Cytoreductive conditioning for severe combined immunodeficiency – help or hindrance?

Abstract

Use of chemotherapy-based conditioning-facilitated engraftment in patients with severe combined immunodeficiency (SCID) is contentious. In T- and NK lymphocyte-negative, B-lymphocyte-positive (T-B+NK+) and T-B-NK+ SCID, the osteo-medullary space is occupied by recipient hematopoietic stem cells and mature B-lymphocytes. The thymic niche is empty in T-B+NK+ SCID but fully occupied by developmentally arrested T-lymphocyte precursors in T-B-NK+ SCID. The outcome of infusion of donor stem cells differs and is dependent on genetic defect and the lymphocyte developmental arrest stage. At best, donor hematopoietic stem cell osteo-medullary engraftment induces normal B-lymphocyte function and long-term thymopoiesis; at worst, peripheral expansion of donor T-lymphocytes from the stem cell source results in a restricted T-lymphocyte receptor repertoire with possible B-lymphocyte failure. Conditioning improves immunoreconstitution but causes short- and long-term toxicities, and increased mortality. Newborn screening for SCID will propel the search for safe, effective methods of achieving donor cell engraftment and full immunoreconstitution without toxic sequalae.

Keywords:

• chemotherapy conditioning
• hematopoietic stem cell transplantation
• severe combined immunodeficiency
• thymopoiesis

Severe combined immunodeficiencies (SCID) are a heterogenous group of inherited primary immunodeficiencies characterized by absent thymopoiesis, T-lymphocyte maturation and function and affect cellular and humoral acquired immunity; without definitive treatment, the condition is invariably fatal, usually within the first year of life. Classical presentation is with persistent viral respiratory or gastrointestinal infection in infancy, with failure to clear virus and persistent and deteriorating symptoms [1]. Multiple pathogens may co-exist, and opportunistic infection, for example with Pneumocystis jiroveci, is common. The genetic bases of 75–80% of SCID types are known. Definitive treatment is predominantly by allogeneic hematopoietic stem cell transplantation (HSCT), although gene therapy and enzyme replacement therapy are available for some specific genetic sub-types. Depending on the genetic defect, recipient B-lymphocyte and/or NK cells may be present. In contrast to treatment of patients with malignancies, where eradication of malignancy is required, the objective of HSCT in patients with SCID is to provide normal hematopoietic stem cells, facilitating correction of the immune defect. Peripheral T-lymphocytes confer medium-term but finite immune protection, with a limited T-lymphocyte receptor repertoire. Therefore, it is important to establish effective long-term thymopoiesis and immune-reconstitution while minimizing potential sequelae of treatment. T-lymphocyte progenitors enter the thymus and undergo intense proliferation during the thymocyte double negative (DN) 1 and 2 stages, before DN3, where cells stop proliferating, and undergo rearrangement of the T-lymphocyte receptor. The lack of recipient T-lymphocytes to facilitate rejection and the idea of host environments which are either permissive or non-permissive to engraftment following infusion has provoked debate over whether infusion of donor stem cells, without administering immunosuppressive and myeloablative pre-transplant chemotherapy conditioning, is more effective and safe in realizing a long-term cure than transplantation of donor stem cells following administration of myeloablative chemotherapy conditioning. As new information is published, there is clarification about which molecular diagnoses permit good long-term thymopoiesis after graft infusion without conditioning and which procedures likely require conditioning. Unfortunately, reported cohorts of patients are rarely matched for age at diagnosis, molecular defect, pre-transplant co-morbidities and pre-transplant conditioning regimens administered. Important aspects to consider when deciding if pre-transplant conditioning should be administered include the presence of pre-existing infection with end-organ damage, the molecular diagnosis, the type of donor available, the likelihood of full immune reconstitution and the risk of short-term and long-term sequalae of chemotherapy.

Frequently used pre-transplant conditioning agents include busulfan, cyclophosphamide, melphalan, fludarabine, thiotepa and treosulfan, a less toxic analogue of busulfan [2]. Regimens targeting busulfan to a sub-myeloablative dose are also increasingly used to reduce unwanted toxic sequalae [3], which include short-term complications such as hematopoietic aplasia, infection, alopecia, mucocitis, veno-occlusive disease and lung toxicity, and long-term sequalae including growth failure, endocrine dysfunction and infertility.

Infusion of unmanipulated or T-lymphocyte-depleted marrow from HLA-matched sibling or haplo-identical parental donors without conditioning enables T-lymphocyte immune reconstitution and thymopoeisis, particularly in patients with IL-2 receptor gamma chain (IL-2RG)-, janus-associated kinase 3 (JAK3)- or adenosine deaminase-(ADA) deficient SCID, with survival of around 80% [4], although recipient B-lymphocyte function remains impaired without the development of donor lymphocytes, at least in IL-2RG- or JAK3-SCID [5]. Better results are observed in infants treated in the first month of life, before the onset of infection, who demonstrate a survival of >90%, and fewer long-term infectious and immune-related complications [6,7]. The outcome is also determined by the immunophenotype (and, therefore, also genotype) and is best when using HLA-matched sibling or family donors, compared to matched unrelated or mismatched family donors, with significantly better overall survival and event-free survival in NK– SCID when compared with NK+ SCID. During long-term follow-up, superior thymopoiesis is evident in IL-2RG-, JAK3-, IL-7 receptor α- (IL-7RA) or ADA-deficient SCID compared with recombination activating genes 1 and 2- (RAG1/2) and DCLRE1C-deficient SCID. However, in the absence of preconditioning, donor B-lymphocyte and myeloid chimerism is generally absent in either group [8]. Graft versus host disease, a significant complication of HSCT, may occur after infusion of an HLA-matched sibling donor innoculum, but is more common when HLA-matched unrelated donors are used, although the use of serotherapy reduces the risk of graft versus host disease and improves survival when HLA-matched unrelated donors are used [9].

The likelihood of long-term B-lymphocyte function and thymopoiesis is dependent on donor HSCs or lymphoid progenitors occupying osteo-medullary or thymic niches. ADA-deficient SCID is a disorder of purine metabolism resulting in the accumulation of toxic metabolites that arrest T-lymphocyte, B-lymphocyte and NK cell development and function, but affect other hematopoietic and non-hematopietic cells. Survival is best in non-conditioned transplants, and despite the absence of chemotherapy, a high engraftment rate is achieved, including donor B-lymphocyte engraftment, although donor myeloid engraftment is less likely to occur. However, in ADA-SCID, long-term thymopoiesis is reduced in non-conditioned procedures, compared to those where reduced intensity conditioning is used [10]. T-lymphocyte immunoreconstitution of patients with IL-2RG- or JAK3-deficient SCID demonstrate sustained thymopoeisis following infusion of unmanipulated matched, or haplo-identical T-lymphocyte-depleted HSC innoculi, although thymopoiesis may diminish more rapidly than in the normal population, at least as demonstrated in one study using SBA-SRBC processed T-lymphocyte-depleted grafts [11]. In these SCID sub-types, donor HSC engraftment is not required for sustained thymopoiesis, but B-lymphocyte function is dependent on donor B-lymphocyte chimerism, which is predicted by donor myeloid donor chimerism, a likely surrogate for donor stem cell engraftment [12]. These genetic defects are associated with recipient mature B-lymphocyte development (and thus full occupation of osteo-medullary niches), although B-lymphocytes demonstrate defective proliferation, plasmablast differentiation and antibody secretion following stimulation with CD40L and IL-21, and thus are non-functional [5]. T-lymphocyte development is arrested at the double-negative 1 stage of thymocyte development, and thus thymic niches are devoid of developing T-lymphocytes. Infusion of donor stem cells facilitates engraftment of T-lymphocyte progenitors in the thymic niche, giving rise to ongoing thymopoiesis. However, the lack of open osteo-medullary niches, because of complete, albeit non-functional, B-lymphocyte development, precludes engraftment of donor B-lymphocyte pre-cursors. Administration of chemotherapy creates space within the osteo-medullary niche for sufficient donor stem cell engraftment, to facilitate donor B-lymphocyte development and continuous export of T-lymphocyte progenitors to the thymic niche, resulting in continued replenishment of the thymic niche with T-lymphocyte progenitors giving rise to long-term thymopoiesis. Mutations in the IL7RA gene give rise to T-B+NK+ SCID. B-lymphocyte development is normal and function is potentially normal providing normal donor T cells are present. T-lymphocyte development is arrested at DN1. Infusion of donor HSC without conditioning in IL7RA-deficient SCID facilitates donor thymocyte engraftment within unoccupied thymic niches facilitating long-term thymopoiesis. Recipient B-lymphocyte interaction with functional donor T-lymphocytes results in B-lymphocyte function [13,14]. Recombination activating genes 1 and 2 (RAG1/2) and DCLRE1C which codes for artemis are critical for the recombination of gene segments required to construct the T- and B-lymphocyte antigen receptor, occurring during the DN3 progenitor stage of thymocyte development. Mutations in these genes give rise to T-B-NK+ SCID. Osteo-medullary and thymic niches are occupied by proliferating lymphocyte progenitors, blocked at the pre-B-1 and DN3 stages, respectively. Infusion of haplo-identical T-lymphocyte-depleted donor inoculum without clearing either niche generally leads to graft failure, which is resolved if chemotherapy conditioning is given [15]. Infusion of unmanipulated HLA-matched donor cells has a better outcome, although T-lymphocyte counts remain low and B-lymphocyte engraftment is variable. However, use of alkylating agents in patients with artemis-deficient SCID, while improving engraftment, results in significant long-term sequalae, including growth deficiency, requirement for nutritional support and dental abnormalities [15].

The use of chemotherapy pre-HSCT improves the quality of engraftment and immunoreconstitution, particularly when an HLA-matched sibling donor is unavailable. Critically, infants aged 3.5 months of age or younger or infants of any age without active infection at time of HSCT have high survival rates regardless of donor type or conditioning. Although the survival in patients with active infection at the time of transplant is significantly better when a haplocompatible donor is used and no conditioning is given, immunoreconstitution is better when conditioning is used. However, the benefit for the individual patient has to be balanced against the risks of adverse effects [7]. Antibody-based conditioning alone [16,17] appears to be less effective than as part of a minimal intensity regimen [18].

In future, urgent steps should be taken to implement newborn screening for SCID in Europe, as in the USA, given the excellent results of treatment for young, non-infected infants [7,19,20]. Further investigation of antibody combinations, possibly with targeted or reduced toxicity chemotherapy, should be performed, preferably in a trial setting, to improve the quality and durability of immunoreconstitution. Finally, detailed long-term assessment of immunoreconstitution, organ function and life quality of patients treated in the early era of HSCT for SCID should be performed, to gain a holistic view of the long-term outcome of SCID patients transplanted with or without chemotherapy.

Financial & competing interest’s disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.