Reconstitution of humoral immunity

In the period immediately following hematopoietic progenitor cell transplantation, preparative regimen-resistant recipient plasma cells are largely responsible for the production of IgG. By about 4–6 months, progenitors contained in the allograft begin to give rise to naïve B cells producing IgM. Isotype switching, resulting in production of IgG, and subsequently IgA, occurs later and depends on the availability of CD4 'help' such that serum levels of IgG can remain abnormal for up to 1 year following transplant. Furthermore, even when normal levels of IgG are achieved, an oligoclonal pattern may be present for several months following transplantation. The presence of grade 3 or 4 acute GVHD or chronic extensive GVHD is correlated with reduced numbers of B cells for at least 1 year following hematopoietic progenitor cell transplantation, as well a diminished availability of CD4+-mediated T-cell help.^{5, 6} Thus, recovery of a complete repertoire of donor-derived IgG- and IgA-producing B cells is typically delayed for several months following transplantation, and in the presence of GVHD the period of B-cell deficiency often extends to 1–2 years.

T-cell immune reconstitution

T-cell reconstitution occurs via two predominant pathways: a thymic-dependent pathway that recapitulates ontogeny and a thymic-independent pathway termed 'homeostatic peripheral expansion' (HPE) that involves expansion of mature T cells which survive the preparative regimen and/or are contained within the allograft or within donor lymphocyte infusions. Although the existence of homeostatic peripheral expansion has long been appreciated,^{7, 8} it has recently been the subject of intense research. Multiple lines of evidence from murine models and human studies have demonstrated that, in contrast to thymic production of T cells, HPE generates a T-cell pool with both quantitative and qualitative deficiencies, resulting in impaired functional immunity to stringent antigens.^{9, 10} One factor contributing to the limited effectiveness of HPE is the loss of T cells due to a high rate of apoptosis.^{10, 11} Furthermore, unlike the broad T-cell repertoire which is generated via thymopoiesis, the T-cell repertoire that is generated during HPE is restricted by the TCR specificities contained in the starting inocula and by the antigens that drive this process.¹² Recently, it has become clear that although cognate antigens can dramatically skew the homeostatically expanding repertoire, naïve cells present in the lymphopenic hosts acquire the capacity to response to weak or low-affinity antigens, thus preserving the repertoire of the naïve pool as much as possible in the absence of thymopoiesis.^{13, 14} Thus, while recovery of thymic function provides the optimal pathway for T-cell immune reconstitution, substantial restoration of immune function can occur through homeostatic peripheral expansion alone. Furthermore, because homeostatic peripheral expansion results in a propensity for augmented proliferation in response to weak antigens in the lymphopenic setting, it is possible that exploitation of this process using tumor vaccines or other strategies could result in augmented antitumor effects.

Thymic recovery does not generally occur for several months following allogeneic stem cell transplantation. This has been documented using phenotypic markers of thymic emigrants (e.g. CD45RA and CD62L expression on CD4+ cells),¹⁵ as well as using newer techniques used to measure thymic function, including measurement of T-cell receptor excision circles¹⁶ and T-cell receptor repertoire analysis.¹⁷ Thymic function is limited by age-related declines in thymic function, and thymic toxicity induced by cytotoxic therapy, radiation or GVHD. Furthermore, because selection of the peripheral T-cell repertoire is largely dictated by MHC expression on the thymic epithelium (rather than by MHC expression on bone marrow derived cells), the thymic selection process which will occur in MHC-mismatched stem cell transplantation is not predicted to select as complete a repertoire as can be achieved in the fully MHC matched setting.¹⁸ As a result, depending on age, degree of MHC mismatch and the presence of GVHD, recovery of normal T-cell number and function often does not occur within the first year following hematopoietic progenitor cell transplantation.

The interaction between GVHD and immune reconstitution is critical to consider as one seeks to develop new ways to enhance immune recovery following allogeneic stem cell transplantation (Figure 2). The graft-versus-host reaction is directly toxic to the thymic microenvironment and contributes significantly to delayed thymic recovery.^{19, 20} Furthermore, alloreactivity may result in impaired negative selection of host reactive T cells. Thus, the net effect is that GVHD results in poor thymic function with altered T-cell selection, leading, in turn, to increased GVHD. In addition, GVHD also significantly attenuates the effectiveness of homeostatic peripheral expansion by causing widespread bystander apoptosis and thus further limiting the repertoire regenerated via this pathway.²¹ Finally, immunosuppressants used to prevent or control GVHD diminish both thymic function and homeostatic peripheral expansion. Thus, the pathophysiology of GHVD as well as the immune suppression required to treat GVHD result in long-term immune deficiency, which presents a barrier to the application of tumor- or infection-specific immunotherapies. These findings support the tenet that optimal reconstitution of adaptive immunity following allogeneic stem cell transplantation can occur only in the absence of GVHD.

Figure 2.

Efforts to improve immune reconstitution following hematopoietic progenitor cell transplantation could (a) target thymopoiesis by augmenting thymic throughput or by augmenting expansion of recent thymic emigrants or (b) augment the number of antigen-specific populations present in vivo, which recognize problematic pathogens or tumor-associated antigens. Whereas thymopoietic approaches are likely to provide global improvements in immunity, they may not result in rapid antitumor effects. Similarly, whereas adoptive therapies may allow rapid targeting of specific antigens, these approaches are not likely to result in 'global' immune reconstitution. Depending upon the specificity of the approach, either approach could worsen GVHD and ultimately diminish immune competence; hence, careful attention must be given to the effects of immunorestorative therapy on alloreactivity/GVHD.

Full figure and legend (55K)

Paradoxically however, the use of T-cell depletion to prevent GVHD also adversely effects immune reconstitution by limiting the efficiency of homeostatic peripheral expansion. The most serious and obvious impact of impaired immune reconstitution following T-cell depletion is the higher rate of leukemic recurrence observed in this setting. As leukemic control is largely mediated by reconstituting T cells, the development of new approaches which can specifically enhance immune reconstitution toward problematic pathogens and tumor antigens without increasing the numbers of T cells responding to alloantigens is necessary in order to improve the outcome for patients undergoing allogeneic stem cell transplantation.

Potential strategies to enhance immune reconstitution

Potential strategies for enhancing immune reconstitution can be largely divided into adoptive cellular therapies specifically aimed toward restoring immunity toward problematic pathogens and toward tumor antigens vs hormonal or cytokine-based therapies designed to improve overall immunity by augmenting thymopoiesis itself and/or homeostatic peripheral expansion of recent thymic emigrants. With regard to adoptive cellular therapy, the simple transfer of unmanipulated donor leukocytes has proven remarkably effective in controlling disease recurrence in chronic myelogenous leukemia, but much less success has been observed with other malignancies.^{22, 23, 24} Furthermore, since unmanipulated DLI contain large numbers of alloreactive T cells, the therapeutic index between a clinically significant response and clinically significant GVHD is narrow; thus DLI may ultimately adversely effect immune reconstitution if GVHD is induced. To improve the effectiveness of DLI, a number of centers are exploring the use of adoptive therapy using lymphocytes enriched for tumor-reactive T cells and/or depleted of potentially pathogenic alloreactive cells. In the autologous setting, impressive responses have been observed when tumor-infiltrating lymphocytes are expanded in vitro and reinfused following lymphodepleting therapy.²⁵ Presumably, this relates, in part, to a 'receptive environment' which occurs in the setting of lymphopenia and results in dramatic clonal expansions described above as 'homeostatic peripheral expansion'. Other investigators have used viral specific cell lines and clones to target CMV and/or EBV following allogeneic progenitor cell transplantation, and it is likely that approaches using tumor-reactive lines or clones will soon be administered in allogeneic setting. Importantly, the use of therapeutic vaccines to further expand adoptively transferred cells in vivo is an attractive approach. The use of tumor vaccines alone to target leukemic antigens has already been undertaken following allogeneic hematopoietic transplantation²⁶ and it is likely that this will ultimately be combined with adoptive transfers of cell populations. Further studies are needed, however, in order to optimize the vaccine approaches and antigenic targets for individual malignancies.

As noted earlier, the optimal pathway for full T-cell recovery is through restoration of thymic T-cell differentiation. Alternatively, augmenting homeostatic peripheral expansion of recent thymic emigrants (which are depleted of alloreactive cells during thymic selection) would also be predicted to improve the overall immune competence. Thus, there has been much interest in strategies to enhance thymic throughput or to enhance the expansion of recent thymic emigrants. Although age-related thymic involution contributes significantly to delayed T-cell recovery, it is evident that toxicity related to the preparative regimen can also contribute. Thus, one way to improve immune reconstitution is to protect the thymic epithelium from the toxic effects of radiation and chemotherapy. Keratinocyte growth factor (KGF) has been studied as a possible means to achieve this goal. In murine chimeras, KGF has been shown to provide remarkable protection from thymotoxic effects of irradiation, resulting in improved immune reconstitution.²⁷ Whether these effects in a highly controlled setting as occurs in murine GVHD can be translated to the more protracted setting of ongoing thymic toxicity in clinical transplantation remains to be seen.

IL-7 receptor signaling is required for thymic T-cell development, as demonstrated by the fact that individuals bearing a mutation in the IL-7R chain show T-cell-deficient SCID.²⁸ IL-7 is also required for the HPE of mature T cells following T-cell depletion, and can dramatically augment this process in murine models. Interestingly, although treatment with IL-7 increases the number of thymic-derived T-cell progeny following murine bone marrow transplantation, non-human primate models reveal a decline in TREC levels in IL-7-treated animals.²⁹ Thus, although some augmentation of thymic throughout by IL-7 cannot be ruled out, current studies suggest that IL-7's primary effect is on peripheral T cells, where it can remarkably enhance the homeostatic peripheral expansion of both recent thymic emigrants as well as other resident T cells. As might be predicted therefore, IL-7-mediated augmentation of antigen-specific proliferation reduces the T-cell threshold at which GVHD is observed in murine models.³⁰ Indeed, although augmented immune reconstitution is observed in mice treated with T-cell-depleted progenitor cell transplants and IL-7, this does not occur when T cells are present in the graft.³⁰ Thus, while IL-7 remains the most potent immunorestorative described thus far, clinical use of IL-7 following allogeneic progenitor cell transplantation will likely be most effective when GVHD is prevented through T-cell depletion.

Finally, we have demonstrated that HPE is critically dependent on the availability of appropriate antigen presentation and that expansion of dendritic cells with Flt3 ligand can increase the HPE of mature cells. Furthermore, Flt3 ligand also had a beneficial effect on the generation of new T cells via the thymus, evidenced by increased thymic cellularity and increased TREC levels.³¹ Thus, Flt3 ligand is another agent which in murine models has been shown to augment immune reconstitution, and it appears to work through a combination of increasing thymopoiesis and increasing homeostatic peripheral expansion. Notably, Flt3 ligand also substantially enhances NK cell reconstitution.³¹ While NK cells are the most rapid to recover following allogeneic stem cell transplantation and therefore NK cell deficiency is not a clinical problem, there is emerging evidence that manipulations of NK populations post-transplant could potentially provide important antitumor effects. Ruggieri et al³² demonstrated that donors which have NK clones reactive with recipient hematopoietic populations (so-called alloreactive NK cells) result in a decreased risk of myeloid leukemic relapse post-transplant. Thus, it is possible that Flt3 ligand could serve several roles in immune reconstitution by augmenting the early recovery of NK cells, which may provide important graft-versus-leukemia effects, as well as by augmenting both homeostatic peripheral expansion and thymopoiesis, which are likely to be important for antileukemia effects and global immune reconstitution, respectively.

Other approaches under study to modulate thymic function include the administration of growth hormone or its analogs,³³ as well as inhibitors of sex steroids, which appear to diminish thymic function as evidenced by thymic rebound following castration of laboratory animals.³⁴

Summary

Many advances have occurred in the last 40 years, which have led to the remarkable success of hematopoietic progenitor cell transplantation. While hematopoietic recovery and recovery of the innate immune system are now reliably achieved and toxicity can be minimized, future progress largely hinges on the ability to more rapidly and reliably augment immune recovery such that problematic pathogens and tumor antigens can be more effectively targeted by the reconstituting immune system. The issues in developing immunorestorative agents for use in the setting of allogeneic stem cell transplantation are multiple, and reflect the complex interactions between immune reconstitution and GVHD. Future clinical studies are needed to optimize both the cellular approaches, which are predicted to allow more effective targeting of defined pathogens, or tumor antigens combined with the development of new agents, which can augment thymic function without exacerbating GVHD.

References

References