Bone Marrow Transplantation (2008) 41, 119–126 & 2008 Nature Publishing Group All rights reserved 0268-3369/08 $30.00 www.nature.com/bmt

REVIEW transplantation for primary immunodeficiency disease

CC Dvorak and MJ Cowan

Department of Pediatrics, Blood and Marrow Transplant Division, UCSF Children’s Hospital, San Francisco, CA, USA

Hematopoietic stem cell transplantation is the definitive SCID therapy for a variety of rare primary cellular immuno- deficiency syndromes diagnosed in children. All primary SCID is a rare disease caused by a group of genetic immunodeficiencies benefit from early diagnosis and disorders with a shared phenotype of deficient T- and transplantation before the development of serious infec- B-lymphocyte function (with or without abnormal natural tions, which contribute to a significant increased risk of killer (NK) cell development) that leads to early death from mortality following transplant. In the absence of a recurrent infections in affected children (Table 2). In matched sibling, parental haplocompatible, matched un- addition, since SCID patients are unable to reject foreign related donor and cord blood stem cells have all been cells, a significant percentage of patients will have evidence utilized with varying degrees of success and immune of maternally engrafted T lymphocytes, which often leads reconstitution. The role of pretransplant conditioning in to clinical manifestations of GVHD.1 Exceptfor those patients with SCID disease in terms of its effects upon patients with SCID due to deficiency of adenosine T- and B-cell immune reconstitution and late effects is deaminase (ADA), for which a replacementenzyme exists, still under debate and will require further study. the only curative therapy for SCID is allogeneic HSCT. Bone Marrow Transplantation (2008) 41, 119–126; However, early results with gene insertion into autologous doi:10.1038/sj.bmt.1705890; published online 29 October 2007 hematopoietic stem cells for children with X-linked SCID Keywords: primary immunodeficiency; SCID; transplant- and ADA deficiency2 suggest that eventually, this will ation; HSCT become a more common form of curative treatment for many primary immunodeficiency diseases. At the time of diagnosis, patients with SCID generally have already developed, or are at an extremely elevated risk of developing, a life-threatening infection. This necessitates Introduction rapid initiation of HSCT. For all stem cell sources, successful outcomes are more likely to be achieved when Primary cellular immunodeficiencies are a group of the patient is still very young, preferably less than 6 months inherited disorders characterized by severe impairment of age.3,4 Buckley et al.,5 demonstrated that infants of the innate or adaptive immune systems, which transplanted less than 3.5 months of age had a 95% overall generally leads to early death from infectious complica- survival (OS), compared to only 76% OS in older children. tions. These disorders can be further categorized by the cell This is likely due to the development of pulmonary lineage(s) primarily affected (Table 1). While improved infections prior to transplant, which have been associated supportive care has extended the life span of patients with significantly poorer outcomes.4,6 In addition, trans- affected by these diseases, definitive cure is generally only plants performed within the first month of life are achieved by allogeneic hematopoietic stem cell transplanta- associated with more rapid T-cell reconstitution, perhaps tion (HSCT), though recent advances in hold due to superior thymic capacity.7 significant promise that this may soon be a viable alternative. Impact of SCID phenotype

T-B þ NKÀ SCID The mostcommon form of SCID, accountingfor approximately 50% of all cases, is due to a defect in the Correspondence: Dr MJ Cowan, Department of Pediatrics, Blood and gene for the common gamma chain (gc) located on Xq13. Marrow TransplantDivision, UCSF Children’s Hospital,505 Parnassus Another defect, JAK3 deficiency, results in a similar Avenue, San Francisco, CA 94143-1278, USA. E-mail: [email protected] T-B þ NKÀ phenotype (Table 2). Post transplant, patients Received 24 August 2007; revised 14 September 2007; accepted 14 with gc deficiency have been noted to have excellent rates of September 2007; published online 29 October 2007 sustained thymic output, as measured by levels of T-cell BMT for immunodeficiency diseases CC Dvorak and MJ Cowan 120 Table 1 Primary immunodeficiencies potentially treated with HSCT

Absent T- and B- Defective T and B lymphocytes Dysfunctional T lymphocytes with Absent or dysfunctional granulocytes lymphocyte function predisposition to HLH

SCID Wiskott–Aldrich syndrome Familial HLH (defects in perforin, Severe congenital neutropenia MUNC, etc.) HIGM1 Chediak–Higashi syndrome Leukocyte adhesion disorder Griscelli syndrome Chronic granulomatous disease XLP

Abbreviations: HIGM1 ¼ hyper IgM syndrome (CD40 ligand deficiency); HLH ¼ hemophagocytic lymphohistiocytosis; XLP ¼ X-linked lymphoprolifera- tive disease.

Table 2 Genetic subtypes of severe combined immunodeficiency

Name Defect Phenotype Special

X-linked Common g chain T-B+NKÀ JAK3 deficiency Janus kinase 3 T-B+NKÀ Rag 1 or 2 Recombinase-activating T-BÀNK+ Frequently associated with Omenn’s proteins 1 or 2 syndrome: autoreactive GVHD Artemis deficiency Artemis (also known as T-BÀNK+ Athabascan-speaking Native Americans, DCLRE1C) radiosensitive Ligase 4 deficiency Ligase 4 T-BÀNK+ Radiosensitive IL-7Ra deficiency IL-7 receptora T-B+NK+ CD45 deficiency CD45 T-B+NK+ CD3d deficiency CD3d subunitT-B+NK+ CD3e deficiency CD3e subunitT-B+NK+ CD3B deficiency CD3B subunitT-B+NK+ Cartilage hair hypoplasia Endoribonuclease T-B+NK+ Dwarfism, hypoplastic hair Finnish, Amish p56lck deficiency p56lck Protein tyrosine kinase T-B+NK+ ADA deficiency Adenosine deaminase T-BÀNKÀ PNP deficiency Purine nucleoside T-BÀNKÀ Neurologic dysfunction, ataxia phosphorylase Reticular dysgenesis Unknown T-BÀNKÀ Impaired myeloid and erythroid development, sensorineural deafness ZAP70 deficiency B-chain-associated protein kinase CD4+, CD8À B+, NK+ type II HLA class II CD4À(mild), CD8+ North African B+, NK+ SCID with bowel atresia Unknown CD4+, CD8+, B+NK+

Abbreviations: ADA ¼ adenosine deaminase; DCLREIC ¼ DNA cross-link repair enzyme 1C; HLA ¼ human leukocyte antigen.

receptor excision circles (TRECs). This may be due to a has been attributed to a diminished rate of engraftment, pre-transplant low level of thymic precursors in these increased severity of GVHD, higher incidence of chronic patients, which post transplant allows for thymic seeding of GVHD4 and slower recovery of T-cell function.6,9 The very early progenitors.8 relatively poor engraftment in patients with T-BÀNK þ SCID is in large part due to the presence of host NK cells, which are capable of mediating donor stem cell rejection.4 T-BÀNK þ SCID Studies in animal models of NK þ SCID,10 as well as It has been estimated that 20–30% of all SCID cases have unpublished data in our own laboratory with an Artemis- the T-BÀNK þ phenotype with defects in RAG1 or RAG2 deficientmouse model, supporttherole of NK cells in graft being the most common etiology. It has been reported that resistance. B-phenotypes (presumably NK þ ) of SCID have signifi- cantly poorer 3-year OS (36%) compared with B þ phenotypes (64%).6 Bertrand et al.4 demonstrated a worse SCID due to DNA repair defects disease-free survival after T-depleted haplocompatible A subsetof T-B ÀNK þ SCID patients have increased transplant in patients with BÀ SCID (35%) compared cellular sensitivity to alkylating agents and ionizing with B þ SCID (60%). Haddad et al.9 has also shown a radiation, mainly due to a deficiency in the gene for worse long-term outcome in children with BÀ SCID (NK Artemis, a critical protein in the nonhomologous DNA cell phenotype unknown), who were more likely to die repair pathway.11 Unlike other forms of BÀ SCID, patients during the first 6 months post transplant (37%) compared with Artemis deficiency have undergone HSCT with with those who had B þ SCID (13%). This poor survival an excellentOS of 75%. 11 A new type of radiosensitive

Bone Marrow Transplantation BMT for immunodeficiency diseases CC Dvorak and MJ Cowan 121 T-BÀNK þ SCID has been recently described with a defect cells from this source makes this option unfeasible for most in DNA ligase IV, an enzyme in the nonhomologous DNA cases of SCID, in which HSCT is urgently indicated. repair pathway distal to Artemis.12 Haplocompatible-related donors T-B þ NK þ SCID With the advent of effective T-cell depletion strategies, the Another subset of SCID patients is characterized by the use of haplocompatible family members as donors for presence of both B cells and NK cells. A variety of cellular children with SCID has become a viable strategy. Three- defects have been reported to result in this phenotype, year survivals of 53–79% have been reported, with including deficiencies of IL-7 receptor-a, CD45 and the significantly better success rates in more recent years and CD3d, CD3e and CD3B subunits. Most of these deficiencies in more experienced centers,3,5,6,14 especially if the trans- have only recently been discovered and very small numbers plantis performed early before thedevelopmentof life- of these patients have been followed long term after HSCT, threatening infections. In SCID patients with evidence of making definitive statements regarding their clinical course maternal engraftment,1 the mother is typically utilized as difficult. the stem cell source (unless an HLA-matched sibling is available), since the recipient is already tolerant to the maternal cells. A potential risk with this approach is the T þ SCID Rare forms of SCID are characterized by the selective development of GVHD by the maternally engrafted T cells following the infusion of maternal bone marrow cells. In absence of CD4 þ T cells (for example, HLA class II our own experience this has not been a significant problem deficiency) or CD8 þ T cells (for example, ZAP70 deficiency). Due to partial T-cell function, children with provided the donor graft is adequately T-cell depleted. A these diseases are often diagnosed later than other patients significant problem with haplocompatible transplants is with SCID. A high incidence of preexisting viral infections that the required T-cell depletion can lead to a slow rate has been associated with high rates of severe GVHD and of T-cell reconstitution and a prolonged time of risk of poor outcomes after HLA-identical HSCT in patients with opportunistic infections and EBV-related lymphoproli- HLA class II deficiency.6,13 ferative disease. The median time to achieve normal T-cell numbers is 8.5–10 months,3,19 only slightly slower than that seen following HSCT from an HLA-identical sibling (6 months).19 However, functional T cells generally take Stem cell sources 4–9 months to be detected following haplocompatible HSCT, as opposed to 2 months following HLA-identical The preferred choice of stem cell donor for a patient with sibling HSCT.3,9,19 B-cell numbers generally recover by SCID is an HLA-identical sibling, in which the OS now 3–12 months post transplant, while their function can take 5,14 exceeds 90% if the transplant is performed promptly. In 6–15 months to return to normal.3,9 In addition, the slow patients without a matched sibling, the choice is whether to recovery of regulatory T cells may explain the increased use an immediately available -depleted haplocompa- risk for the development of autoimmune hemolytic tible family member or to perform a search for an HLA- anemia.19 matched unrelated donor or cord blood unit. Some of the reports on transplantation with different stem cell sources are summarized in Table 3. In cases with a known genetic Matched unrelated donors mutation, recent improvements in preimplantation genetics Due to a slow rate of immune reconstitution following raise the possibility of creation of a SCID-free HLA- T cell-depleted haplocompatible HSCT as well as the matched sibling for possible cord blood transplantation. inability to do T-cell depletion, some centers have However, both the cost and time needed to produce stem advocated utilization of matched unrelated donors. Despite

Table 3 Survival following HSCT for SCID based on stem cell source and conditioning regimen

Year Reference MRD Haplo Haplo MUD MUD Cord

Conditioning None None MA MA RI MA Dror et al. 1993 3 — 67% (12) 50% (12) — — — Buckley et al. 1999 5 100% (12) 78% (77) — — — 66% (3) Bertrand et al. 1999 4 — 46% (50) 54% (129) — — — Dalal et al. 2000 15 — — — 67% (9) — — Knutsen and Wall 2000 16 —————88%(8) Antoine et al. 2003 6 81% (104) — — 63% (28) — — Rao et al. 2005 17 — — — 71% (7) 83% (6) — Bhattacharya et al. 2005 18 —————80%(10)a Grunebaum et al. 2006 14 92% (13) — 53% (40) 81% (41) — —

Stem cell source: MRD (matched related donor) vs haplo (haplocompatible family donor) vs MUD (matched unrelated donor) vs Cord (unrelated cord blood stem cells). Conditioning: none vs MA (myeloablative) vs RI (reduced intensity). Percentage indicates overall survival (absolute number of patients). aSome patients received no conditioning.

Bone Marrow Transplantation BMT for immunodeficiency diseases CC Dvorak and MJ Cowan 122 a later median age of transplant due to the time needed to sustained thymopoiesis, except for patients with ILgc perform a search, some groups reportgood outcomes(OS deficiency.8 As with ILgc deficiency, in patients with 63–67%) with the use of unrelated donors.6,15 Grunebaum JAK3 deficiency, HSCT without conditioning successfully et al.14 reported that matched unrelated donors had reconstitutes T-cell immunity, but rarely accomplishes superior 2-year OS (81%) compared with mismatched restoration of NK cell function, and does not produce related donors (53%). In those patients tested, evidence of donor B cells.20 Successful achievementof donor B-cell functional T cells was seen as early as 2 months post- chimerism is usually associated with normal B-cell func- HSCT, and specific production by 5 months.15 tion, while recipient B cells only receive enough support It should be noted that in the recipients of the unrelated from donor T cells for normal function in a fraction of donor transplants, conditioning with high-dose chemother- cases, and none with ILgc deficiency.9,22,24 In cases apy such as busulfan was utilized. Since it has been shown of inadequate B-cell reconstitution, prolonged administra- that the use of a conditioning regimen for haplocompatible tion of intravenous immunoglobulin is required to prevent donors produces faster recovery of T-cell numbers,19 itis infections.5 difficult to directly compare the immune reconstitution Possible explanations for selective T-cell donor chimer- following HSCT from unrelated donors with that of ism following unconditioned HSCT include: (1) engraft- nonconditioned haplocompatible donors. In addition, the ment of marrow stem cells does occur, but only selective long-term complications with the utilization of high-dose T-lymphocyte lineage differentiation takes place; (2) early chemotherapy-based conditioning regimens have not been self-renewing progenitor cells seed the thymus8 or (3) reported in any detail. Thus, until a prospective rando- mature donor T cells expand to fill a homeostatic niche. In mized trial comparing haplocompatible-related donors to the latter case, it might be expected that eventually thymic matched unrelated donors has been performed, the optimal output would diminish and holes would appear in the T-cell choice of stem cell source is not clear. However, some receptor repertoire, leading to recurrence of opportunistic recent success with reduced intensity conditioning regimens infections later in life.25 Some groups do notreporta demonstrates that this may prove to be a feasible significant difference in late infectious complications alternative to fully myeloablative conditioning prior to regardless of thymic output,8 while others have reported transplant of unrelated donor stem cells, especially in that some long-term survivors of HSCT for SCID do show patients who begin transplant with a known infection or a propensity for the development of chronic human organ dysfunction.17 papillomavirus disease, though a link to the presence or absence of a conditioning regimen has not been estab- lished.26 At least two studies have identified donor stem Cord blood cells in the marrow of recipients of unconditioned Utilization of cord blood for HSCT for SCID has been haplocompatible transplants.23,27 In one, a sibling was advocated over adult-unrelated donors due to more rapid diagnosed with SCID and was HLA matched with her availability of the product, increased permissiveness of sister who had previously undergone a successful haplo- HLA mismatch with a reduced risk of GVHD. Most compatible transplant from their father. The sibling reports describing this strategy include only small numbers received a bone marrow transplant from the older sister of patients, but demonstrate rapid recovery of T-, B- and and successfully engrafted without any conditioning being 5,16,20,21 NK cell function. However, these types of trans- given. Donor cells were shown to contain stem cells from plants required high-dose chemotherapy to achieve success- the father at the time of the second transplant.23 ful engraftment, further raising the issue of conditioning It has been argued that SCID phenotypes with preserva- therapy in very young infants. In one report of 10 cord tion of NK cells, which have been shown to mediate graft blood transplants for children with SCID, 3 transplants rejection in murine models, would benefit from condition- did notuse conditioningand were from 6/6 or 8/8 ing regimens that are designed to inhibit NK activity.4 HLA-matched unrelated cord blood units, all of whom Grunebaum et al.14 was not able to detect a difference in 18 engrafted. The same could prove to be true for perfectly outcome based on B-cell phenotypes, possibly due to a high matched unrelated donor transplants for SCID as well and use of unrelated donors, for whom busulfan and cyclophos- needs to be further evaluated. phamide was utilized. However, some forms of B-NK þ SCID have increased sensitivity to alkylating agents and ionizing radiation, and are at significant risk of increased Conditioning short- and long-term toxicity from conditioning regimens utilizing these agents.4,11 Alternative approaches need to be One unique feature of HSCT for the treatment of SCID is considered for these genotypes. the lack of an absolute requirement for conditioning, which When conditioning is utilized, complete donor chimerism has been demonstrated to restore antigen-specific T-cell in all cell lines is frequently accomplished.3,8,22 Patients with function within 8 weeks (for unfractionated marrow reticular dysgenesis present with the typical features of recipients)19 to 4–8 months (for T cell-depleted marrow SCID, such as absent lymphocytes and , but also recipients).3,5,9 It is unclear, however, whether a true stem demonstrate an early arrest of myeloid differentiation. cell graft occurs in the recipient marrow when ablative Given the more global nature of the defect, it is not conditioning is not used, as B cells, myeloid cells and surprising that HSCT without conditioning results in erythroid cells generally remain of recipient origin.3,5,8,22,23 treatment failure, while myeloablative conditioning can be Host myeloid chimerism has been associated with defective successful.28

Bone Marrow Transplantation BMT for immunodeficiency diseases CC Dvorak and MJ Cowan 123 For those patients requiring conditioning, the ideal Other primary immunodeficiencies regimen has notbeen defined. Many groups utilizea conventional 16 mg/kg of busulfan plus 200 mg/kg of In addition to SCID, many other primary immunodefi- cyclophosphamide.3,14,15 Ithas been proposed thatfully ciencies have been treated with allogeneic HSCT. These ablative regimens utilizing 16 mg/kg of busulfan may have include syndromes with T-cell defects (for example, poor outcomes due to increased infectious complications. Wiskott–Aldrich syndrome (WAS) and hyper IgM1 syn- Bertrand et al.4 reported that their best outcomes follow a drome), inherited predispositions to development of reduced intensity regimen utilizing 8 mg/kg of busulfan plus hemophagocytic lymphohistiocytosis (HLH) (for example, 200 mg/kg of cyclophosphamide, however other groups familial HLH, Chediak–Higashi syndrome (CHS), Griscelli have reported that 8 mg/kg of busulfan is not sufficient to syndrome and X-linked lymphoproliferative disease) and obtain myeloid chimerism.24,28 Table 3 summarizes the phagocytic disorders (for example, severe congenital survival rates in recent reports that indicate the type of neutropenia (SCN), leukocyte adhesion deficiency and conditioning regimen administered based on the stem cell chronic granulomatous disease (CGD)). Because of the source utilized. rarity of most of these disorders, many of the descriptions of treatment with HSCT include very small numbers of patients. When these patients are combined into a single large group for analytical purposes, not surprisingly Long-term outcomes patients with genotypically identical-related donors had significantly improved 3-year OS (71%) compared to other Failure to develop adequate T-cell immunity at 2 years donor types (42–59%).6 after HSCT is the most common indication for retrans- plantation, though B- SCID may require retransplant at an earlier time point.9 Repeattransplanthasbeen required Wiskott–Aldrich syndrome more often following haplocompatible HSCT.14 Many Wiskott–Aldrich syndrome is characterized by the triad of centers have utilized a conditioning regimen during second thrombocytopenia with small platelets, eczema and recur- transplant, especially when changing from a haplocompa- rent infections. The T-cell immunodeficiency predisposes to tible donor to an unrelated donor.3,5,20,22 Others perform the development of autoimmune phenomena and lympho- ‘booster’ transplants without conditioning with reported mas. A defect at Xp11 in the gene for the WAS protein improvementin immune function. 5,23 leads to defective cytoskeletal interactions with CDC42, a While HSCT may correctsome or all of theimmunologic regulator of signaling pathways that controls diverse abnormalities in patients with SCID, it should be noted cellular functions, including migration and cell cycle that nonimmunologic abnormalities, such as dwarfism and progression. Affected males rarely survive past the second hypoplastic hair in cartilage-hair hypoplasia or neurologic decade of life. abnormalities in ADA deficiency,22 may persist. Radio- The only curative strategy for treatment of WAS is sensitive forms of SCID may be at increased risk of other allogeneic HSCT. A report from the International Bone abnormalities, such as failure of secondary tooth develop- Marrow Transplant Registry and National Marrow Donor ment.11 Otherwise, reports generally show few significant Program demonstrated in 170 patients that while the 5-year long-term sequelae of HSCT in infants with SCID8,14 OS of patients transplanted from HLA-identical siblings although detailed careful long-term follow-up studies was 87%, the results for unrelated HSCT were significantly evaluating growth and developmental and other major related to the age at transplant.31 Unrelated donors less organ functions have not been done in the large majority of than 5 years of age had an 85% 5-year OS, while all 15 patients with SCID. patients greater than 5 years of age died.31 This may be at least partly related to the presence of infection prior to conditioning, which significantly decreased the survival of Newborn screening patients with unrelated donors.31 No difference was noted between TBI- and chemotherapy-based conditioning regi- Recent work has led to the development of two tests for mens.31 Haplocompatible-related transplants have been less T-cell lymphopenia: IL-7 levels29 and a polymerase chain successful for WAS, with an OS of 45–52%.6,31 reaction test to detect low numbers of TRECs,30 which can be performed on dried blood spots as part of a population- based screening program atbirth.Since HSCT is most Familial hemophagocytic lymphohistiocytosis likely to be successful when performed at a young age prior Familial HLH is characterized by episodes of fever, to the onset of life-threatening infections,7 plans are hepatosplenomegaly and cytopenias. An autosomal reces- underway in a number of states to test whether implemen- sive defectin one of several genes, including thoseencoding tation of such a program will prove to be a cost-effective perforin or Munc13, causes reduced NK and T-cell strategy to improve OS in these infants. However, large- cytotoxicity. This leads to a widespread accumulation of scale screening programs mustbe designed witha high lymphocytes and mature macrophages with hypercyto- degree of specificity in order to avoid the generation of a kinemia. Familial HLH is invariably fatal. large number of false positive results that would result in The only curative strategy for treatment of familial excessive evaluations for the presence of SCID and undue HLH is allogeneic HSCT. A report from a multicenter parental anxiety. prospective trial, HLH-94, demonstrated a 62% 3-year

Bone Marrow Transplantation BMT for immunodeficiency diseases CC Dvorak and MJ Cowan 124 EFS in 65 children undergoing allogeneic HSCT with a Blood and Marrow Transplantation demonstrated in 23 variety of stem cell sources.32 patients that myeloablative conditioning prior to matched sibling HSCT can be safely performed (85% OS), especially Chediak–Higashi syndrome if the patients were free of infection at the time of HSCT 36 Chediak–Higashi syndrome is characterized by oculocuta- (100% OS). neous albinism, recurrent infections and the presence of giant granules in hematopoietic and other cells. A recessive defect at 1q42 in the lysosomal trafficking regulator gene Conclusions leads to neutropenia, as well as defective phagocyte, NK cell and T-cell function. In those patients that survive the One overriding similarity between all primary immuno- infectious complications, during the first or second decade deficiencies is the importance of early transplantation of life, an accelerated phase develops, in which patients before the development of serious infections that contribute generally die from HLH. to a significant increase in the risk of mortality following A report from the Center for International Blood and HSCT. In families with a history of primary immunode- Marrow Transplant Research demonstrated a 62% 5-year ficiency due to an identified genetic mutation prenatal OS in 35 children undergoing allogeneic HSCT for CHS. diagnosis is available, which should enable initiation of Mortality was highest in those patients undergoing HSCT HSCT in the first several months of life. In addition, several with an alternative related donor or with active disease at attempts at in utero HSCT have been performed for the time of HSCT, suggesting that early referral for HSCT treatment of SCID, though progress with this technique prior to development of the accelerated phase may to date has been limited due to poor engraftment and mixed 37 significantly improve survival.33 chimerism. A population-based screening program at birth may be able to identify children with T-cell lymphopenia and SCID prior to the onset of life-threaten- Severe congenital neutropenia ing infections and should also allow for prompt initiation Severe congenital neutropenia is characterized by an 7,29,30 9 of HSCT, when itis mostlikely tobe successful. In the absolute neutrophil count less than 0.2 Â 10 per liter and future, the success rates for T cell-depleted haplocompa- an early onset of life-threatening bacterial infections. Both tible-related stem cells, as well as unrelated donor sources, autosomal dominant and recessive forms of SCN exist, for treatment of SCID and other primary immunodeficien- with the most common cause due to a defect in the cies will continue to improve as new strategies are neutrophil elastase gene at 19p13. The administration of G- discovered to enhance immune reconstitution, such as CSF has significantly improved the survival of patients with adoptive immunotherapy utilizing alloreactive-depleted SCN, however, these children also demonstrate an elevated donor T cells,38 and to reduce short- and long-term side risk of malignant transformation into AML. A recent study effects by the effective use of reduced toxicity regimens and has demonstrated that approximately one-third of SCN possibly regimens that do not require alkylating agents or patients are relatively unresponsive to G-CSF doses above radiation. Finally, as efforts with gene therapy become 8 mg/kg per day and that these patients have a 10-year more effective and safer, we will be able to move away from incidence of AML of 40%, with a 14% incidence of death 34 the use of allogeneic donors in favor of replacement with from sepsis. gene-corrected autologous hematopoietic progenitors.2 A report from the SCN International Registry demon- strated an 82% OS in 11 children with nontransformed SCN undergoing allogeneic HSCT. The only two deaths References seen in this cohort were in recipients of mismatched grafts.35 Given the extremely elevated risk of AML and/ 1 Muller S, Ege M, Pottharst A, Schulz AS, Schwarz K, or mortality in G-CSF poorly-responsive SCN patients, Friedrich W. Transplacentally acquired maternal T lympho- early referral for HSCT may significantly improve survival. cytes in severe combined immunodeficiency: a study of 121 patients. Blood 2001; 98: 1847–1851. 2 Cavazzana-Calvo M, Fischer A. Gene therapy for severe Chronic granulomatous disease combined immunodeficiency: are we there yet? J Clin Invest Chronic granulomatous disease is characterized by recur- 2007; 117: 1456–1465. rent pyogenic infections in patients with normal neutrophil 3 Dror Y, Gallagher R, Wara DW, Colombe BW, Merino A, numbers. Patients present with deep tissue infections and Benkerrou M et al. Immune reconstitution in severe combined sepsis due to catalase-positive organisms such as Staphy- immunodeficiency disease after lectin-treated, T-cell-depleted lococcus aureus and Aspergillus fumigatus. A defectin one haplocompatible bone marrow transplantation. Blood 1993; of the four genes encoding subunits of the nicotinamide 81: 2021–2030. adenine dinucleotide phosphate-oxidase complex leads to 4 Bertrand Y, Landais P, Friedrich W, Gerritsen B, Morgan G, insufficient production of free protons from which to make Fasth A et al. Influence of severe combined immunodeficiency hydrogen peroxide. With good supportive care, including phenotype on the outcome of HLA non-identical, T-cell- depleted bone marrow transplantation: a retrospective therapy with interferon-g, affected individuals can live up European survey from the European group for bone marrow to the fourth decade of life, but suffer early mortality from transplantation and the European Society for Immuno- recurrentpulmonary infections. deficiency. J Pediatr 1999; 134: 740–748. The only curative strategy for treatment of CGD is 5 Buckley RH, Schiff SE, Schiff RI, MarkertL, Williams LW, allogeneic HSCT. A reportfrom theEuropean Group for Roberts JL et al. Hematopoietic stem-cell transplantation for

Bone Marrow Transplantation BMT for immunodeficiency diseases CC Dvorak and MJ Cowan 125 the treatment of severe combined immunodeficiency. N Engl logic, and molecular analyses of 10 patients and outcomes of JMed1999; 340: 508–516. stem cell transplantation. Blood 2004; 103: 2009–2018. 6 Antoine C, Mu¨ ller S, CantA, Cavazzana-Calvo M, Veys P, 21 Knutsen AP, Wall DA. Kinetics of T-cell development of Vossen J et al. Long-term survival and transplantation of umbilical cord blood transplantation in severe T-cell immu- haemopoietic stem cells for immunodeficiencies: report of the nodeficiency disorders. J Allergy Clin Immunol 1999; 103 European experience 1968–1999. Lancet 2003; 361: 553–560. (5 Part1): 823–832. 7 Myers LA, Patel DD, Puck JM, Buckley RH. Hematopoietic 22 Ho¨ nig M, AlbertMH, Schulz A, Sparber-Sauer M, Schu¨ tzC, stem cell transplantation for severe combined immuno- Belohradsky B et al. Patients with adenosine deaminase deficiency in the neonatal period leads to superior thymic deficiency surviving after hematopoietic stem cell transplant- output and improved survival. Blood 2002; 99: 872–878. ation are at high risk of CNS complications. Blood 2007; 109: 8 Cavazzana-Calvo M, Carlier F, Le DeistF, Morillon E, 3595–3602. Taupin P, Gautier D et al. Long-term T-cell reconstitution 23 Stiehm ER, Roberts RL, Hanley-Lopez J, Wakim ME, after hematopoietic stem-cell transplantation in primary Pallavicini MG, Cowan MJ et al. Bone marrow transplant- T-cell-immunodeficient patients is associated with myeloid ation in severe combined immunodeficiency from a sibling who chimerism and possibly the primary disease phenotype. Blood had received a paternal bone marrow transplant. N Engl J Med 2007; 109: 4575–4581. 1996; 335: 1811–1814. 9 Haddad E, Landais P, Friedrich W, Gerritsen B, Cavazzana- 24 Haddad E, Le DeistF, AucouturierP, Cavazzana-Calvo M, Calvo M, Morgan G et al. Long-term immune reconstitution Blanche S, De SaintBasile G et al. Long-term chimerism and outcome after HLA-nonidentical T-cell-depleted bone and B-cell function after bone marrow transplantation in marrow transplantation for severe combined immuno- patients with severe combined immunodeficiency with B deficiency: a European retrospective study of 116 patients. cells: a single-center study of 22 patients. Blood 1999; 94: Blood 1998; 91: 3646–3653. 2923–2930. 10 Murphy WJ, Kumar V, Bennett M. Rejection of bone marrow 25 Patel DD, Gooding ME, Parrott RE, Curtis KM, Haynes BF, allografts by mice with severe combined immune deficiency Buckley RH. Thymic function after hematopoietic stem-cell (SCID). Evidence that natural killer cells can mediate the transplantation for the treatment of severe combined immu- specificity of marrow graft rejection. J Exp Med 1987; 165: nodeficiency. N Engl J Med 2000; 342: 1325–1332. 1212–1217. 26 LaffortC, Le DeistF, Favre M, Caillat-ZucmanS, Radford- 11 O’Marcaigh AS, DeSantes K, Hu D, Pabst H, Horn B, Li L Weiss I, Debre´M et al. Severe cutaneous papillomavirus et al. Bone marrow transplantation for T-B- severe combined disease after haemopoietic stem-cell transplantation in patients immunodeficiency disease in Athabascan-speaking native with severe combined immune deficiency caused by common Americans. Bone Marrow Transplant 2001; 27: 703–709. gammac cytokine receptor subunit or JAK-3 deficiency. Lancet 12 Gruhn B, Seidel J, Zintl F, Varon R, To¨ nnies H, Neitzel H 2004; 363: 2051–2054. et al. Successful bone marrow transplantation in a patient with 27 Tjnnfjord GE, Steen R, Veiby OP, Friedrich W, Egeland T. DNA ligase IV deficiency and bone marrow failure. Orphanet Evidence for engraftment of donor-type multipotent CD34+ J Rare Dis 2007; 2:5. cells in a patient with selective T-lymphocyte reconstitution 13 Renella R, Picard C, Neven B, Ouachee-Chardin M, Casanova after bone marrow transplantation for B-SCID. Blood 1994; J-L, DeistFL et al. Human leucocyte antigen-identical 84: 3584–3589. haematopoietic stem cell transplantation in major histocom- 28 Bertrand Y, Mu¨ ller SM, Casanova JL, Morgan G, Fischer A, patiblity complex class II immunodeficiency: reduced survival Friedrich W. Reticular dysgenesis: HLA non-identical bone correlates with an increased incidence of acute graft-versus- marrow transplants in a series of 10 patients. Bone Marrow host disease and pre-existing viral infections. Br J Haematol Transplant 2002; 29: 759–762. 2006; 134: 510–516. 29 McGhee SA, Stiehm ER, Cowan M, Krogstad P, McCabe 14 Grunebaum E, Mazzolari E, Porta F, Dallera D, Atkinson A, ERB. Two-tiered universal newborn screening strategy for Reid B et al. Bone marrow transplantation for severe severe combined immunodeficiency. Mol Genet Metab 2005; combined immune deficiency. JAMA 2006; 295: 508–518. 86: 427–430. 15 Dalal I, Reid B, Doyle J, Freedman M, Calderwood S, 30 Chan K, Puck JM. Developmentof population-basednewborn Saunders F et al. Matched unrelated bone marrow transplant- screening for severe combined immunodeficiency. J Allergy ation for combined immunodeficiency. Bone Marrow Trans- Clin Immunol 2005; 115: 391–398. plant 2000; 25: 613–621. 31 Filipovich A, Stone JV, Tomany SC, Ireland M, Kollman C, 16 Knutsen A, Wall D. Umbilical cord blood transplantation in Pelz CJ et al. Impact of donor type on outcome of bone severe T-cell immunodeficiency disorders: two-year experience. marrow transplantation for Wiskott–Aldrich syndrome: colla- J Clin Immunol 2000; 20: 466–476. borative study of the International Bone Marrow Transplant 17 Rao K, Amrolia P, Jones A, Cale C, Naik P, King D et al. Registry and the National Marrow Donor Program. Blood Improved survival after unrelated donor bone marrow 2001; 97: 1598–1603. transplantation in children with primary immunodeficiency 32 Henter J, Samuelsson-Horne A, Arico`M, Egeler R, Elinder G, using a reduced-intensity conditioning regimen. Blood 2005; Filipovich A et al. Treatment of hemophagocytic lympho- 105: 879–885. histiocytosis with HLH-94 immunochemotherapy and bone 18 Bhattacharya A, Slatter M, Chapman C, Barge D, Jackson A, marrow transplantation. Blood 2002; 100: 2367–2373. Flood T et al. Single centre experience of umbilical cord stem 33 Eapen M, DeLaatC, Baker K, Cairo M, Cowan M, Kurtzberg cell transplantation for primary immunodeficiency. Bone J et al. Hematopoietic cell transplantation for Chediak– marrow Transplant 2005; 36: 295–299. Higashi syndrome. Bone Marrow Transplant 2007; 39: 19 Wijnaendts L, Le Deist F, Griscelli C, Fischer A. Development 411–415. of immunologic functions after bone marrow transplantation 34 Rosenberg PS, Alter BP, Bolyard AA, Bonilla MA, Boxer LA, in 33 patients with severe combined immunodeficiency. Blood Cham B et al. The incidence of leukemia and mortality 1989; 74: 2212–2219. from sepsis in patients with severe congenital neutro- 20 Roberts JL, Lengi A, Brown SM, Chen M, Zhou Y-J, O’Shea penia receiving long-term G-CSF therapy. Blood 2006; 107: JJ et al. Janus kinase 3 (JAK3) deficiency: clinical, immuno- 4628–4635.

Bone Marrow Transplantation BMT for immunodeficiency diseases CC Dvorak and MJ Cowan 126 35 Zeidler C, Welte K, Barak Y, Barriga F, Bolyard A, Boxer L the European experience, 1985–2000. Blood 2002; 100: et al. Stem cell transplantation in patients with severe 4344–4350. congenital neutropenia without evidence of leukemic trans- 37 Westgren M. In utero stem cell transplantation. Semin Reprod formation. Blood 2000; 95: 1195–1198. Med 2006; 24: 348–357. 36 Seger RA, Gungor T, Belohradsky BH, Blanche S, 38 Amrolia PJ, Muccioli-Casadei G, Huls H, Adams S, Durett A, Bordigoni P, Di Bartolomeo P et al. Treatment of chronic Gee A et al. Adoptive immunotherapy with allodepleted donor granulomatous disease with myeloablative conditioning T-cells improves immune reconstitution after haploidentical and an unmodified hemopoietic allograft: a survey of stem cell transplantation. Blood 2006; 108: 1797–1808.

Bone Marrow Transplantation