Abstract
Acute graft-vs-host disease (GVHD) is a major obstacle to safe allogeneic hematopoietic stem cell transplantation (HSCT), leading to a significant morbidity and mortality. GVHD occurs when transplanted donor T lymphocytes react to foreign host cells. It causes a wide variety of host tissue injuries. This review focuses on the pathobiological basis, clinical aspects, and current management strategies of acute GVHD. Afferent phase of acute GVHD starts with myeloablative conditioning, i.e., before the infusion of the graft. Total-body irradiation (TBI) or high-dose chemotherapy regimens cause extensive damage and activation in host tissues, which release inflammatory cytokines and enhance recipient major histocompatibility complex (MHC) antigens. Recognition of the foreign host antigens by donor T cells and activation, stimulation, and proliferation of T cells is crucial in the afferent phase. Effector phase of acute GVHD results in direct and indirect damage to host cells. The skin, gastrointestinal tract, and liver are major target organs of acute GVHD. Combination drug prophylaxis in GVHD is essential in all patients undergoing allogeneic HSCT. Steroids have remained the standard for the treatment of acute GVHD. Several clinical trials have evaluated monoclonal antibodies or receptor antagonist therapy for steroid-resistant acute GVHD, with different successes in a variety of settings. There are some newer promising agents like mycophenolate mofetil, glutamic acid-lysine-alanine-tyrosine (GLAT), rapamycin, and trimetrexate currently entering in the clinical studies, and other agents are in development. Future experimental and clinical studies on GVHD will shed further light on the better understanding of the disease pathobiology and generate the tools to treat malignant disorders with allogeneic HSCT with specific graft-vs-tumor effects devoid of GVHD.
Introduction
Allogeneic hematopoietic stem cell transplantation (HSCT) can cure a variety of malignant and nonmalignant disorders [
1
, 2
]. Graft-vs-host disease (GVHD) is a major cause of morbidity and mortality even when siblings are matched at the human leukocyte antigen (HLA) locus [3
, 4
, 5
, 6
, 7
]. GVHD in its chronic form can significantly affect the quality of life of long-term survivors following bone marrow transplantation (BMT) and also lead to mortality [7
, 8
, 9
].GVHD occurs when transplanted donor-derived T cells recognize and react to histoincompatible recipient antigens and cells. Final consequences of the GVHD process are a wide variety of host tissue injuries in varying degrees of clinical severity [
3
, 9
]. The fundamentals of GVHD include the transfer of genetically disparate donor-derived T cells into a host incapable of rejecting them [3
, 10
]. Three factors are required for the occurrence of graft-vs-host (GVH) reaction as outlined by Billingham, in his historical Harvey lecture in 1966 [10
]. The first requirement for GVH reaction is that the graft must contain a sufficient number of immunologically competent cells. The second requirement is that the host should have important transplantation isoantigens lacking in the graft. Hence, the host appears foreign to the graft and is capable of stimulating donor cells antigenically. The third is that the host immune system must be incapable of mounting an effective immune response against the graft, at least for a sufficient time for the latter to manifest its immunological competence.Specific host cells are recognized as foreign antigens by the alloreactive donor-derived T lymphocytes. Clinical manifestations of GVHD depend upon the degree of donor-host histocompatibility and graft alloreactivity to major host antigens. Epithelial cells of the skin and mucous membranes, biliary ducts, and intestinal tract crypts are primary tissue systems damaged during the pathobiological course of GVHD, although other organs in the human body may also be affected [
3
, 7
, 9
]. The aim of this review is to outline the essential pathobiological basis and current management strategies of acute GVHD.Genetic basis of acute GVHD
Major histocompatibility antigens encoded by the major histocompatibility complex (MHC) genetic loci have a major impact on transplantation and on the biological progress of GVHD [
3
, 11
]. MHC is a closely linked, highly polymorphic multi-gene and multi-allelic complex playing a central role in both cell-mediated and humoral immune responses. MHC genes are found in all mammals and vertebrates and consist of a number of closely linked genetic loci that function as a system. It is located on the short arm of chromosome 6 at the p21 position in humans and encodes HLA [12
]. Particularly class I and class II HLA are cell surface molecules controlling T-cell recognition and histocompatibility [3
, 12
, 14
]. HLA class I antigens (HLA-A, HLA-B, and HLA-C) have a wide distribution and are found on all nucleated cells [3
]. Matching BMT recipients with sibling donors sharing identical HLA antigens significantly improved engraftment kinetics and decreased GVHD severity [1
, 13
, 15
]. HLA class II antigens (DR, DQ, and DP) are found more selectively on the cells of the immune response system [3
]. CD4+ T cells recognize foreign antigens via the presentation of class II HLA molecules. The structure of the MHC class II molecule is strikingly similar to the structure of the class I molecule; however, there are some significant differences. The structure of the MHC class II region has been elucidated by X-ray crystallography [16
]. The DR1 molecule is a heterodimer. This pair of dimers may allow for simultaneous interaction with two T cell receptor complexes. Alternatively, in antigen presenting cells (APCs), this dimerization may induce expression of costimulatory molecules. Since class II HLA products are particularly induced on the skin and intestinal tract epithelial tissues, they may promote specific targeting during acute GVHD [17
, 18
].Minor histocompatibility antigens (miH) are peptides derived from intracellular proteins presented by MHC molecules to donor T cells [
19
]. Genetic polymorphisms of endogenous cellular proteins represent the miH. The miH are critical in matched-sibling allogeneic bone marrow graft. It has been demonstrated that T cells do not recognize antigens alone, but in conjunction with the MHC of the antigen presenting cells (usually self) [20
]. MHC molecule, with a single cleft where binding takes place, acts as a receptor for the antigen. This complex is then recognized by the T cell receptor (TCR). For that reason, alloreactivity is the recognition of different nonself peptides by the TCR bound and carried by the recipient MHC rather than the molecules themselves [21
]. Following the presentation of miH (foreign peptide) by MHC to donor T cell, i.e., CD4+ in the context of MHC class II and CD8+ in class I, the presence of nonself peptide bound to the MHC molecules trigger the T cell and induction of GVHD occurs. The roles of class III MHC and non-MHC genes in GVHD are less well defined [3
, 22
].Pathophysiology of acute GVHD
Distinct clinical forms of GVHD including acute and chronic GVHD are, in large part, a consequence of damage to host tissues by activated donor-derived T lymphocytes in response to the MHC disparities. However, the pathophysiological mechanisms of the different GVHD syndromes are not the same.
A three-phase model elucidates the three major processes that lead to GVHD. The first phase involves tissue damage secondary to the conditioning regimen, while the second phase consists of donor T-cell activation, stimulation, and proliferation. Those two phases make the afferent phase of GVHD. Finally, the effector phase comprises the third phase of GVHD pathophysiology [
23
, 24
].Phase 1 (conditioning)
Conditioning regimen: afferent phase of acute GVHD starts before the infusion of the graft.
The changes of acute GVHD start even before the allograft is infused during the conditioning regimen. The BMT conditioning regimen has a great impact in the pathogenesis of acute GVHD [
24
]. Myeloablative regimens with either total-body irradiation (TBI) or high-dose chemotherapy to cytoreduce underlying disease, and to suppress host defense for preventing graft rejection, can also cause extensive damage in host tissues including the intestinal mucosa, liver, and other tissues. Activated cells from damaged recipient tissues secrete many inflammatory cytokines, such as interleukin-1 (IL-1), tumor necrosis factor–alpha (TNF-α), granulocyte-macrophage colony-stimulating factor (GM-CSF), and interferon-γ (IFN-γ) [24
, 25
]. Dysregulated release of the cytokines may upregulate adhesion molecules and enhance recipient MHC antigens [- Xun C.Q.
- Thompson J.S.
- Jennings C.D.
- Brown S.A.
- Widmer M.B.
Effect of total body irradiation, busulfan-cyclophosphamide, or cyclophosphamide conditioning on inflammatory cytokine release and development of acute and chronic graft-vs-host disease in H-2–incompatible transplanted SCID mice.
Blood. 1994; 83: 2360
26
, 27
, 28
, 29
]. Increased expression of donor-tissue antigens may augment recognition of histoincompatible host-tissue antigens by alloreactive donor T cells [24
, 29
].The critical correlation between intensive conditioning regimens and increased risk of GVHD has been previously observed in human BMT [
30
, 31
]. Animal models have also demonstrated the important relationships between intensive conditioning, dysregulated inflammatory cytokine network, and GVHD [32
]. Furthermore, delaying the transfer of donor cells well after the intensive conditioning results in a decreased risk of acute GVHD as demonstrated in some experimental and clinical studies [33
, 34
, 35
].Phase 2 (induction and expansion)
Recognition of the foreign host antigens by donor T cells and activation, stimulation, and proliferation of T cells is crucial in the afferent phase of acute GVHD.
Presentation of recipient antigens to donor T cells, activation of donor T cells, and subsequent proliferation and differentiation of these activated T cells is crucial in this phase of acute GVHD. After the cellular component of the graft is infused, immunologically competent donor T cells recognize foreign host antigens presented by antigen presenting cells (APCs) in the context of MHC. APCs digest and process large proteins into smaller peptides; these peptides bind to MHC molecules and are presented on the surface of APCs as allopeptide-MHC complexes. Donor T-cell recognition and activation involves the interaction between the allopeptide-MHC and antigen-specific T cell receptors [
36
]. In murine models of GVHD, in which genetic discrepancies between multiple strains of mice combinations can be controlled, it has been established that donor CD4+ T cells induce GVHD to MHC class II molecule disparities, and donor CD8+ T cells induce GVHD via activation secondary to MHC class I molecule disparities [23
, 24
].When donor and recipient are not HLA-identical, a dramatic GVH reaction commonly occurs. Clinical data also support the occurrence of severe GVHD, even with single antigen difference [
37
, 38
]. When the recipient and donor are HLA-identical, the T cell and its TCR recognize the different peptides bound to the same MHC, the so-called minor histocompatibility antigens (miH), and GVHD takes place [23
, 39
]. Clinical trials have also validated the experimental models showing that mismatches of miH antigens between HLA-identical donors and recipients may be associated with significant GVHD [40
]. In a clinical study, five previously characterized miH antigens (HA-1, HA-2, HA-3, HA-4, and HA-5) that are recognized by T cells in association with HLA A1 and A2 were analyzed in 148 BMT patients. Mismatching of HA-1 alone was correlated with acute GVHD of grade II or higher (p = 0.02), and mismatching at HA-1, HA-2, HA-4, and HA-5 was also significantly associated with GVHD (p = 0.006) in HLA-identical sibling allogeneic BMT. In all cases where a HA-1+ patient received a HA-1− graft, acute GVHD developed; a mismatch at HA-3 had no effect [40
]. HA-1 miH antigen, a nonapeptide encoded by an allele of KIAA0223 gene, was found to have one amino acid difference from the HA-1+ counterpart [19
]. Moreover, peptide analysis of the HA-2 antigen has also been shown to belong to the class I myosin family [39
].After infusion of the graft, the systemic vasculature, including the capillary beds, represents the potential first and extensive area of contact with new alloantigens for the mature donor T cell [
23
]. Therefore, vascular antigens have been studied as potential miH antigens [41
, 42
, 43
]. One of the most common and highly expressed vascular antigens in the endothelial adhesion molecule is CD31 (PECAM-1) [42
, 44
]. This adhesion molecule represents a member of the immunoglobulin gene superfamily of adhesion molecules [45
]. CD31 is expressed on platelets, endothelial cells, monocytes, granulocytes, lymphocytes, and a variety of leukemic cell lines [23
]. Studies suggest that vascular antigens may be important in the pathogenesis of GVHD, although the data are conflicting [42
, 46
, 47
].Host APCs are particularly essential at the second phase of the GVH reaction. The APCs stimulate T cells via IL-1 and costimulatory signals to produce IL-2 and express the IL-2 receptor [
48
, 49
]. Under the effect of IL-2, alloreactive T cells clonally expand and differentiate into effector cells. It has been demonstrated that inactivating host APCs of hematopoietic origin can promote tolerance and reduce GVHD [50
, 51
].Donor T-cell adhesion and activation
As discussed before, host vascular endothelium antigens not only represent the first area of contact and new alloantigens that the donor-derived mature T cell may recognize when grafted into the recipient, but they also serve as molecules that donor T cells may adhere to. Donor T cells in the host bloodstream must adhere long enough to the endothelium to become activated. T cells roll along the endothelial surfaces with their TCR in contact with a variety of different antigens. It has been suggested that the binding Km of TCR to the MHC molecules is relatively weak, which raises the likelihood of TCR engaging with a large number of MHC molecules [
52
]. If its TCR recognizes any particular antigen that will activate the donor T cell, adhesion molecules play an important role in firmly anchoring the T cell and preventing further rolling. A variety of adhesion molecules are important in this process and in setting up the conditions necessary for the activation of this T cell. CD62E (E-selectin), α4β1 integrin (VLA-4), αLβ2 integrin (LFA-1, CD11a/CD18), ICAM-1 (intercellular adhesion molecule-1), PECAM-1 (platelet and endothelial cell adhesion molecule-1, CD31), and VCAM-1 (vascular cell adhesion molecule-1) are some of the adhesion molecules involved in these processes [23
, 53
].Donor T-cell activation and costimulation
Donor T-cell activation requires two signaling events. TCR-peptide-MHC interaction, and a lattice formation between allopeptide bound to host MHC and donor T cell receptor is the first signal [
54
, 55
]. The second, i.e., costimulatory signal, is provided by APCs and requires cell-to-cell contact [56
, 57
]. The outcome of the first signal is regulated by the second signal. Three outcomes may occur according to the interaction of first and second signaling events: complete activation, partial activation, or anergy, i.e., a long-lasting state of antigen-specific unresponsiveness.There are several costimulatory molecules for resting T cells, antigen-primed T lymphocytes, and T helper cell clones. B7 antigens, B7-1 (CD80) and B7-2 (CD86), and the CD40/CD40L receptor-ligand pairs are among the best characterized costimulatory molecules. CD28 and CTLA-4 are two T-cell surface receptors on which B7 ligands bind. Normally, a signal from the TCR, a costimulatory signal from CD28, and an inhibitory signal from CTLA-4 determine the outcome of T-cell activation. This process was described in mice deficient for either CD28 or CTLA-4 [
58
, 59
]. In the absence of a CD28 costimulatory signal, signaling through the TCR complex results in a signal that leads to anergy. On the other hand, absence of CTLA-4 results in loss of the inhibitory signal, finally resulting in augmented and uncontrolled cytokine production and proliferation [58
, 59
]. Experimental data from murine BMT models suggest that CTLA4-Ig blocks T-cell costimulation through the B7-CD28 signal pathway. In addition, the anti-CD40L (CD154) monoclonal antibody (mAb), which can interfere with the interaction of CD154 on T cells and CD40 on APCs, can induce long-term graft [60
]. Experimental data illustrate the importance of costimulatory molecules in T-cell activation or creating a state of anergy [- Seung E.
- Iwakoshi N.
- Woda B.A.
- et al.
Allogeneic hematopoietic chimerism in mice treated with sublethal myeloablation and anti-CD154 antibody absence of graft-vs-host disease, induction of skin allograft tolerance, and prevention of recurrent autoimmunity in islet-allografted NOD/Lt mice.
Blood. 2000; 95: 2175
61
]. The use of CTLA4-Ig to block B7:CD28 interaction to inhibit alloreactive donor T cells and induce a state of anergy has been shown in a clinical phase I trial [51
]. In that clinical study, 12 patients received a one full haplotype-mismatched allogeneic BMT and all successfully engrafted. Moreover, only 3 patients developed gut GVHD, which was far less than anticipated GVHD survival in the murine model.In summary, in vivo T-cell activation is very complex. Costimulatory requirements for T cells depend on their state of activation-induced maturation. T-cell activation depends upon both the state of activation of the T cell (resting vs activated, naı̈ve vs mature) and the nature of the APC (professional vs nonprofessional, resting vs activated) [
61
].Phase 3 (effector phase)
Effector phase of acute GVHD: Direct and indirect damage to host cells.
A variety of complex mechanisms are involved in this phase of GVHD, which is responsible for end-organ dysfunction and tissue damage. Donor T cells either directly, or by using secondary mechanisms, attach to host cells in the efferent phase of GVHD. Once T lymphocytes are activated and proliferated, they release a variety of inflammatory cytokines having dual roles in complex immune responses of acute GVHD [
3
].TH1 and TH2 cytokines
Both CD4+ and CD8+ cells secrete cytokines such as IL-2, GM-CSF, TNF-α, and IFN-γ when MHC-mismatched BMT is performed in murine models [
62
]. These cytokines can then activate other T cells or other cell types, such as monocytes, NK cells, or NK-like cells including residual host cells [3
].There are at least two types of T helper (TH) cells that play intricate roles in GVH reactions, TH1 and TH2 cells. This is based on the cytokines that these T cells generate, and the cytokines to which these T cells respond. TH1 cells produce IL-2 and γ-interferon; in contrast, TH2 cells produce IL-4, IL-6, and IL-10. TH1 cells make and use IL-2 for their growth, TH2 cells produce IL-4 and require IL-1 or IL-4 for proliferation. A preponderance of TH1 cells could lead to activation of cytotoxic T lymphocytes (CTLs) and subsequent GVHD, compared to an abundance of TH2 cells, leading to a humoral response and prevention of GVHD [
63
, 64
]. Therefore, if donor T cells are driven by IL-10 to TH2 phenotype cells prior to the transfer of these cells into the host, these cells do not cause GVHD. Allogeneic T cells (CD4+) activated in the presence of IL-10 result in a state of long-term anergy [65
]. These experimental data have also been observed in recent clinical allogeneic BMT [66
, 67
]. Although the donor T-cell dose in the graft is higher in allogeneic peripheral blood stem cell transplantation (allogeneic PBSCT) compared to unmobilized allogeneic BMT, this did not seem to result in an increased risk of acute GVHD. This protection from increased risk of GVHD may be attributable to donor progenitor cell mobilization with granulocyte colony-stimulating factor (G-CSF) or GM-CSF, which may polarize donor T cells towards IL-10–driven TH2-type cells [66
]. Although there is no increased risk of acute GVHD with allogeneic PBSCT, there may be a tendency toward increased chronic GVHD [67
, 68
, 69
]. There is a decreased relapse risk with allogeneic PBSCT. Although the roles of TH1-type cytokines (i.e., IL-2 and IFN-γ); i.e., in inducing GVHD and being deleterious, and TH2-type cytokines (IL-4 and IL-10); i.e., being suppressive and protective of GVHD, are widely accepted, this paradigm may not be so clear-cut, and contradictory data also exist [70
, 71
]. Experimental data note the use of IL-12 during early posttransplantation abrogating GVHD in an experimental model [72
]. Donor CD4+ T cells secrete IL-2 in the initial days after experimental allogeneic BMT [73
]. Inhibiting IL-2 with antibodies to IL-2 receptor or IL-2 ligand can inhibit the development of experimental GVHD. In clinical studies, IL-2 producing T cells pre-BMT has been used to predict the risk of GVHD [74
, 75
], and soluble IL-2 receptor levels show a close relationship with disease severity and may be a sensitive indicator of GVHD onset [76
].Cytokines
The term “cytokine storm” was used and appropriately defined what is observed clinically in patients [
23
, 24
]. Among the proinflammatory cytokines, IL-1 production by activated monocytes takes place early in the inflammatory responses and may further lead to expression of TNF-α as well as other cytokines [24
, 77
, 78
]. IL-2 added exogenously during the first week of GVH reaction can lead to enhanced severity and mortality associated with class I MHC-mismatched BMT in mice; however, high doses of IL-2 administered over a brief period beginning on the day of BMT appears to mediate a protective effect against GVHD in murine models while preserving the allogeneic engraftment [79
, 80
]. It appears that the graft-vs-leukemia effect is protected. The mechanism of GVHD protection by IL-2 administration might involve splenic cells with CD3+CD4−CD8− phenotype that were increased in IL-2–treated animals 3 or 4 days during early post-BMT period [81
]. Limited data are available on the roles of IL-6 and IL-8 (neutrophil activating peptide) in the pathophysiology of GVHD [82
, 83
]. Serum IL-6 receptor and IL-6 concentrations are elevated in patients who developed acute GVHD [82
]. In a clinical study among patients who underwent BMT for thalassemia, IL-8 concentrations were found to be higher in patients with acute GVHD than the patients who did not develop this complication [83
]. In an experimental murine model across MHC and miH antigen barriers, it has also been shown that IL-11 may drive TH2-type cytokines with significant increases in IL-4 levels and decreasing TNF-α levels, thus decreasing GVHD-related mortality [84
]. In a severe combined immunodeficient (SCID) mouse model, an early increase in IFN-γ secretion with a synchronized increase in activated T cells and mRNA expression of IL-12, IL-18, and their respective receptors has been demonstrated after the establishment of acute GVHD by injecting donor spleen cell transfer [85
].T cells can mediate their cytotoxicity through soluble mediators such as TNF-α. The role of TNF-α in GVHD was elucidated from the studies evaluating the role of gut flora in GVHD. Experimental data suggest that there is a reduction in GVHD following BMT in gnotobiotic (free of pathogens) mice [
86
]. Clinical studies also support these experiments, suggesting that gut decontamination and laminar air flow rooms may decrease the severity of GVHD in selected allogeneic BMT patients [87
, 88
]. Radiation and chemotherapy cause damage to the gut. This allows gut flora (i.e., bacteria) and endotoxins (lipopolysaccharides; LPS), a well-known stimulus for TNF and other cytokines, to enter the circulation. Data supporting this explanation came from a trial of monoclonal antibody against endotoxin, which seems to protect mice against GVHD [- Beelen D.W.
- Elmaagacli A.
- Müller K.D.
- Hirche H.
- Schaefer U.W.
Influence of intestinal bacterial decontamination using metronidazole and ciprofloxacin or ciprofloxacin alone on the development of acute graft-vs-host disease after marrow transplantation in patients with hematologic malignancies final results and long-term follow-up of an open-label prospective randomized trial.
Blood. 1999; 93: 3267
89
]. Increased serum TNF levels were found in patients who developed GVHD [90
]. Furthermore, the use of monoclonal antibodies against TNF-α receptor in patients with severe acute GVHD refractory to conventional therapy has resulted in a significant reduction in GVHD-related skin, liver, and intestinal lesions, but unfortunately GVHD recurred when therapy was discontinued [91
]. Anti–TNF-α antibody can reduce skin GVHD lesions [92
]. Recent experimental data suggest that keratinocyte-growth factor (KGF) may prevent mucositis via prevention of bacterial and LPS translocation from gut to the circulation, may be protective of GVHD, and graft-vs-leukemia (GVL) effect is also preserved [93
, 94
].Cell mediated cytotoxicity
Although tissue damage in the effector phase of GVHD can result from the cytolytic function of CTLs, other effector cells such as natural killer (NK) cells may also contribute to the GVHD damage through release of inflammatory cytokines and nitric oxide [
95
, 96
]. On the contrary, it has also been suggested that NK cells may suppress GVH reactions and contribute to GVL effects [97
, 98
]. Besides TNF-α–mediated cytotoxicity, CTLs and other effector cells like NK cells can mediate their cytotoxicity through two other important contact-dependent pathways: Perforin-granzyme B–mediated cytolysis and Fas-Fas ligand (FasL)–mediated apoptosis [99
, 100
]. After direct contact and binding of perforin to target cell, the penetration of cell granule contacts (i.e., granzyme A and B) occurs, which leads to activation of caspase cascade, and thus cytolysis takes place [23
, 24
, 99
]. The roles of Fas and FasL have been well defined in initiation of apoptosis through caspase system; the significance of death receptors has also been explicated in DNA fragmentation and apoptosis [101
]. Though perforin- and FasL-mediated pathways are important in effective cell-mediated cytotoxicity, they are not the sole effector mechanisms involved in GVHD. Mice that receive FasL (CD95L)-defective T cells develop significant GVHD-associated cachexia but only minimal GVHD-associated changes in liver or skin [102
]. Anti-FasL antibody can protect from hepatic damage as demonstrated in a murine model [103
]. FasL-mediated cytotoxicity may be an essential pathway in hepatic GVHD. Mice that receive perforin-deficient T cells develop all signs of GVHD, but with a significant delay in the time of onset [102
]. These results suggest that perforin plays a role in the kinetics of GVHD and is possibly more critical in the affector phase than effector phase. Experimental murine models suggest that granzyme B–deficient CD8+ cells have significantly diminished GVHD induction capability compared to wild-type controls [104
, 105
]. Recent investigations keep on elucidating the intricate molecular pathways and signals between cytotoxic and target cells in this effector phase of GVHD.Regulatory cells
Double negative T cells (CD3+CD4−CD8−), which are usually NK1.1+, make up an unusually high proportion of the bone marrow T cell population. NK1.1+ T cells can suppress the MLR and can prevent GVHD in vivo [
106
, 107
, 108
]. Presumably these regulatory cells arise following sensitization as a method of controlling the overall response of activated T cells. The marrow NK1.1+ T cells secrete high levels of both IFN-γ and IL-4, and marrow NK1.1− T cells secrete high levels of IFN-γ with little IL-4 [106
]. The balance between suppressor NK1.1+ T cells (usually CD3+CD4−CD8−) and reactive NK1.1− T cells (usually CD3+CD4+CD8+) may regulate the reduction or induction of GVHD [106
, 109
].Clinical aspects of acute GVHD
GVHD is clinically divided as acute and chronic GVHD based on the time of onset, distinct pathobiological pathways, and different clinical presentations. Historically, GVHD occurring within the first 100 days following allogeneic BMT is called acute GVHD. Chronic GVHD is defined as GVHD that occurs after 100 days post-BMT, though this time distinction is not always so clear-cut.
Hyperacute GVHD is a severe fulminant form of acute GVHD that is frequently fatal but fortunately is rare in the era of GVHD prophylaxis [
110
]. It occurs in the first week post-BMT and is characterized by fever, generalized erythroderma, desquamation, severe hepatitis, widespread inflammation, and vascular leakage [110
].Acute GVHD is a clinicopathological syndrome involving mostly three organ systems the skin, the gastrointestinal tract, and the liver. Any one organ or combination of these organs may be affected. Acute GVHD occurs within the first 100 days, usually between 2 and 6 weeks following allogeneic BMT. Acute GVH reaction is directed against many cells of the host including epithelial cells of skin and mucosa, hair follicle cells, bile ducts of liver, crypt cells of intestinal tract, airways, mucous membranes, bone marrow, and immune system [
3
, 7
, 111
, 112
].Clinically significant acute GVHD, defined as grade II–IV and higher, occurs in 9–50% of patients who receive an allogeneic HLA-matched BMT, even when intensive prophylaxis with immunosuppressive drugs such as methotrexate, cyclosporine, and corticosteroids is used [
3
, 111
]. Risk of clinically significant GVHD can be 100% if no prophylaxis has been used [110
, 113
]. The incidence of acute GVHD varies with the degree of histoincompatibility, recipient age, the source and number of infused donor T lymphocytes, GVHD prophylaxis strategy, and to a lesser degree with other factors [3
, 7
, 114
]. The incidence of clinically significant GVHD may be greater than 70% in unrelated HLA-matched allogeneic BMT, and even as high as 80–90% in HLA-haploidentical (MHC-mismatched) transplantation [115
, 116
]. In related HLA-nonidentical allogeneic BMT, the risk of grade II–IV GVHD are 75%, 78%, and 80% for one-, two-, or three-HLA locus mismatches, respectively [37
, 117
, 118
].Multiple factors play a role in the induction of GVHD. Besides the utmost importance of histoincompatibility, other risk factors include conditioning regimen, dose of TBI, type of acute GVHD prophylaxis given, patient and donor age, underlying primary disease, state of primary alloimmunization (multiple transfusion), gender (such as female multiparous donors for male recipients), prior splenectomy, viral infection, and source of stem cells and graft composition [
3
, 31
, 111
, 119
, 120
, 121
, 122
, 123
, 124
, 125
, 126
, 127
, 128
]. Risk factors of acute GVHD are summarized in Table 1.Table 1Etiology and risk factors of acute GVHD
| Risk Factor | Background | Findings and Comments |
|---|---|---|
| Histocompatibility | Donor-recipient genetic disparity in major histocompatibility complex (MHC) and/or minor histocompatibility antigens (miH) increases the risk of both graft rejection and GVHD. | • miH antigen differences mainly contribute to GVHD in HLA-identical sibling transplants.40 • HLA-identical siblings exhibit less grade II–IV GVHD (approximately 20–30%) than HLA-identical unrelated donors (may be as high as 80%).3–6,119,120 • Molecular typing of HLA alleles is critical in the reduction of the incidence of GVHD in unrelated donor transplantation.121,122 |
| Conditioning regimens with either TBI or high-dose chemotherapy | The intensity of myeloablative conditioning chemotherapy and the dose of radiation contribute to the pathogenesis of GVHD. | • Myeloablative/cytoreductive conditioning approach can cause extensive damage and activation in host tissues.23,24 • Activated cells from damaged recipient tissues release inflammatory cytokines and enhance recipient MHC antigens. • Alloreactive donor T cells recognize and attach to histoincompatible host tissue antigens. |
| Microenvironment | Sterile microenvironment including gut decontamination is essential for the prevention of GVHD. | • Intestinal anaerobic bacterial microflora contribute to the pathogenesis of acute GVHD. • Augmented release of cytokine macrophages stimulated by bacterial breakdown products, such as lipopolysaccharide, or cross-reactions of donor T cells with bacterial antigens may facilitate the allorecognition process. • Antimicrobial chemotherapy targeted to intestinal anaerobic bacteria in recipients significantly reduces the severity of acute GVHD.88 |
| Patient and donor age | Advanced recipient and donor age represent a significant risk factor for the development of acute GVHD. | • Recipients aged less than 20 years old exhibit less grade III–IV acute GVHD than patients aged 51 to 62 years.123,124 |
| Donor: recipient gender | Gender mismatching may increase the risk of GVHD. | • Gender mismatching, especially female multiparous donor and male recipient, has an increased risk of acute GVHD (may reach approximately 68%).31,123 • Allosensitization to the putative H-Y male antigen through pregnancy may be the possible mechanism. |
| Source of stem cells and graft cell composition | • Allogeneic unrelated umbilical cord blood transplantation appears to result in a lower rate of GVHD than what would be anticipated in an unrelated donor or the same-degree related HLA-mismatched allogeneic BMT.125 | |
| • There is no apparent increase in the rates of acute GVHD in HLA-identical sibling alloPBSCT compared to alloBMT. However, there is a concern about possible tendency towards increased risk of chronic GVHD development with allogeneic PBSCT.68,69,126 | ||
| OTHER RISK FACTORS3,9,123,127,128 | ||
| State of primary donor alloimmunization | Multitransfusions significantly affect the occurrence of GVHD. | |
| Underlying primary disease status | Patients with leukemia, particularly chronic myelogenous leukemia, are at increased risk for GVHD. | |
| Viral infection | Positive donor and recipient cytomegalovirus serologies increase the risk for GVHD. | |
| Prior splenectomy | Splenectomy before transplantation may increase GVHD risk. | |
| Type of acute GVHD prophylaxis | The dose of methotrexate and cyclosporin delivered affects the development of GVHD. | |
Clinical presentations of acute GVHD
Skin
The first and the most common clinical manifestation of acute GVHD is often a pruritic maculopapular skin rash. The rash usually occurs at or near the time of white blood cell engraftment [
112
]. In the early stages of acute cutaneous GVHD, involvement of the nape of the neck, ears, and the shoulders, as well as the palms of the hands and the soles of the feet, can be seen and may look like a “sunburn.” Mild skin GVHD may resolve spontaneously or may leave postinflammatory dyspigmentation in response to therapy. In severe GVHD, the skin lesions may progress to generalized erythroderma, bullae formation, or desquamation and may even evolve to epidermal necrolysis. The progression of skin GVHD can be clinically defined into four stages depending on the extent of involvement of the skin [3
, 9
, 112
] (Table 2).Table 2Acute GVHD staging by the affected organ systems
| Organ | Grade 1 | Grade 2 | Grade 3 | Grade 4 |
|---|---|---|---|---|
| Skin | Rash over <25% of body area | Rash over 25–50% of body area | Generalized erythroderma | Generalized erythroderma with bullous formation |
| Liver | Bilirubin 2–3 mg/dL | Bilirubin 3.1–6 mg/dL | Bilirubin 6.1–15 mg/dL | Bilirubin >15 mg/dL |
| Gut | Diarrhea >500 mL/day | Diarrhea >1000 mL/day | Diarrhea >1500 mL/day | Diarrhea >2000 mL/day or severe abdominal pain with or without ileus |
The clinical diagnosis of acute skin GVHD may be complicated by other precipitating factors. Chemotherapy-induced and/or drug-related skin eruptions and viral exanthems should be considered in the differential diagnosis of skin GVHD [
112
].Histopathologic examination is essential for acute cutaneous GVHD, although the histologic differential diagnosis requires clinicopathologic correlation [
3
, 129
]. Dyskeratotic epidermal keratinocytes, exocytosed lymphocytes, Langerhans cell depletion, follicular involvement, satellite lymphocytes adjacent to or surrounding dyskeratotic epidermal keratinocytes vacuolar degeneration of the basal cell layer, intercellular edema, basal cell necrosis, acantholysis, and epidermolysis are all the histological features of skin GVHD. Unfortunately, the histology is not always pathognomonic [9
]. One possible marker for GVHD is an increase in the expression of HLA-DR in the keratinocytes [130
]. Studies of acute skin GVHD demonstrated that the infiltrating dermal lymphocytes are almost entirely T cells with a relative predominance of CD4+ cells, whereas epidermal lymphocytes exhibit a fairly striking predominance of CD8+ cells [131
]. Further characterization of skin infiltrating T lymphocytes during acute GVHD from the skin biopsy of a patient undergoing acute GVHD after an one-antigen HLA-mismatched related allogeneic BMT was demonstrated to be almost all CD4+ phenotype [132
]. NK cells may also be secondarily attracted to the skin by antigen-specific T cells [95
].Liver
The liver is the second most commonly affected organ in acute GVHD [
134
, 135
]. Infrequently, patients without skin involvement may have moderate to severe liver GVHD. The earliest manifestation of liver GVHD is jaundice with conjugated hyperbilirubinemia and an elevated alkaline phosphatase concentration. This is a reflection of damage to the bile canaliculi leading to cholestasis. Other diseases may cause abnormal liver function tests. These may include hepatic veno-occlusive disease (secondary to high-dose therapy), infection (primarily viral), sepsis, and drug toxicity used for GVHD prophylaxis (cyclosporin, methotrexate, or FK 506) [3
, 111
, 135
]. Aminotransferase levels may remain normal. Histopathological findings include lymphocytic infiltration and damage of the small bile ducts with nuclear polymorphism and epithelial cell drop-out. Cholestatic jaundice is a common feature, but hepatic failure with encephalopathy is unusual unless the GVHD is long-standing. Liver biopsy may be necessary to diagnose hepatic GVHD from other conditions leading to hepatic damage. However, the risk of acute hemorrhage is a big concern for the biopsy process following BMT [3
]. Transjugular approaches may be preferred to percutaneous liver biopsies, since the bleeding risk is lower if performed by an experienced operator.Gastrointestinal tract
The gut is the third organ system involved in GVHD. Gut GVHD is frequently the most severe and difficult to treat. Gut GVHD may involve any location throughout the gastrointestinal (GI) tract. Diarrhea and abdominal cramping are generally the hallmarks of gut involvement. Clinical manifestations include nausea, vomiting, crampy abdominal pain, distention, paralytic ileus, intestinal bleeding, and voluminous, often bloody, diarrhea [
133
, 135
]. Enteric cultures are needed to rule out infection. Voluminous secretory diarrhea may persist even with cessation of oral intake, and can be in excess of 10 liters per day. Initially, the diarrhea may be watery secondary to salt and water reabsorption defect, but often becomes bloody with increasing transfusion requirements. It may be difficult to maintain an adequate fluid balance and to treat severe abdominal pain [3
, 133
, 135
].Within the first weeks following BMT, the diarrhea may be related to the conditioning regimen. Antibiotics and superinfections such as Clostridium difficile should be considered in the differential diagnosis of gut GVHD. Since the symptoms are nonspecific, endoscopic biopsy confirmation is often needed. Endoscopic findings of gut GVHD may range from normal to extensive edema, mucosal sloughing, and diffuse bleeding throughout the whole GI tract. Lesions may be most prominent in the cecum, ileum, and ascending colon, but also commonly involve the stomach, duodenum, and rectum [
3
, 133
]. Histology reveals crypt-cell necrosis and drop-outs [137
]. Immunostaining for several adhesion molecules, such as VCAM-1 expression in enterocytes, might be useful for the diagnosis of gut GVHD [138
]. Nitric oxide (NO) was implicated in the intestinal pathology of GVHD in one experimental study [139
].Acute upper GI GVHD was described as a distinct entity initially in older patients. This form of acute gut GVHD presents clinically with anorexia, dyspepsia, food intolerance, nausea, and vomiting and can be recognized by upper GI tract endoscopy and biopsies [
136
, 140
]. Acute upper GI GVHD seems to be more responsive to immunosuppressive treatment [136
].Hematolymphoid organs
GVHD can also involve the hematolymphoid organs as described in early animal studies on GVHD. Host's immune competency may be affected [
3
]. In lymph nodes, there is diminution of germinal centers lasting for many months after BMT. Abnormal CD4/CD8 ratios can be found in both the circulating blood and lymph nodes [141
]. This compromised immune state may lead to frequent and serious infectious complications. GVHD may also affect hematopoiesis, and it may cause a reduction in peripheral blood counts, particularly thrombocytes. This has been observed in murine models [142
] and, clinically, it is occasionally observed in patients who receive donor lymphocyte infusion. The mechanism is probably similar to the myelosuppression of the well-described transfusion-associated GVHD [143
].Grading of acute GVHD
The stage of each organ involved is combined for the clinical staging and grading of acute GVHD. The overall clinical grade of acute GVHD has a major impact on survival after BMT and is used to assess response to the prophylaxis or treatment. GVHD is graded from I to IV according to severity (Table 2, Table 3). One difficulty in grading is the concordance for the diagnosis and grading between the physicians [
144
]. This underscores the importance of further studies and revisions of the scoring criteria as new data accumulate.Table 3Overall grading of acute GVHD
| Overall grade | Skin | Liver | Gut | ECOG performance | |
|---|---|---|---|---|---|
| I (very limited) | +1 to +2 | 0 | 0 | 0 | |
| II (moderately severe) | +1 to +3 | +1 | and/or | +1 | 0–1 |
| III (severe multiorgan) | +2 to +3 | +2 to +4 | and/or | +2 to +3 | 2–3 |
| IV (life-threatening) | +2 to +4 | +2 to +4 | and/or | +2 to +4 | 3–4 |
| +1 | 15 mg/m2 | IV | |||
| +3 | 10 mg/m2 | IV | |||
| +6 | 10 mg/m2 | IV | |||
| +11 | 10 mg/m2 | IV | |||
| <1.5 | 100 | 100 | |||
| 1.5–1.7 | 75 | 75 | |||
| 1.8–2.0 | 50 | 50 | |||
| >2.0 | Hold dose | Hold dose | |||
| <2.0 | 100 | ||||
| 2.1–3.0 | 50 | ||||
| 3.1–5.0 | 25 | ||||
| >5.0 | Hold dose | ||||
| IV = intravenous; QD = once daily; PO = oral; BID = twice daily. | |||||
Prophylaxis of acute GVHD
The first approach for the prevention of GVHD is minimizing risk factors whenever possible. The incidence of acute GVHD can approach 100% without prophylaxis [
110
]. Therefore, GVHD prophylaxis is essential in all patients undergoing allogeneic BMT with the present approaches [3
].Combination drug prophylaxis for acute GVHD
Pharmacologic prophylaxis of GVHD with methotrexate (MTX) was adopted early in human BMT protocols after experimental evidence in dogs indicated that it decreased the incidence and severity of GVHD [
1
]. Different combinations of methotrexate, cyclosporine (CsA), FK 506, and glucocorticoids have been widely used to prevent acute GVHD. GVHD pharmacologic prophylaxis is commonly administered in the immediate posttransplant period, gradually tapering off after 100 days and stopping around day 180 [3
]. Table 4 depicts “standard” two-drug regimen (CsA/MTX) prophylaxis for acute GVHD.Table 4“Standard” GVHD prevention with cyclosporin/methotrexate regimen
| Day of transplantation | Cyclosporin-A dosage | Cyclosporin-A route |
|---|---|---|
| −2 to +3 | 5 mg/kg | IV QD by infusion over 20 hours |
| +4 to +14 | 3 mg/kg | IV QD by infusion over 20 hours |
| +15 to +35 | 3.75 mg/kg | IV QD by infusion over 20 hours |
| +36 to +83 | 5 mg/kg | po BID |
| +84 to +97 | 4 mg/kg | po BID |
| +98 to +119 | 3 mg/kg | po BID |
| +120 to +180 | 2 mg/kg | po BID |
Immunosuppression with pharmacologic agents such as CsA, MTX, and corticosteroids are more effective when used in combination than as single agents [
145
, 146
, 147
]. Survival advantage for those patients who received the combination of CsA and MTX compared to either one of those drugs utilized alone has been demonstrated [148
, - Ringden O.
- Klaesson S.
- Sundberg B.
- Ljungman P.
- Lonnqvist B.
- Persson U.
Decreased incidence of graft-vs-host disease and improved survival with methotrexate combined with cyclosporin compared with monotherapy in recipients of bone marrow from donors other than HLA identical siblings.
Bone Marrow Transplant. 1992; 9: 19
149
]. Some major randomized trials in GVHD prophylaxis [6
, 150
, 151
, 152
, 153
, 154
, 155
, 156
, 157
, 158
, 159
] have been summarized in Table 5.Table 5Combination drug prophylaxis for acute GVHD
| Published trial | Diseases | Drug prophylaxis (patient numbers) | Incidence of grade II–IV Acute GVHD (p-value) |
|---|---|---|---|
| Ramsay150, 1982 | Aplastic anemia and hematologic malignancy | MTX+ATG+Pred. (32) vs MTX (35) | 21% vs 48% (p = .01) |
| Storb151, 1986 | Hematologic malignancy | MTX+CsA (43) vs CsA (50) | 33% vs 54% (p = .0014) |
| Santos152, 1987 | Nonmalignant and malignant disorders | CsA+MetPred. (42) vs CTX + MetPred. (40) | 32% vs 68% (p = .005) |
| Forman153, 1987 | Leukemia | MTX+Pred. (53) vs CsA+Pred. (54) | 47% vs 28% (p = .05) |
| Storb154, 1989 | Aplastic anemia | MTX+CsA (22) vs MTX (24) | 18% vs 53% (p = .01) |
| Sullivan155, 1989 | Hematologic malignancy | Long MTX (44) vs short MTX (40) vs Long MTX+DBC (25) | 25% vs 59% vs 82% |
| Storb156, 1990 | Nonmalignant and malignant disorders | MTX+CsA+Pred. (59) vs MTX+CsA (63) | 46% vs 25% (p = .02) |
| Chao6, 1993 | Hematologic malignancy | CsA+Pred. (74) vs MTX+CsA+Pred. (75) | 23% vs 9% (p = .02) |
| Deeg157, 1997 | Hematologic malignancy | CsA (60) vs CsA+MetPred. (62) | 73% vs 60% (p = .01) |
| Ratanatharathorn158, 1998 | Hematologic malignancy | MTX+FK506 (165) vs MTX+CsA (164) | 31.9% vs 44.4% (p = .01) |
| Chao159, 1999 | Leukemia | CsA+MTX+Pred. (90) vs CsA+MTX (96) | 20% vs 18% (p = NS) |
| Abbreviations: ATG = antithymocyte globulin; DBC = donor buffy-coat cells; CML = chronic myeloid leukemia; CsA = cyclosporine; CTX = cyclophosphamide; MetPred. = methylprednisolone; MTX = methotrexate; Pred. = prednisone; NS = not significant. | |||
The first randomized trial that proved the superiority of CsA + MTX to CsA alone, included 93 patients with acute myelogenous leukemia (AML) in first complete remission, or chronic myeloid leukemia (CML) in chronic phase [
151
]. The long-term follow-up reports of the trials also supported the initial results [160
, 161
]. In another study, CsA alone, T-cell depletion, and the combination of CsA and MTX were compared for GVHD prophylaxis in 140 patients with CML. The results suggested that the combination of CsA and MTX is the best among the three options for GVHD prophylaxis with an improved disease-free survival [162
]. Previous studies demonstrated a decrease in the incidence of acute GVHD with the addition of MTX to CsA and prednisone (PSE) in patients with leukemia [6
]. A prospective randomized trial comparing the three-drug regimen (CsA/MTX[three doses]/PSE) to the “standard” two-drug regimen (CsA/MTX [four doses]) was performed in order to investigate the benefit of PSE used up front for the prevention of acute and chronic GVHD. A total of 193 patients were randomized and 186 patients were included in the analysis (5 were not evaluable due to death before engraftment, 2 were deemed ineligible). All patients received bone marrow from a fully histocompatible sibling donor. The preparatory regimen consisted of fractionated total body irradiation (FTBI) and etoposide in all but 13 patients, who received FTBI/cyclophosphamide. The patients were randomized to receive either CsA/MTX/PSE or CsA/MTX. The two groups were well balanced with respect to diagnosis, age, donor-recipient gender match, and parity. In an intent-to-treat analysis, the incidence of acute GVHD was 18% (confidence interval [CI]: 12–28) for CsA/MTX/PSE compared to 20% (CI: 10–26) for CsA/MTX (p = 0.60). Overall survival was 65% for those receiving CsA/MTX/PSE and 72% for CsA/MTX (p = 0.10) with a relapse rate of 15% for the CsA/MTX/PSE group and 12% for the CsA/MTX group (p = 0.83). The incidence of chronic GVHD was similar (46% vs 52%, p = 0.38), with a follow-up of 0.7–6 years. Of interest, 19 patients went off study due to GVHD, 4 in the group receiving CsA/MTX/PSE and 15 of those receiving CsA/MTX (p = 0.02) and 11 patients went off study because of alveolar hemorrhage, 3 in the CsA/MTX/PSE arm and 8 in the CsA/MTX arm (p = 0.22). The addition of PSE did not result in a higher incidence of infectious complications: bacterial (66% vs 58%), viral (77% vs 66%), and fungal (20% vs 20%) in those receiving CsA/MTX/PSE or CsA/MTX, respectively. These data suggest that the addition of PSE is associated with a somewhat lower incidence of early posttransplant complications but did not have a positive impact in overall incidence of acute or chronic GVHD, event-free or overall survival [159
].Tacrolimus (FK506), a macrolide antibiotic, though it is structurally unique, has a similar mechanism of action as CsA. It exerts its activity through binding to FK-binding protein. It has been found to be equally as effective as CsA for the prevention of rejection of cell or organ allografts and is potentially less toxic [
163
, 164
, 165
]. Clinical studies have evaluated a FK506 + MTX combination, and the results of these trials suggest that this combination is effective in the prevention of GVHD. There have been two randomized studies comparing FK506 to CsA [166
, 167
]. Both studies found that FK506 appeared to be better for the prophylaxis of acute GVHD, but without a survival advantage. In a recent phase III randomized multicenter trial, the role of tacrolimus/methotrexate vs cyclosporine/methotrexate for GVHD prophylaxis after HLA-identical sibling marrow transplantation in patients with hematologic malignancy was tested [158
]. The incidence of grade II–IV acute GVHD was found to be significantly lower in patients who received tacrolimus compared to patients in the cyclosporine group (31.9% vs 44.4%, p = 0.01), though grade III–IV acute GVHD (13.3%. vs 17.1%) and chronic GVHD between groups (55.9% vs 49.4%, p = 0.8) was similar and relapse rates were also not different. The patients who received cyclosporine had a significantly better survival than patients who received the tacrolimus (57.2% vs 46.9%). A second approach using FK506 has been following donor pretreatment with FK506 [168
]. A single dose of FK506 given to the allogeneic bone marrow donor can significantly prolong the mean GVHD-free interval after allogeneic BMT.Serum immunoglobin (Ig) levels decrease significantly in the following months post-BMT. Serum IgA levels may remain low for up to two years. Other studies have reported subclass deficiencies of IgG2 and IgG4 [
169
, 170
]. In a large randomized study by Sullivan et al., there was a reduction in the incidence of gram-negative septicemia and local infections in patients receiving intravenous Ig as well as a significant decrease in the incidence of acute GVHD in patients aged 20 years and older who received intravenous Ig compared to the patients who did not receive intravenous Ig [171
]. These data suggested that the role of intravenous IG resulted in an immunomodulatory effect specifically in preventing GVHD.Marrow T-cell depletion for GVHD prevention
One attractive method to prevent GVHD is to eliminate donor T lymphocytes. T-cell depletion (TCD) is an effective technique in preventing acute GVHD in murine models. TCD can be performed by physical separation techniques, such as density gradients, selective depletion with lectins, treatment with cytotoxic drugs, or the use of anti–T cell serum or monoclonal antibodies (either alone, with complement, or conjugated to toxins). TCD reduces the incidence of GVHD but has potential adverse effects. TCD adversely affects engraftment due to graft rejection by residual recipient T cells that survived the conditioning regimen. TCD also increases leukemic relapse, infections, and secondary malignancies. The result was no improvement in overall survival [
3
, 172
, 173
, 174
].Among the strategies for TCD, the most common method is to use ex vivo treatment of the donor bone marrow with monoclonal antibodies. Broadly reactive anti–T cell agents, such as anti-CD2, anti-CD3, and anti-CD5, as well as more restricted reactivity such as anti-CD8 monoclonal antibodies, have been used [
175
, 176
, 177
]. Moreover, a broadly reactive human monoclonal antibody against lymphoid tissues, Campath-1, has been utilized [178
]. TCD with those antibodies is primarily indicated for patients who are at high risk for GVHD, because of the above mentioned serious adverse effects. According to an analysis from the International Bone Marrow Transplant Registry (IBMTR), CML patients in chronic phase had a relapse rate of approximately 12% following unmodified BMTs. This risk is increased to 50% in patients who received T cell–depleted transplants [179
]. In a recent intriguing study of 70 patients who received HLA-matched alloBMT, Campath-1G in vivo administration and Campath-1M ex vivo T cell depletion with no posttransplant prophylaxis was compared to historical IBMTR data, and acute GVHD was 4% in the Campath group vs 35% in the CsA + MTX group. Leukemia relapse at 5 years was 30% vs 29% [180
]. Several studies [181
, 182
, 183
, 184
, 185
, 186
, 187
, 188
] with TCD are summarized in Table 6.- Drobyski W.R.
- Hessner M.J.
- Klein J.P.
- Kabler-Babbitt C.
- Vesole D.H.
- Keever-Taylor C.A.
T-cell depletion plus salvage immunotherapy with donor leukocyte infusions as a strategy to treat chronic-phase chronic myelogenous leukemia patients undergoing HLA-identical sibling marrow transplantation.
Blood. 1999; 94: 434
Table 6Studies of T-cell depletion for the prevention of graft-versus-host disease (GVHD)
| Antibody | n | Other GVHD prophylaxis | GVHD incidence | Graft failure | Relapse risk | Ref. |
|---|---|---|---|---|---|---|
| 8 monoclonals | 20 | CsA, MTX | 15% | 15% | 35% | 181 |
| Anti-CD2 | 20 | CsA | 15% | 25% | 35% | 182 |
| Anti-CD2, 5, 7, or Anti-CD4, 5, 8 | 58 | none | 5% | 19% | 24% | 183 |
| Anti-CD8 | 36 | CsA | 28% | 11% | 8% | 184 |
| Campath-1 | 282 | — | 12% | 15% | 185 | |
| Anti-CD6 | 112 | none | 18% | 2.7% | — | 186 |
| SBA/E-rosette | 31 | none | 9.6% | 16% | 187 | |
| T10B9/Complement | 25 | CsA | 8% | 0 | 49% (OAS = 80% with DLI) | 188 |
| Campath1M–In vitro | 70 | None | 4% | 6% | 30% | 180 |
| Campath1G–In vitro | vs | vs | vs | vs | vs | |
| vs | 459 | CsA+MTX | 35% | 2% | 29% | |
| CsA+MTX (IBMTR) | ||||||
| Abbreviations: CsA = cyclosporine, MTX = methotrexate, DLI = donor lymphocyte infusion | ||||||
Newer agents for GVHD prevention
Newer pharmacologic agents and approaches that recently entered in clinical studies, such as rapamycin, MMF, trimetrexate, GLAT, and agents in development like PG27, hold some promise. Studies to develop pharmacologic agents that can block signaling pathways (for example, ZAP-70) may be of interest. Some newer agents [
189
, 190
, 191
, 192
, 193
, 194
, 195
, 196
, - Schlegel P.G.
- Aharoni R.
- Chen Y.
- et al.
A synthetic random basic copolymer with promiscuous binding to class II major histocompatibility complex molecules inhibits T-cell proliferative responses to major and minor histocompatibility antigens in vitro and confers the capacity to prevent murine graft-vs-host disease.
Proc Natl Acad Sci U S A. 1996; 93: 5061
197
] under investigation for the prevention of acute GVHD are listed in Table 7. The possible use of nucleoside analogs such as fludarabine, 2-CDA (2 deoxy-chloro-adenosine) is gaining some interest for prevention of GVHD, since antigen-activated T cells depend on the purine de novo synthesis. Fludarabine has already been tested in some clinical studies as part of the conditioning regimen [198
, 199
]. An experimental study of a murine GVHD model has demonstrated an improved survival in animals treated with fludarabine [200
].Table 7Newer agents for GVHD prevention
| Agent | Structure | Action | Setting |
|---|---|---|---|
| Mycophenolate Mofetil (MMF)193 | An ester product of mycophenolic acid produced by several species of penicillum molds | Exerts its immunosuppressive activity by inhibiting de novo synthesis of guanine nucleotides | Improvement in acute GVHD was found in 65% of patients treated with MMF in combination with CSP and prednisolone in a trial. |
| Rapamycin (Sirolimus)194 | A lipophilic macrolide produced by a strain of Streptomyces hygroscopicus | Binds to FK-binding proteins (FKBP) and inhibits the progression of cells from G1 into the S-phase | Inhibits both host-versus-graft and graft-versus-host reactions in a MHC-mismatched rat model. Phase I trials are currently underway. |
| Trimetrexate195,196 | 2,4-diaminoquinazoline folate analog | Similar action like MTX and metabolized by the liver | A clinical study in HLA-mismatched related alloBMT demonstrated that in combination with CSP, is tolerated well for prevention of GVHD. One major advantage over MTX is lack of nephrotoxicity. |
| Deoxyspergualin197 | A derivative of spergualin | Has immunosuppressive properties | Inhibits alloreactive cytotoxic activity in GVHD in experimental models. |
| Chloroquine198 | 4-aminoquinoline derivative | Blocks the transport of the peptides in conjunction with the MHC molecules to the cell surface | This approach was tested in a murine model and found to prevent GVHD. |
| PG27199 | An active fraction purified from an extract of a Chinese herb, Tripterygium wilfordii hook f | Has potent anti-inflammatory and immunosuppressive properties | In a murine model, treatment with PG27 prevented GVHD while inducing host-specific tolerance and retained the GVL effect. |
| CTLA4Ig and Monoclonal antibodies against CD40ligand61 | Humanized monoclonal antibodies | Inhibits costimulatory signals | A pilot study showed promising results in HLA-mismatched alloBMT. Clinical trials are currently underway. |
| GLAT200 | A large copolymer glutamic acid-lysine-alanine-tyrosine (GLAT) | Promiscuous binding to class II MHC intervenes TCR signal recognition | Significantly reduced GVHD in murine models. Phase I clinical trials are ongoing. |
| Neuraminidase201 | An enzyme | Impairs lymphocyte interactions through modifications of cell surface glycoproteins | Neuraminidase pretreatment of donor lymphocytes decreased acute GVHD incidence in murine models. |
Treatment of acute GVHD
Although the basic goal is to prevent the occurrence of GVHD following allogeneic BMT, this goal is not always achieved and therapy is required. Despite vigorous GVHD prophylaxis, 10–50% of the patients will still develop significant acute GVHD. Treatment of GVHD (grades II to IV) should then be aggressive, since survival correlates directly with the response.
Steroids have remained the standard for the treatment of acute GVHD [
3
, 201
]. The mechanism of action of steroids is presumably related to its lympholytic action. Methylprednisolone (MP) 2 mg/kg/day is generally the first choice. A prospective randomized trial was performed in patients with acute GVHD who were responding by day 14 of corticosteroid treatment. They received either a fast PSE taper over approximately 3 months (86 days) or a slow PSE taper over 5 to 6 months (147 days) [202
]. The median time to resolution of acute GVHD was 42 days (range 12–74 days) in the short-taper group vs a median of 30 days (range 6–30 days) for patients on the long-taper schedule (p = 0.01). There was no difference between groups in the number of patients with acute GVHD flare, or in the rates of infection or survival. Several studies analyzed the efficacy of this GVHD therapy. Among 469 patients receiving allogeneic BMT in Minneapolis, 179 (42%) developed grade II or greater acute GVHD. Seventy-two patients (41%) achieved complete and continued resolution of acute GVHD after a median of 21 days of therapy. Though most of the responders received corticosteroids as primary treatment, other immunosuppressive agents, including CsA, anti–T cell immunotoxin, or antilymphocyte globulin were also used. Chronic GVHD incidence was 70% among those who developed acute GVHD. Martin and colleagues analyzed 740 patients who developed grade II to IV GVHD [203
]. Primary treatment was with steroids in 532 patients, CsA in 170, and antithymocyte globulin (ATG) in 156. There was an improvement in the skin lesions in 43%; liver, 35%; and intestinal tract in 50% of patients—with an overall complete response of 18% and a partial response of 26% [204
]. A subsequent analysis included 427 patients who failed primary therapy [205
]. Most common manifestations of these patients were skin rash (75%) and liver dysfunction (59%). GI complications were the third most common, seen in 53% of the patients. Secondary treatment was glucocorticoids in 249 patients, CsA in 80, ATG in 214, or monoclonal antibody in 19. There was an improvement or resolution of GVHD in 45% of the patients with skin disease, 25% of patients with liver disease, and 35% of the patients with gut disease. Some response has been observed in 40% of patients. There is some suggestion that following an unrelated matched allogeneic BMT, acute GVHD may respond to a very high dose of steroids like 5 mg/kg/day for 4 days, escalating to 10 mg/kg/day for nonresponders [206
]. Unfortunately, increased risk of serious infections is a major complication after very high doses of steroids.If GVHD persists despite MP, additional immunosuppressive agents may be tried. ATG was studied as a single agent and 30–50% of patients responded. Although the results seemed to be inferior compared to steroids, it did not reach significance [
207
]. Another study investigated the usefulness of a combination of ATG, MP, and CsA [208
]. Approximately 60% of patients who received triple combination responded but, owing to increased risk of infections, survival was low in this group.Monoclonal antibodies for the treatment of acute GVHD have been increasingly used [
209
, 210
, 211
, 212
, 213
, 214
]. Table 8 depicts several trials of monoclonal antibody or receptor antagonist therapy for steroid-resistant acute GVHD. Daclizumab, a humanized monoclonal IgG1 directed against the α chain of the interleukin-2 receptor (IL-2R), is a competitive inhibitor of IL-2 on activated lymphocytes [214
]. This agent has been tested in a dose-finding study of 43 patients (14 patients received HLA-identical, 15 received HLA-mismatched related, and another 14 underwent unrelated donor allogeneic BMT) with advanced or steroid-refractory GVHD. The first cohort of 24 patients was treated with daclizumab IV 1 mg/kg on days 1, 8, 15, 22, and 29 and the complete response (CR) rate was 29%. A second cohort of 19 patients was treated with daclizumab 1 mg/kg on days 1, 4, 8, 15, and 22. The complete resolution of GVHD was 47% and survival on day 120 was 53% in this second cohort. Daclizumab has been tolerated well with no infusion-related reactions. There was a reduction in the serum concentrations of soluble IL-2R after treatment. OKT3, anti–IL-2 receptor monoclonal antibody (B-B10, CD25) [210
], monoclonal antibody against TNF-α [211
], Xomazyme [212
], and Anti-Ly1 monoclonal antibody conjugated to yttrium-90 [213
] have been studied with different successes in a variety of settings. Some new monoclonal antibodies hold promise in future use for the therapeutic intervention in GVHD [215
].Table 8Monoclonal antibody or receptor antagonist therapy for steroid-resistant acute GVHD
| Trial | Other GVHD therapy | Antibody | Description | Dose | Response |
|---|---|---|---|---|---|
| Blood 1990; 75:1426 | CSA+MP | H65-RTA (Zomazyme) | Anti-CD5 antibody labeled with ricin A chain | 0.05 mg/kg/day to 0.33 mg/kg/day IV for up to 14 consecutive days | 16–34 durable complete and partial responses |
| Blood 1990; 75:1426 | CSA+MP | B-C7 | Anti–TNF-α antibody | 0.1–0.4 mg/kg IV daily × 4 days then every other day × 2 | 74% partial response in 3 days, relapse in most when therapy stopped |
| Blood 1990; 75:1426 | CSA+MP | IL-1ra | IL-1 receptor antagonist | 400–3200 mg a day continuous IV infusion for 7 days | 10/16 improved |
| Transplant Int 1991; 4:3 | CSA+MP | 25.3 | Murine anti–LFA-1 (CD11a) antibody | 0.1 mg/kg IV over 4 hours daily × 5 days | 8/10 (80%) partial response |
| Blood 1990; 75:1426 | Cyc-A+Pred. | Humanized anti-Tac antibody | IL-2α receptor antibody | 0.5, 1.0, or 1.5 mg/kg IV over 1 hour single dose, repeated once between 11–48 days after first dose in responding patients | 4/20 complete response, 4/20 partial response |
| BMT 1994; 13:563 | TCD, CSA+MP | BT 563 (B-B10) | Murine anti-human IL-2α receptor antibody | 0.2 mg/kg IV over 30 min daily (mean 27 days, range 12–70 days) until GVHD < grade II for 48 hours | 11/15 complete remission, 2/15 partial remission, 6/13 relapsed |
| Blood 1990; 75:1426 | CSA+MP | BT 563 (B-B10) | Murine anti-human IL-2α receptor antibody | 5.0 mg IV bolus daily × 10 days then every other day for 10 days | 21/32 complete response, 6/32 partial response, 10/27 relapse |
| Blood 2000; 95:83 | CSA or Tacrolimus +MP | Daclizumab | Humanized monoclonal IgG1 against IL-2 receptor | 1.0 mg/kg IV infusion over 30 min on days 1, 4, 8, 15, 22 | 16/43 complete response (37%) with an overall response rate of 22/43 (51%) |
| TCD = T-cell depletion. | |||||
Future prospects
Allogeneic hematopoietic stem cell transplantation can be curative of a variety of malignant and nonmalignant conditions. Although significant improvements have been made, early transplant-related mortality and GVHD remain as major obstacles to safe transplantation. Recent, less damaging nonmyeloablative allogeneic stem cell transplantation approaches seem promising, with two possible advantages: first, decreasing the intensity of conditioning results in a lesser early transplant-related mortality, and second, decreasing the intense conditioning regimen, which clearly plays a role in graft-vs-host disease, may have further benefits in reducing GVHD. Purine nucleoside analogs such as fludarabine and cladribine are used in the recent nonmyeloablative preparative regimens and have shown some tendency toward a lower incidence of GVHD. These drugs also eliminate T cells responsible for alloreactions, and clinical studies evaluating their role in GVHD postallografting are intriguing. The value of this promising treatment modality needs to be validated in future randomized studies. As information from research and clinical studies in this field accumulates, our understanding of the biology of GVHD continues to evolve. In the future, we may have the tools to treat malignant diseases with allogeneic transplantation with specific GVL effects devoid of GVHD.
References
- Bone-marrow transplantation (second of two parts).N Engl J Med. 1975; 292: 895
- Bone marrow transplantation.N Engl J Med. 1994; 330: 827
Chao NJ (1999) Graft-vs-host disease (2nd edition). Austin: R.G. Landes Co.
- Treatment of chronic myeloid leukemia with allogeneic bone marrow transplantation after preparation with BuCy2.Blood. 1992; 80: 1352
- Pathogenesis of graft-vs-host disease.in: Atkinson K. Clinical bone marrow and blood stem cell transplantation. 2nd edition. Cambridge University Press, Cambridge2000 (147p)
- Cyclosporine, methotrexate, and prednisone compared with cyclosporine and prednisone for prophylaxis of acute graft-vs-host disease.N Engl J Med. 1993; 329: 225
- Graft-vs-host disease.N Engl J Med. 1991; 324: 667
- Long-term survival and late deaths after allogeneic bone marrow transplantation.N Engl J Med. 1999; 341: 14
- Graft vs host disease.in: Thomas E.D. Blume K.B. Forman S.J. Hematopoietic cell transplantation. 2nd edition. Blackwell Scientific Publications, Oxford1999 (515p)
- The biology of graft-vs-host reactions. In The Harvey lectures. Academic Press, New York1966 (62:21)
- Overview of marrow transplantation immunology.in: Thomas E.D. Blume K.B. Forman S.J. Hematopoietic cell transplantation. 2nd edition. Blackwell Scientific Publications, Oxford1999 (Oxford, 19)
- The HLA system in clinical marrow transplantation.Hematol Oncol Clin North Am. 1990; 4: 507
- Histocompatibility.in: Thomas E.D. Blume K.B. Forman S.J. Hematopoietic cell transplantation. 2nd edition. Blackwell Scientific Publications, Oxford1999 (28p)
- The major histocompatibility complex.N Engl J Med. 1976; 295: 872
- HLAs and risk of acute graft-vs.-host disease after marrow transplantation from an HLA-identical sibling.Biol Blood Marrow Transplant. 1998; 4: 128
- Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1.Nature. 1993; 364: 33
- Class II antigen expression by keratinocytes and enterocytes.Transplantation. 1988; 46: 402
- Induction of MHC class II gene products in rat intestinal epithelium during graft-vs-host disease and effects on the immune function of the epithelium.Immunology. 1992; 75: 366
- The minor histocompatibility antigen HA-1.Science. 1998; 279: 1054
- MHC-restricted cytotoxic T cells.Adv Immunol. 1979; 27: 51
- T-immunogenic peptides are constituted of rare sequence patterns. Use in the identification of T epitopes in the human immunodeficiency virus gag protein.Eur J Immunol. 1988; 18: 1547
- Implication of a gene distal to HLA-A in the etiology of graft-vs-host disease.Transplant Proc. 1989; 21: 2971
- Graft-vs-host disease.Biol Blood Marrow Transplant. 1997; 3: 1
- Pathophysiologic mechanisms of acute graft-vs.-host disease.Biol Blood Marrow Transplant. 1999; 5: 347
- Effect of total body irradiation, busulfan-cyclophosphamide, or cyclophosphamide conditioning on inflammatory cytokine release and development of acute and chronic graft-vs-host disease in H-2–incompatible transplanted SCID mice.Blood. 1994; 83: 2360
- ICAM-1 expression on epidermal keratinocytes in cutaneous graft-vs-host disease.Transplantation. 1991; 51: 1203
- Effects of tumor necrosis factor on the interferon-γ-induced major histocompatibility complex class II antigen expression by human endothelial cells.Eur J Immunol. 1988; 18: 1469
- Overlapping patterns of activation of human endothelial cells by interleukin-1, tumor necrosis factor, and immune interferon.J Immunol. 1986; 137: 1893
- The role of cytokines in acute graft-vs-host disease.Cytokines Cell Mol Ther. 1997; 3: 257
Clift RA, Buckner CD, Appelbaum FR, et al. (1990) Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: a randomized trial of two irradiation regimens. Blood 76:1867
- Risk factors for acute graft-vs-host disease.Br J Haematol. 1987; 67: 397
- Total body irradiation and acute graft-vs-host disease.Blood. 1997; 90: 3204
- Delayed infusion of immunocompetent donor cells after bone marrow transplantation breaks graft-host tolerance allows for persistent antileukemic reactivity without severe graft-vs-host disease.Blood. 1995; 85: 3302
- Delaying transplantation after total body irradiation is a simple and effective way to reduce acute graft-vs-host disease mortality after major H2 incompatible transplantation.Transplantation. 1997; 64: 297
- T cell–depleted bone marrow transplantation and delayed T cell add-back to control acute GVHD and conserve a graft-vs-leukemia effect.Bone Marrow Transplant. 1998; 21: 543
- Molecular recognition of antigen involves lattice formation between CD4, MHC class II and TCR molecules.Immunol Today. 1995; 16: 581
- Marrow transplantation from related donors other than HLA-identical siblings.N Engl J Med. 1985; 313: 765
- Risk factors for acute graft-vs-host disease in histocompatible donor bone marrow transplantation.Transplantation. 1991; 51: 1197
- Identification of a graft vs host disease–associated human minor histocompatibility antigen.Science. 1995; 268: 1476
Goulmy E, Schipper R, Pool J, et al. (1996) Mismatches of minor histocompatibility antigens between HLA-identical donors and recipients and the development of graft-vs-host disease after bone marrow transplantation N Engl J Med 334:281
- The cell adhesion molecule CD31 is phosphorylated after cell activation.Down-regulation of CD. 1992; 31: 5243
- Polymorphism of adhesion molecule CD31 and its role in acute graft-vs-host disease.N Engl J Med. 1996; 334: 286
- Evidence that CD31, CD49b, and CD62L are immunodominant minor histocompatibility antigens in HLA identical sibling bone marrow transplants.Blood. 1998; 92: 2169
- CD31/PECAM-1 is a ligand for α v β 3 integrin involved in adhesion of leukocytes to endothelium.J Cell Biol. 1995; 130: 451
- PECAM-1 (CD31) cloning and relation to adhesion molecules of the immunoglobulin gene superfamily.Science. 1990; 247: 1219
- Interaction of CD31 with a heterophilic counterreceptor involved in downregulation of human T cell responses.J Exp Med. 1996; 184: 41
- Polymorphism of adhesion molecule CD31 is not a significant risk factor for graft-vs-host disease.Blood. 1996; 88: 4429
- Coblockade of the LFA-1/ICAM and CD28/CTL4:B7 pathways is a highly effective means of preventing acute lethal graft-vs-host disease induced by fully MHC-disparate donor grafts.Blood. 1995; 85: 2607
- Kinetics of Th1 and Th2 cytokine production during the early course of acute and chronic murine graft-vs-host disease.J Immunol. 1995; 155: 2396
- Prevention of graft vs host disease by inactivation of host antigen-presenting cells.Science. 1999; 285: 412
- Transplantation of anergic histoincompatible bone marrow allografts.N Engl J Med. 1999; 340: 1704
- Kinetics of T-cell receptor binding to peptide/I-Ek complexes.Proc Natl Acad Sci U S A. 1994; 91: 12862
- Lymphocyte homing and homeostasis.Science. 1996; 272: 60
- Interaction of antigenic peptides with MHC and TCR molecules.Clin Immunol Immunopathol. 1995; 76: 68
- Molecular recognition of antigen involves lattice formation between CD4, MHC class II and TCR molecules.Immunol Today. 1995; 16: 581
Yang Y, Wilson JM (1996) CD40 ligand-dependent T cell activation: requirement of B7-CD28 signaling through CD40. Science 273:1862
- The B7 and CD28 receptor families.Immunol Today. 1994; 15: 321
- Absence of B7-dependent responses in CD28-deficient mice.Immunity. 1994; 1: 501
- Cloning of B7-2.Science. 1993; 262: 909
- Allogeneic hematopoietic chimerism in mice treated with sublethal myeloablation and anti-CD154 antibody.Blood. 2000; 95: 2175
- B7-mediated costimulation and the immune response.Blood Rev. 1996; 10: 111
- T cell subsets and graft-vs-host disease.Transplantation. 1987; 44: 335
- Subsets of T cells in the rat mediating lethal graft-vs-host disease.Transplantation. 1981; 32: 222
- Alloreactive murine CD8+ T cell clones secrete the Th1 pattern of cytokines.J Immunol. 1990; 144: 1744
- Interleukin-10 induces a long-term antigen-specific anergic state in human CD4+ T cells.J Exp Med. 1996; 184: 19
Grewal IS, Foellmer HG, Grewal KD, et al. (1996) Requirement for CD40 ligand in costimulation induction, T cell activation, and experimental allergic encephalomyelitis. Science 273:1864
- A prospective, randomized trial of peripheral blood stem cells or marrow for patients undergoing allogeneic transplantation for hematologic malignancies.Blood (suppl. 1999; 1): 1637a
- Randomized trial of bone marrow vs lenograstim-primed blood cell allogeneic transplantation in patients with early-stage leukemia.J Clin Oncol. 2000; 18: 537
- Allogeneic blood and bone-marrow stem-cell transplantation in haematological malignant diseases.Lancet. 2000; 355: 1231
- Donor-derived interferon-γ is required for inhibition of acute graft-vs-host disease by interleukin-12.J Clin Invest. 1998; 102: 1742
- Differential effects of the absence of interferon-γ and IL-4 in acute graft-vs-host disease after allogeneic bone marrow transplantation in mice.J Clin Invest. 1998; 102: 1742
- Interleukin-12 inhibits murine graft-vs-host disease.Blood. 1995; 86: 2429
- Critical role of interleukin-2 in the development of acute graft-vs-host disease.Int Immunol. 1993; 5: 565
- Host-specific interleukin-2–secreting donor T-cell precursors as predictors of acute graft-vs-host disease in bone marrow transplantation between HLA-identical siblings.N Engl J Med. 1992; 327: 1613
- Frequency of anti-recipient alloreactive helper T-cell precursors in donor blood and graft-vs-host disease after HLA-identical sibling bone-marrow transplantation.Lancet. 1993; 341: 203
- Serum concentration of the soluble interleukin-2 receptor for monitoring acute graft-vs-host disease.Bone Marrow Transplant. 1996; 17: 185
- Interleukin-1 and interleukin-1 antagonism.Blood. 1991; 77: 1627
- Cytokine dysregulation and acute graft-vs-host disease.Blood. 1994; 80: 2964
- Inhibition of graft-vs-host disease by interleukin-2 treatment is associated with altered cytokine production by expanded graft-vs-host-reactive CD4+ helper cells.Transplantation. 1995; 60: 481
- Dissociating graft-vs-host disease from the graft-vs-leukemia effect of allogeneic T cells.Bone Marrow Transplant (suppl. 1992; 1): 10
- The mechanism of IL-2–mediated protection against GVHD in mice. II. Protection occurs independently of NK/LAK cells.Transplantation. 1992; 53: 1063
- Soluble interleukin-6 receptors in hematology patients undergoing bone marrow transplantation.Transplantation. 1996; 62: 138
- Elevated interleukin-8 serum concentrations in β thalassemia and graft-vs-host disease.Blood. 1993; 81: 2252
- Interleukin-11 promotes T cell polarization and prevents acute graft-vs-host disease after allogeneic bone marrow transplantation.J Clin Invest. 1998; 102: 115
- Kinetics of interferon-γ secretion and its regulatory factors in the early phase of acute graft-vs-host disease.Immunology. 1999; 98: 379
- Mitigation of secondary disease of allogeneic mouse radiation chimeras by modification of the intestinal microflora.J Natl Cancer Inst. 1974; 52: 401
- Graft-vs-host disease and survival in patients with aplastic anemia treated by marrow grafts from HLA-identical siblings. Beneficial effect of a protective environment.N Engl J Med. 1983; 308: 302
- Influence of intestinal bacterial decontamination using metronidazole and ciprofloxacin or ciprofloxacin alone on the development of acute graft-vs-host disease after marrow transplantation in patients with hematologic malignancies.Blood. 1999; 93: 3267
- Prevention of the graft-vs-host reaction in newborn mice by antibodies to tumor necrosis factor–α.Transplantation. 1989; 47: 1057
- Increased serum levels of tumor necrosis factor–α precede major complications of bone marrow transplantation.Blood. 1990; 75: 1011
- Phase I-II trial of a monoclonal anti-tumor necrosis factor α antibody for the treatment of refractory severe acute graft-vs-host disease.Blood. 1992; 79: 3362
- Tumor necrosis factor/cachectin is an effector of skin and gut lesions of the acute phase of graft-vs.-host disease.J Exp Med. 1987; 166: 1280
- Keratinocyte growth factor administered before conditioning ameliorates graft-vs-host disease after allogeneic bone marrow transplantation in mice.Blood. 1998; 92: 3960
- Keratinocyte growth factor (KGF) separates graft-vs-leukemia effects from graft-vs-host disease.Blood. 1999; 94: 825
- Evidence that large granular lymphocytes of donor origin mediate acute graft-vs-host disease.Transplantation. 1989; 47: 50
- Involvement of nitric oxide in target-cell lysis and DNA fragmentation induced by murine natural killer cells.Blood. 1996; 87: 5136
- Relevance and potential of natural killer cells in stem cell transplantation.Biol Blood Marrow Transplant. 2000; 6: 90
- Suppression of graft-vs-host disease and amplification of graft-vs-tumor effects by activated natural killer cells after allogeneic bone marrow transplantation.J Clin Invest. 1998; 101: 1835
- Cytolytic T-cell cytotoxicity is mediated through perforin and Fas lytic pathways.Nature. 1994; 370: 650
- A major role for the Fas pathway in acute graft-vs-host disease.J Immunol. 1996; 157: 5387
- Death receptors.Science. 1998; 281: 1305
- The role of cell-mediated cytotoxicity in acute GVHD after MHC-matched allogeneic bone marrow transplantation in mice.J Exp Med. 1996; 183: 2645
- Differential effects of anti-Fas ligand and anti-tumor necrosis factor–α antibodies on acute graft-vs-host disease pathologies.Blood. 1998; 91: 4051
- The role of granzyme B in murine models of acute graft-vs-host disease and graft rejection.Blood. 1996; 87: 1232
- Perforin/granzyme-dependent and independent mechanisms are both important for the development of graft-vs-host disease after murine bone marrow transplantation.J Clin Invest. 1997; 100: 904
- Bone marrow NK1.1(−) and NK1.1(+) T cells reciprocally regulate acute graft-vs-host disease.J Exp Med. 1999; 189: 1073
- Role of natural suppressor cells in bone marrow transplantation.Transplant Proc. 1987; 19: 88
- Cloned natural suppressor cells prevent lethal graft-vs-host disease.J Immunol. 1987; 138: 699
- Unusual T cell populations in adult murine bone marrow.Prevalence of CD. 1990; 3+CD4−CD8−: 3209
- Hyperacute graft-vs-host disease in patients not given immunosuppression after allogeneic marrow transplantation.Blood. 1986; 67: 1172
- Pathophysiology and treatment of graft-vs-host disease.Hematol Oncol Clin North Am. 1999; 13: 1091
- Graft-vs-host reactions in dermatology.J Am Acad Dermatol. 1998; 38: 369
- Incidence of acute graft-vs-host disease with and without methotrexate prophylaxis in allogeneic bone marrow transplantation patients.Blood. 1984; 64: 215
- Unrelated donor hematopoietic cell transplantation.in: Thomas E.D. Blume K.B. Forman S.J. Hematopoietic cell transplantation. 2nd edition. Blackwell Scientific Publications, Oxford1999 (915p)
- Increased disparity for minor histocompatibility antigens as a potential cause of increased GVHD risk in marrow transplantation from unrelated donors c