A 2-hit model for chronic GVHD

Robertson Parkman

In this issue of Blood, Dertschnig and colleagues1 demonstrate in mice that acute graft-versus-host disease (GHVD) results in a marked reduction of autoimmune receptor–expressing medullary thymic epithelial cells (Aire+ mTEC) and a decrease in the diversity of Aire-dependent tissue-restricted peripheral self-antigens (TRAs) required for effective negative thymic selection. Both of these abnormalities are reversed by the peritransplant administration of the epithelial protectant drug, fibroblast growth factor 7(Fgf7).

Chronic GVHD continues to be a major cause of both morbidity and mortality after allogeneic hematopoietic stem cell transplantation (HSCT).2 Chronic GVHD has been assumed to be caused by the continuation of the pathogeneic mechanisms that cause acute GVHD, primarily donor-derived T lymphocytes specific for histocompatibility antigens uniquely expressed by recipient cells.3 As a consequence, therapy for chronic GVHD has traditionally been directed at suppressing the donor antirecipient immune response. However, during the last 25 years, a series of murine experiments have indicated that donor-derived, autoreactive T lymphocytes (ie, T lymphocytes specific for antigens expressed by both donor and recipient cells) are present in murine HSCT recipients with chronic GVHD and that the chronic GVHD could be transferred by the donor-derived T lymphocytes into both donor and recipient mice.4-6

More recently, a clinical trial of low-dose subcutaneous IL-2 in human HSCT recipients with established chronic GVHD demonstrated that an increase in circulating regulatory T lymphocytes (Treg) resulted in clinical improvement, suggesting that deficiencies in Treg lymphocytes play a role in the pathogenesis of human chronic GVHD.7

Dertschnig et al report that mice with acute GVHD have a profound decrease in the frequency of Aire+ mTEC, which are necessary for the thymic production of naturally occurring Treg lymphocytes. Other investigators have previously demonstrated that the diverse expression of TRA by Aire+ mTEC is required for the effective thymic elimination of autoreactive T lymphocytes by negative selection.8 Using microarray analyses of isolated Aire+ mTEC, the present investigators report the decreased expression and diversity of TRA with a selective decrease in the TRA associated with the tissues that are the target organs of human chronic GVHD (skin, liver, salivary glands, lung, eye, and gastrointestinal tract). The result of the restricted diversity of TRA expression would be the extrathymic presence of autoreactive T lymphocytes. Their present results suggest a 2-hit model for chronic GVHD, in which the presence of extrathymic autoreactive T lymphocytes (Hit 1) in the context of immune dysregulation (deficiencies in Treg lymphocytes, Hit 2) can result in the development of chronic GVHD.

The peri-HSCT administration of the epithelial protectant drug, Fgf7, does not affect the initial post-HSCT decrease in mTEC but does hasten the recovery of Aire+ mTEC and improves the diversity of TRA expression. Fgf7 acts by stimulating the proliferation and differentiation of TEC progenitors and the proliferation of residual mTEC.9 The relative contribution of the 2 populations to the recovery of Aire+ mTEC is unclear. The presence of normal numbers of Aire+ mTEC with normal TRA diversity during the recapitulation of immunologic ontogeny, which occurs after the engraftment of donor hematopoietic stem cells, may result in the presence of adequate numbers of circulating Treg lymphocytes and effective negative thymic selection, which would eliminate the peripheral presence of autoreactive T lymphocytes, and an absence of chronic GVHD. As such, clinical trials to determine whether the peritransplant administration of Fgf7 results in a decreased incidence of chronic GVHD in human HSCT recipients are indicated.

The present murine experiments, however, do not address several potentially important clinical questions: (1) What would be the impact of the administration of Fgf7 to patients with established chronic GVHD?; (2) Do HSCT recipients with established chronic GVHD have an adequate number of TEC progenitors and residual mTEC for the Fgf7 to be effective?; (3) Do the TEC progenitors and mTEC in chronic GVHD patients become refractory to the action of Fgf7?; and (4) Is the loss of Aire+ mTEC during acute GVHD paralleled by a decrease in TRA diversity, or are they separable biological processes? If they differ, then some HSCT recipients may have a deficiency of Treg lymphocytes without the concomitant presence of peripheral autoreactive T lymphocytes, whereas other recipients may have adequate numbers of Treg lymphocytes with the presence of peripheral autoreactive T lymphocytes. Only HSCT recipients with both the presence of peripheral autoreactive T lymphocytes (Hit 1) and deficiencies in Treg lymphocytes (Hit 2) would be at risk of developing chronic GVHD. The immunophenotypic identification of functional human Treg lymphocytes will aid in the evaluation of human chronic GVHD; however, the inability to identify immunophenotypic human autoreactive T lymphocytes is still a major limitation to our better understanding.10

Finally, the present results suggest that the focus of future research into new therapies to prevent or treat chronic GVHD should be directed at the thymic microenvironment rather than at lymphohematopoietic cells.


  • Conflict-of-interest disclosure: The author declares no competing financial interests.