeneic Hematopoietic Stem Cell Transplantation
The easiest situation: hematopoietic stem cell transplantation from an HLA identical sibling
Since the pioneering experiment in 1968, hundreds of SCID patients worldwide have undergone HSCT. At present, HSCT from an HLA MSD confers a 90% chance of a cure. For patients without pre-existing infections (such as neonates with SCID), the outcome is even better. This setting does not require myeloablative therapy to enable donor cell engraftment and is associated with an extremely infrequent occurrence of GVHD, for reasons that are not clear. The second favorable aspect of HLA-identical HSCT is the fast rate of T-cell recovery because of the homeostatic- and also antigen-driven expansion of the mature/memory donor T cells present in the graft. In HLA-identical HSCT, T cells can be observed as early as 10 to 15 days after transplantation and reach normal levels within 1 to 2 months. Later on (usually 3 to 4 months after transplantation), the peripheral detection of naïve T cells indicates that neothymopoiesis is also taking place. The rapid restoration of the T cell compartment and the quasi-absence of GVHD explain the excellent prognosis for HLA-identical HSCT.
Matched unrelated donors
This source of hematopoietic stem cells generally can provide results close to a HLA genoidentical-related HSCT, rendering this graft source theoretically preferable for SCID patients. Grunebaum and colleagues have recently reported good clinical results in SCID patients (80.5% survival) treated with MUD grafts. The overall European experience is slightly different. In the 2000–2005 period, the survival rates for MUD and related HLA mismatched treated patients were 69% and 66%, respectively. The major obstacle to the routine use of this HSC source is the median time that can be required from diagnosis to MUD HSCT, which is 4 to 6 months. This delay can be detrimental to infected patients as it increases the risk of clinical deterioration in patients with poorly treatable infection (for instance, severe respiratory viruses). Thus, in order to more accurately evaluate the role of MUD source for SCID patients, a comparison should be made assessing patients in similar clinical conditions as “intention to treat.” This type of comparison is not available today. Likely, the use of a MUD requiring several weeks or months for transplantation to be organized should be restricted to SCID patients in good clinical condition or for atypical SCID patients over the age of 2 with residual T-cell immunity.
The most difficult situation: hematopoietic stem cell transplantation from an HLA partially matched familial donor or umbilical cord blood
Haploidentical HSCT was introduced in this setting in 1982 to 1983, when techniques were developed to efficiently deplete contaminating donor mature T cells in the graft. Substantial numbers of SCID patients have received transplants across the major HLA barrier, and it seems that this type of donor provides a more favorable outcome in SCID than results generally reported for other clinical settings treated with haploidentical transplantation. Taking into account the obstacles raised by haploidentical HSCT, the question can be raised of the value of cord blood–derived HSC in this particular setting. The preliminary results of an international survey comparing the outcome of haploidentical versus cord blood HSCT in primary immunodeficiencies seem to indicate that event-free survival is essentially the same in both groups. This emphasizes the importance of the transplantation center’s expertise (rather than the donor source) in the HSCT outcome.
More than 40 years’ experience has provided essential information that may enable building up a la carte treatments for the four main groups of SCID patients. The defective cytokine-dependent survival signaling in T cell precursors (and sometimes NK-cell precursors) represents the most favorable situation—even for haploidentical HSCT. The combination of a lack of NK cells (in γc and JAK3 deficiencies) and a very early block in T-cell precursors completely abolishes graft rejection, resulting in high engraftment rates in the absence of any conditioning. The immediate consequence is greater than 70% survival rate in these patients, as long as severe, pre-existing, OIs are not present, a result similar to that reported by Buckley for X-linked SCID transplanted patients. In the absence of a myeloablative regimen, characteristic split chimerism is observed in this condition, characterized by the detection of donor T cells and host B cells in most cases. Despite the use of immunoglobulin substitution therapy, the absence of B-cell engraftment is responsible for recurrent respiratory tract infections more than 10 years after HSCT in some of the patients. In addition to these infectious episodes, the long-term prognosis of such patients is hampered by the occurrence in up to 30% of severe human papilloma virus infections, probably related to the nonhematopoietic consequences of γc and JAK3 deficiencies.
The poorest prognosis is observed in ADA -deficient patients transplanted with an HLA partially compatible donor. The toxic accumulation of ADA metabolites in organs, such as lung and thymus, and the high toxicity of the required conditioning regimen to achieve engraftment are responsible for this poorer outcome. As a result, stem cell transplantation using a haploidentical donor is not recommended for this particular SCID form, except for neonates.
Patients affected by defective V(D)J rearrangements of the TCR and B-cell receptor genes and transplanted with an HLA partially incompatible donor exhibit an intermediate outcome. The presence of functional NK cells potentially able to reject the graft and the occupation of the stromal niches by double-negative T-precursors in Rag 1 or Rag 2 or Artemis deficiencies diminish the likelihood of adequate engraftment in the absence of an appropriate conditioning regimen.
Impact of immune reconstitution and graft-versus-host disease on outcome
SCID patients with faster and better immune reconstitution have more successful outcomes. In a survey of 90 patients transplanted in Paris at Hôpital Necker, it has been clearly demonstrated that the 1- and 2-year post-HSCT event-free survival rates of patients with low CD4+ T-cell counts were significantly lower than for patients with normal CD4+ T-cell counts. The occurrence of acute GVHD (aGVHD) and chronic GVHD (cGVHD) is strongly associated with poor immune reconstitution, so GVHD prevention is mandatory in this setting. The results of this long-term outcome of a single-center cohort is difficult to compare with the survey reported by the team at Duke University on the long-term immune reconstitution in 128 transplanted SCID patients. There are significant differences between these two large cohorts, largely due to the different prevalence of genetic subsets. In Sarzotti-Kelsoe and colleagues’ survey, 86 of 128 patients have a defect in the cytokine signaling pathway and only seven presented with SCID due to a V(D)J rearrangement defect and only one with Artemis deficiency. Nevertheless, the Sarzotti-Kelsoe and colleagues’ analysis clearly indicates that the ADA and Rag-deficient patients are not as healthy as those with the other molecular types of SCID in line with their lower T-cell numbers, function, and thymic output after transplantation. In the survey from Paris, the same number of cytokine-dependent signaling defect and V(D)J recombination defects have been analyzed (n = 38), thus sufficiently powering the analysis to conclude that the poorest prognosis is for patients with an Artemis deficiency SCID form. Artemis is a protein ubiquitously expressed and is a key element of the nonhomologous end-joining process involved in double-stand DNA break repair. Thus, in the group of patients with this defective DNA repair pathway, there was a higher incidence of long-term complications, such as persistent aGVHD, auto-immunity/inflammation, requirement for some form of long-term nutritional support, opportunistic infection, and growth failure. These results indicate that further efforts should be made to design more tailored regimens in the future for these two different SCID forms.
Poor or slow immune reconstitution can result from thymic damage. In these patients, the latter is the sum of (1) the profound abnormalities in the distribution and function of cortical and medullar thymic epithelial cells (TECs) and thymic dendritic cells caused by genetic defects and thymic damage due to infection prior to transplant, (2) GVHD-induced thymic dysplasia, and (3) possibly toxicity of conditioning regimen. These conditions can profoundly affect the mechanisms of central tolerance and likely contribute to the autoimmune/inflammatory complications observed in mismatched related donor (MMRD) HSCT transplant recipients. The specific contribution of each of these mechanisms to the long-lasting immune reconstitution is difficult to determine. Clave and colleagues recently reported that in patients with grade I aGVHD, the median naïve positive TCR excision circles T-cell count was closer to that of patients with grades II–IV GVHD than that of patients with no detectable aGVHD. Subclinical, GVHD-mediated thymic lesions may thus play an important role in persistent immunodeficiency. Hence, any factor or cell therapy approach able to improve post-HSCT immune reconstitution is highly warranted. For example, it may be worth investigating the use of an anti–interferon (IFN)-γ antibody to prevent the thymic damage due to alloreactivity. Fibroblast growth factor 7 (also known as keratinocyte growth factor) is a potent epithelial cell mitogen and is able to protect against radiotherapy- and chemotherapy-induced damage and could improve TEC recovery after HSCT. Another strategy for speeding up immune reconstitution in this setting involves providing recipients with mature T cells devoid of specific antihost alloreactivity or pathogen-specific mature donor T cells. Other potentially feasible approaches include the use of Notch ligands to preferentially expand lymphoid progenitors. Thus, understanding of the precise mechanisms underlying de novo thymus-dependent generation of T cells serves as a basis for new strategies aimed at improving T-cell reconstitution; anticytokine antibodies, growth factors, and Notch-based culture systems are currently under investigation.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
