Update on graft versus host disease.

Abstract

Exciting new advances have occurred in the field of hematopoietic transplantation biology in the last two decades. More powerful reagents and techniques for classification of HLA antigens to match donors and recipients, new conditioning regimens to prepare recipients for transplantation and new methods of transplantation (autologous transplantation, cord blood and stem cell transplantation) are in use. Also, entirely new concepts about graft versus malignancy reactions have come from data collected over years of experience with transplantation in humans and in valuable mouse models. However, graft versus host disease (GVHD) continues to be a major complication of hematopoietic cell transplantation, occurring in 30-80 percent of recipients. The concept of GVHD was first formulated by Rupert Billingham [[1]Billingham R.E. The biology of graft versus host disease.Harvey Lectures. 1966; 62: 21PubMed Google Scholar]. It occurs when "the graft contains immunologically competent cells, the recipient expresses tissue antigens that are not present in the transplant donor, and the recipient is incapable of mounting an effective response to destroy the transplanted cells" [[2]Ferrara J.L. Deeg H.J. Graft-versus-host disease.New England Journal of Medicine. 1991; 324: 667Crossref PubMed Scopus (1126) Google Scholar]. We know now that the mediators of GVHD are dendritic cells, T cells, NK cells, macrophages, cytokines and surface markers on immune cells (major and minor histocompatibility antigens, MHC and MiHC respectively) in the recipient. The recipient is usually immunocompromised either chemically by irradiation and chemotherapy or physiologically, as in neonates. Risks for developing GVHD include donor-recipient gender mismatch, HLA mismatch, older age of recipient, high numbers of T cells transfused from donor, host exposure to previous blood product transfusion, and low concentrations of immunosuppressants in recipient (Table I). The combination of immunocompromise due to conditioning regimen, slow recovery of the transplanted immune system and immunosuppression for GVHD all contribute to increased susceptibility of recipients to life-threatening infections and mortality due to bone marrow failure.Table IRisks for Graft versus Host Disease•Donor-recipient gender mismatch•Histocompatibility antigen mismatch (major and minor)•Older age of recipient•Host exposure to previous blood product transfusion•High numbers of T cells in the donor inoculum•Low concentrations of immunosuppressants in recipient Open table in a new tab This review focuses on emerging concepts of GVHD, the different manifestations of GVHD, the treatment of GVHD and new transplantation regimens. We will concentrate on cutaneous manifestations and therapies that dermatologists would encounter. Most importantly, new information gained through advances in transplantation biology has contributed to our rapidly increasing fund of knowledge about the immune system. Acute GVHD Clinical. Acute GVHD can occur within the first weeks after transplantation, with greater frequency and severity for HLA nonidentical or unrelated donors. The clinical manifestations include "dermatitis, hepatitis, and enteritis", which reflect injury to epithelia by activated immune cells. There is also injury to the immune system itself, leading to increased susceptibility to infection. Skin is an early target (Table II). GVHD usually begins on the palms, soles, ears, and oral mucosa. In acute GVHD, skin findings generally precede liver and gastrointestinal tract involvement. Tender erythematous macules are seen first. These are often folliculocentric, reflecting injury to hair follicle epithelium as well as epidermis. The macules may coalesce into confluent erythema, and in severe cases, subepidermal bullae can form with blistering. A fulminant form evolving into a lichen planus-like dermatitis can be seen. Acute GVHD is graded according to a system that includes percentage of total skin that is inflamed, amount of diarrhea per 24 hours, and serum level of bilirubin. The clinical differential diagnosis is drug eruption, viral exanthem, subacute radiation dermatitis, cutaneous reaction of lymphocyte recovery, and in severe cases, erythema multiforme and toxic epidermal necrolysis.Table IICutaneous and mucosal manifestationsof GVHDAcute GVHD•Macular erythema with pruritus first on palms, soles, ears and oral mucosa•May progress to confluent erythema with blistering if severe•Bacterial, viral and fungal infections•Oral: lacy patches in oral mucosaChronic GVHD•Lichenoid papules in flexures (lichenoid variant)•Sclerodermatous plaques and joint contractures (sclerodermoid variant)•Bacterial, viral and fungal infections•Hyperkeratosis•Alopecia•Pigmentary changes, mottling•Onychodystrophy•Oral: xerostomia, mucositis, ulceration, lichen planus-like changes Open table in a new tab Histopathology. Interface dermatitis with epidermal injury out of proportion to the inflammation is the key to cutaneous and mucosal GVHD (Figure 1A, B). Hair follicle and eccrine duct involvement and location of apoptotic cells in suprabasilar keratinocyte layer are also helpful in making the diagnosis of GVHD. Apoptosis is the mechanism of keratinocyte death in human and mouse GVHD[3Gilliam A.C. Whitaker-Menezes D. Korngold R. Murphy G.F. Apoptosis is the predominant form of epithelial target cell injury in acute experimental graft-versus-host disease.Journal of Investigative Dermatology. 1996; 107: 377Crossref PubMed Scopus (75) Google Scholar, 4Langley R.G. Walsh N. Nevill T. Thomas L. Rowden G. Apoptosis is the mode of keratinocyte death in cutaneous graft-versus-host disease.Journal of the American Academy of Dermatology. 1996; 35: 187Abstract Full Text PDF PubMed Google Scholar]. The histologic differential diagnosis is usually drug eruption. A definitive diagnosis of GVHD versus hypersensitivity reaction is controversial. Using the classic criteria (apoptotic keratinocytes in epidermis and appendages, basal cell vacuolization, satellitosis) investigators in a correlative study found that no single or combined feature above could predict clinical GVHD [[5]Kohler S. Hendrickson M.R. Chao N.J. Smoller B.R. Value of skin biopsies in assessing prognosis and progression of acute graft-versus-host disease.American Journal of Surgical Pathology. 1997; 21: 988Crossref PubMed Scopus (73) Google Scholar]. Nevertheless, dermatopathologists are still asked to make the call if possible because of the devastating consequences of not recognizing early GVHD. Treatment. The standard treatment for acute GVHD is prophylaxis with cyclosporin A and methotrexate, and methyl prednisone for approximately 2 weeks, with a glucocorticoid taper if GVHD manifestations are resolving. However, glucocorticoids fail to control GVHD in a large number of patients (up to 80% of transplanted patients in some early studies)[[6]Carpenter P.A. Sanders J.E. Steroid-refractory graft-vs.-host disease: past, present and future.Pediatric Transplantation 7 Suppl. 2003; 3: 19Crossref Scopus (30) Google Scholar]. This is an urgent problem which is addressed with a line of secondary therapies including polyclonal anti-thymus globulin (ATG), mega-dose glucocorticoids in the ranges of 5-20 milligrams/kg/day, macrolides such as tacrolimus and sirolimus, mycophenolic acid, and exciting new monoclonal antibodies to a variety of immune cell surface markers and to cytokine receptors[7Bruner R.J. Farag S.S. Monoclonal antibodies for the prevention and treatment of graft-versus-host disease.Seminars in Oncology. 2003; 30: 509Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar, 8Dean R.M. Bishop M.R. Graft-versus-host disease: emerging concepts in prevention and therapy.Current Hematology Reports. 2003; 2: 287PubMed Google Scholar, 9Davies J.K. Lowdell M.W. New advances in acute graft versus host disease prophylaxis.Transfusion Medicine. 2003; 13: 387Crossref PubMed Scopus (10) Google Scholar]. PUVA (psoralen plus UVA light) is effective for steroid-resistant acute GVHD of the skin[[10]Furlong T. Leisenring W. Storb R. Anasetti C. Appelbaum F.R. Carpenter P.A. Deeg H.J. Doney K. Kiem H.P. Nash R.A. Sanders J.E. Witherspoon R. Thompson D. Martin P.J. Psoralen and ultraviolet A irradiation (PUVA) as therapy for steroid-resistant cutaneous acute graft-versus-host disease.Biology of Blood & Marrow Transplantation. 2002; 8: 206Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar]. Extracorporeal photophere sis has also been used successfully[11Greinix H.T. Volc-Platzer B. Rabitsch W. Gmeinhart B. Guevara-Pineda C. Kalhs P. Krutmann J. Honigsmann H. Ciovica M. Knobler R.M. Successful use of extracorporeal photochemotherapy in the treatment of severe acute and chronic graft-versus-host disease.Blood. 1998; 92: 3098PubMed Google Scholar, 12Zic J.A. Miller J.L. Stricklin G.P. King Jr., L.E. The North American experience with photopheresis.Therapeutic Apheresis. 1999; 3: 50Crossref PubMed Scopus (40) Google Scholar]. Chronic GVHD Clinical. Chronic GVHD is distinguished from acute GVHD by clinical symptoms that can resemble an overlap of several connective tissue diseases (lupus erythematosus, mixed connective tissue disease, scleroderma, Sjogren syndrome, biliary cirrhosis, idiopathic pulmonary fibrosis). The classical definition of chronic GVHD is GVHD that persists or occurs de novo beyond 80-100 days post BMT. Chronic GVHD occurs in approximately 30 to 60% of patients, particularly when hematopoietic stem cells are used instead of bone marrow, and when the donor is not an HLA identical sibling[[6]Carpenter P.A. Sanders J.E. Steroid-refractory graft-vs.-host disease: past, present and future.Pediatric Transplantation 7 Suppl. 2003; 3: 19Crossref Scopus (30) Google Scholar]. In limited chronic GVHD, mild skin involvement, mild involvement of ocular, oral, or vaginal mucosa, and small changes in liver function tests are seen. Systemic therapy is usually not used for these individuals. In severe chronic GVHD, degree of epithelial involvement is more extensive, leading to serositis, hepatitis, biliary cirrhosis, severe gastrointestinal injury, and bronchiolitis obliterans. Individuals with chronic GVHD can develop alopecia and pigmentary changes, and are more susceptible to skin cancers, and to cutaneous viral and fungal infections (herpes, molluscum, candida, and opportunistic deep fungi) (Table II). Jaundice is a late occurrence. In chronic GVHD, lichenoid papules resembling lichen planus in flexural surfaces (lichenoid GVHD) and white mucosal patches occur. In some patients (approximately 15%) sclerodermatous plaques are present (sclerodermatous graft versus host disease), with esophageal dysmotility, skin fibrosis, and joint contractures[[13]Penas P.F. Jones-Caballero M. Aragues M. Fernandez-Herrera J. Fraga J. Garcia-Diez A. Sclerodermatous graft-vs-host disease: clinical and pathological study of 17 patients.Archives of Dermatology. 2002; 138: 924Crossref PubMed Google Scholar]. A polymyositis-like disorder with muscle weakness, pain, and necrotic muscle fibers on biopsy can occur in small numbers (less than 1%) of individuals with chronic GVHD[[14]Stevens A.M. Sullivan K.M. Nelson J.L. Polymyositis as a manifestation of chronic graft-versus-host disease.Rheumatology. 2003; 42: 34Crossref PubMed Scopus (78) Google Scholar]. Sicca syndrome is common. Individuals with chronic GVHD can develop autoantibodies to nuclear antigens similar to those in autoimmune diseases[[15]Rouquette-Gally A.M. Boyeldieu D. Prost A.C. Gluckman E. Autoimmunity after allogeneic bone marrow transplantation.A study of 53 long-term-surviving patients. 1988; 46 (Transplantation): 238Google Scholar]. When extensive skin involvement, thrombocytopenia and progressive type onset are present, non-relapse mortality is high[[16]Akpek G. Lee S.J. Flowers M.E. Pavletic S.Z. Arora M. Lee S. Piantadosi S. Guthrie K.A. Lynch J.C. Takatu A. Horowitz M.M. Antin J.H. Weisdorf D.J. Martin P.J. Vogelsang G.B. Performance of a new clinical grading system for chronic graft-versus-host disease: a multicenter study.Blood. 2003; 102: 802Crossref PubMed Scopus (122) Google Scholar]. Histology. The histology of chronic cutaneous GVHD can be lupus-like or scleroderma-like. In the lichenoid form, chronic-interface dermatitis leads to sawtoothing of epidermal rete and hyperkeratosis (Figure 1C, D). In the sclerodermoid form, dermal fibrosis begins in the papillary dermis and extends to subcutaneous fat. Ultimately sclerotic collagen and adnexal atrophy are seen (Figure 1E, F). Treatment. Treatment of severe chronic GVHD has historically been with early administration of long-term glucocorticoids and cyclosporin or tacrolimus. Many of the immunomodulatory agents mentioned above are also useful in chronic GVHD, including thalidomide and etretinate[[8]Dean R.M. Bishop M.R. Graft-versus-host disease: emerging concepts in prevention and therapy.Current Hematology Reports. 2003; 2: 287PubMed Google Scholar]. Of note are some interesting studies with extracorporeal photopheresis, in which a significant number of patients improve clinically[11Greinix H.T. Volc-Platzer B. Rabitsch W. Gmeinhart B. Guevara-Pineda C. Kalhs P. Krutmann J. Honigsmann H. Ciovica M. Knobler R.M. Successful use of extracorporeal photochemotherapy in the treatment of severe acute and chronic graft-versus-host disease.Blood. 1998; 92: 3098PubMed Google Scholar, 12Zic J.A. Miller J.L. Stricklin G.P. King Jr., L.E. The North American experience with photopheresis.Therapeutic Apheresis. 1999; 3: 50Crossref PubMed Scopus (40) Google Scholar]. Supportive care includes artificial tears for ocular dryness, pilocarpine for xerostomia, and nutritional supplements for severe mucositis. Anti-osteoporosis regimens are helpful to decrease bone loss from long term corticosteroid therapy. Overall, the best treatment for chronic GVHD is prevention. Transfusion associated GVHD. This is a rare and usually fatal complication of transfusion with blood products, including red blood cells, platelet concentrates, fresh plasma and granular cell concentrations. It results when donor and recipient share HLA haplotypes, and can occur when donations are obtained from first or second-degree relatives or from populations with a restricted pool of HLA haplotypes. Transfusion associated GVHD can also occur when the recipient has deficient cell mediated immunity, either inherited, as in severe combined immunodeficiency, or acquired, as in Hodgkin's disease. Clinical manifestations include a subtle skin rash, fever, diarrhea, hepatic dysfunction, and acute bone marrow failure approximately 1 to 2 weeks after transfusion. Transfusion associated GVHD is prevented by giving blood products that are previously treated with gamma irradiation. Treatment is usually not effective in severe disease[[17]Pamphilon D. Transfusion-associated graft versus host disease.2002http://www.blood.co.uk/hospitals/guidelines/pdf/GvHD.pdfGoogle Scholar]. Syngeneic GVHD. A rare and unexpected variant of graft versus host disease can occur with identical twin-twin transplantation, despite the identical genetic composition of donor and recipient. This is also the form of GVHD that occurs in autologous hematopoietic cell transplantation (transplantation with an individual's own hematopoietic cells). Syngeneic GVHD can be modeled in inbred strains of mice. It is thought that self-tolerance is lost in the setting of an irradiated thymus and cyclosporin or IFN-gamma administered during recovery of the transplanted immune system. Graft versus malignancy reaction. One of the most important observations to come from transplantation biology for hematopoietic malignancies is that transplanted individuals with GVHD appear to have a lower rate of relapse of their malignancy than individuals without GVHD. It is thought that the donor lymphocytes attack not only host target tissues (skin, liver, gut), but also residual host malignant cells. This is been designated the "graft versus malignancy" or "graft versus leukemia" effect. Supporting this hypothesis is the observation that relapsed individuals can return to remission after another infusion of donor lymphocytes. Therefore the trend in recent years has been to achieve a delicate balance with less intensive and non-myeloblative conditioning and sufficient immunosuppression to maintain the engraftment and hematopoiesis in order to enhance the graft versus malignancy effect. Another advance using graft versus malignancy effect has been the use of hematopoietic stem cell transplantation in non-myeloblative regimens for solid tumors such as melanoma, and for carcinoma of ovary, lung, colon, kidney, prostate, and breast[[18]Renga M. Pedrazzoli P. Siena S. Present results and perspectives of allogeneic non-myeloablative hematopoietic stem cell transplantation for treatment of human solid tumors.Annals of Oncology. 2003; 14: 1177Crossref PubMed Scopus (21) Google Scholar]. Allogeneic bone marrow transplantation (BMT) (transplantation of bone marrow from one genetically unique individual to another unique individual). The prototypic transplantation scenario in the past has been of bone marrow from one sibling to the other. In this situation, major histocompatibility antigens (HLA) are most easily matched because of close genetic identity of the two individuals. Nevertheless, a perfect HLA match does not prevent GVHD because other genetic systems are also involved. The minor histocompatibility antigens (MiHC) are less well understood and characterized and can differ between siblings. There are also genetically determined polymorphisms of genes for cytokines, chemokines, and molecules of the innate immune system which recognize bacteria and viruses that may be important. This area of investigation is a newly emerging one with great promise for the future. Better methods of genetic and immunologic matching will continue to improve the survival of transplanted individuals. Hematopoietic stem cell transplantation (HSC). This method of transplantation has been used for approximately a decade. HSC has been a great improvement because the transplant infusion material can be enriched for CD34-positive hematopoietic stem cells. This allows transplantation of fewer cells, and more rapid recovery of the recipient immune system. The risk for GHVD is thought to be higher with stem cell transplantation, however. Cord blood transplantation. A recent development in transplantation involves the use of immune cells in neonatal cord blood from placenta[19Isoyama K. Ohnuma K. Kato K. 26 Takahashi T.A. Kai S. Kato S. Takanashi M. Sato N. Sato H. Kitajima K. Naoe T. Saito H. Nishihira H. Japanese Cord Blood Bank. N. Cord blood transplantation from unrelated donors: a preliminary report from the Japanese Cord Blood Bank Network.Leukemia & Lymphoma. 2003; 44: 429Crossref PubMed Scopus (16) Google Scholar, 20Laughlin M.J. Barker J. Bambach B. Koc O.N. Rizzieri D.A. Wagner J.E. Gerson S.L. Lazarus H.M. Cairo M. Stevens C.E. Rubinstein P. Kurtzberg J. Hematopoietic engraftment and survival in adult recipients of umbilical-cord blood from unrelated donors.New England Journal of Medicine. 2001; 344: 1815Crossref PubMed Scopus (734) Google Scholar]. Because the neonatal immune system is immature (naive), it is possible to transplant across large disparities in HLA antigens without causing severe GVHD. This remarkable property of cord blood enlarges the numbers of potential donors for transplantation. The downside of cord blood transplantation is that the transplanted immune system remains immature for a long period of time (up to 18 months), making the recipient highly susceptible to fungal and viral infections. Non-myeloblative stem cell transplantation for human solid tumors. Incomplete immune ablation is a recent development in transplantation that avoids the known deleterious effects of intense conditioning x-ray and chemotherapy regimens. It adds to the beneficial effects of graft versus malignancy reaction in which residual malignant cells can be eliminated by the transplanted immune cells. Although still in its infancy, this area holds promise for older patients who cannot undergo intensive conditioning and those with highly immunogenic tumors such as renal cell carcinoma. However, no benefits have been demonstrated for patients with extensive metastatic disease[[18]Renga M. Pedrazzoli P. Siena S. Present results and perspectives of allogeneic non-myeloablative hematopoietic stem cell transplantation for treatment of human solid tumors.Annals of Oncology. 2003; 14: 1177Crossref PubMed Scopus (21) Google Scholar], and the results with melanoma, breast, ovarian, prostate carcinoma, to name a few, are still very preliminary. Life-threatening acute GVHD can still occur with nonmyeloablative transplantation. Older patients who develop severe GVHD after nonmyeloablative transplantation are also more likely to die of GVHD than younger ones[[21]Anagnostopoulos A. Giralt S. Critical review on non-myeloablative stem cell transplantation.Crit Rev Onc/Hematol. 2002; 44: 175Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar]. GVHD results from transplanting cells from individuals that are genetically different in cell surface molecules called histocompatibility antigens. The histocompatibility antigens were first identified in rodents as antigens responsible for rejection of tissue grafts between different strains of animals. Human major histocompatibility complex (MHC) genes encode antigens for tissue typing, many of which are involved in the steps leading to T cell activation. They lie on the short arm of chromosome 6 called the HLA (human leukocyte antigen) region and are designated MHC I and II. MHC Class I and Class II antigens are usually typed by molecular methods now, replacing the earlier serologic or cellular typing techniques. The average patient has approximately a 20 to 30 percent chance of having an HLA match within his immediate family. These numbers are higher for individuals from Japan with less genetic diversity, and lower for individuals of African descent with more polymorphisms than most other racial groups. Minor histocompatibility antigens (MiHC) are also important in GVHD but are less well understood and characterized. Depending on the degree of MHC and MiHC mismatch and the presence of certain critical genetic differences in donor and recipient, GVHD can occur. Ferrara et al.[22Teshima T. Ferrara J.L. Understanding the alloresponse: new approaches to graft-versus-host disease prevention.Seminars in Hematology. 2002; 39: 15Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 23Ferrara J.L. Pathogenesis of acute 30 graft-versus-host disease: cytokines and cellular effectors.Journal of Hematotherapy & Stem Cell Research. 2000; 9: 299Crossref PubMed Scopus (140) Google Scholar] have proposed a 3 phase model of GVHD (Table III): (1) The conditioning regimen, with high dose radiation and chemotherapy causes injury to the gut and other tissues, and release of inflammatory cytokines such as tumor necrosis factor alpha (TNF-alpha), interferon-gamma (IFN-gamma), interleukin 2 (IL-2) and interleukin 12 (IL-12). These changes promote activation of host antigen presenting cells (APCs) (2) Donor cell activation and expansion. Donor T cells recognize foreign recipient antigens presented by activated host APCs. Under the influence of altered cytokine environments, autoreactive T cells proliferate. (3) Target tissue injury. Self-reactive T cells then activate macrophages and natural killer cells which along with the T cells secrete additional cytokines that damage host tissue. NK cells and cytotoxic T cells attack keratinocytes directly.Table IIIThree Phase Model of Acute GVHD 22Teshima T. Ferrara J.L. Understanding the alloresponse: new approaches to graft-versus-host disease prevention.Seminars in Hematology. 2002; 39: 15Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 23Ferrara J.L. Pathogenesis of acute 30 graft-versus-host disease: cytokines and cellular effectors.Journal of Hematotherapy & Stem Cell Research. 2000; 9: 299Crossref PubMed Scopus (140) Google Scholar1.The conditioning regimen with XRT and chemotherapy produces host tissue injurya.Chemokines and cytokines are releasedb.Host dendritic cells in spleen and peripheral target tissues (skin, liver, gut) are activated2.Donor T cells are activated by dendritic cells and proliferatea.More chemokines and cytokines are released and expansion of autoreactive T cells occursb.Nonspecific downstream effector cells such as NK cells and monocyte/ macrophages are activated3.Target tissue injury occursa.Keratinocytes die by apoptosis in skin, liver and gut via "cytokine storm" and direct injury (satellite cell necrosis)b.Autoreactive donor T cells attack the remaining host immune system Open table in a new tab The skin, liver and mucosa as target organs of GVHD. All these targets of injury in GVHD are "barrier epithelia" which can respond rapidly to external insults with inflammation. Increased chemokines, cytokines, and antigen presenting capacity by dendritic cells may predispose the barrier organs to GVHD after conditioning that leads to enhanced nonspecific inflammation and tissue damage. One possible scenario is that gut injury allows leakage of LPS via gut flora into the bloodstream, leading to cytokine release, particularly TNF-alpha, a major cytokine of GVHD which can induce keratinocyte apoptosis. The specific molecules targeted in GVHD are not known, but are assumed to be cell surface molecules unique to epithelial cells. In skin, suprabasilar keratinocytes express VCAM-1 and keratin 15 and may be a target for injury[24Kim J.C. Whitaker-Menezes D. Deguchi M. Adair B.S. Korngold R. Murphy G.F. Novel expression of vascular cell adhesion molecule-1 (CD106) by squamous epithelium in experimental acute graft-versus-host disease.American Journal of Pathology. 2002; 161: 763Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar, 25Whitaker D. Jones S.C. Friedman T.M. Korngold R. Murphy G.F. An epithelial target site in experimental graft versus host disease and cytokine-mediated cytotoxicity is defined by cytokeratin 15.Biol Blood Marrow Transplant. 2003; 9: 559Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar]. Therefore, damage to skin results from a combination of "cytokine storm", and direct cytotoxic T cell and NK cell injury to keratinocytes (satellite cell necrosis). Types of cells that mediate GVHD that can trigger keratinocyte apoptosis([26]Vogelsang G.B. Lee L. Bensen-Kennedy D.M. Pathogenesis and treatment of graft-versus-host disease after bone marrow transplant.Annual Review of Medicine. 2003; 54: 29Crossref PubMed Scopus (165) Google Scholar). Donor cells. Both CD4-positive and CD8-positive T cells can induce GVHD in murine disease, depending on the histocompatibility mismatch. CD56-positive NK cells have also been shown to play a role not only in direct cytotoxic injury to epithelia but also in engraftment and graft versus malignancy effects. T regulatory cells that are CD4-positive CD25-negative may also be important. Recipient cells. Host dendritic cells (DCs) are critical to the development of GVHD ([27]Schlomchik W.D. Couzens M.S. Tang C.B. McNiff J. Robert M.E. Liu J. Shlomchik M.J. Emerson S.G. Prevention of graft versus host disease by inactivation of host antigen-presenting cells.Science. 1999; 285: 412Crossref PubMed Scopus (983) Google Scholar). DCs from old mice express more TNF-alpha and IL- 12, and stimulate better allo-reactions than DCs from young mice, possibly explaining the higher risk of severe GVHD in older patients[[28]Ordemann R. Hutchinson R. Friedman J. Burakoff S.J. Reddy P. Duffner U. Braun T.M. Liu C. Teshima T. Ferrara J.L. Enhanced allostimulatory activity of host antigen-presenting cells in old mice intensifies acute graft-versus-host disease.Journal of Clinical Investigation. 2002; 109: 1249Crossref PubMed Scopus (94) Google Scholar]. Lastly, expression of MHC molecules on host DCs only and not on epithelial cells is sufficient to initiate GVHD[[29]Teshima T. Ordemann R. Reddy P. Gagin S. Liu C. Cooke K.R. Ferrara J.L. Acute graft-versus-host disease does not require alloantigen expression on host epithelium.. 2002; 8 ([see comment][erratum appears in Nat Med 2002 Sep;8(9):1039]. Nature Medicine): 575Google Scholar]. Cytokines, chemokines and growth factors of GVHD. The Th1 cytokines predominate in classic cytotoxic GVHD. TNF-alpha, IFN-gamma and IL-2 are major cytokines that are increased in serum of patients with GVHD. TNF-alpha can cause apoptosis of cells directly; IFN-gamma can cause upregulation of MHC and costimulatory molecules on APCs and enhance Fas-mediated apoptosis. Keratinocyte growth factor can downregulate TNF-alpha and IFN-gamma and has a protective effect on GI epithelium, decreasing GVHD in a murine model[[29]Teshima T. Ordemann R. Reddy P. Gagin S. Liu C. Cooke K.R. Ferrara J.L. Acute graft-versus-host disease does not require alloantigen expression on host epithelium.. 2002; 8 ([see comment][erratum appears in Nat Med 2002 Sep;8(9):1039]. Nature Medicine): 575Google Scholar]. Chemokines play a critical role in recruitment of leukocytes to target tissues. In autologous/syngeneic GVHD, autoreactive CD8+ T cells recognize MHC class II antigens in the setting of markedly elevated IL-10 and MIP-1 alpha mRNA levels by PBMCs in patients. Patients predisposed to develop autologous/syngeneic GVHD have IL-10(-1082) G/G polymorphic alleles that determine increased baseline IL-10 production([30]Hess A.D. Thoburn C.J. Immunobiology and immunotherapeutic implications of syngeneic/autologous graft-versus-host disease.Immunological Reviews. 1997; 157: 111Crossref PubMed Scopus (55) Google Scholar), a possible explanation for their susceptibility to syngeneic/autologous GVHD. In sclerodermatous GHVD, where disease is characterized by skin and visceral fibrosis rather than epithelial injury, increased production of TGF-beta by T cells and monocyte/macrophages leads to upregulated collagen synthesis in a murine model[[31]McCormick L.L. Zhang Y. Tootell E. Gilliam A.C. Anti-TGF-beta treatment prevents skin and lung fibrosis in murine sclerodermatous graft-versus-host disease: a model for human scleroderma.Journal of Immunology. 1999; 163: 5693PubMed Google Scholar]. The cytokine profile in sclerodermatous GVHD is mixed, but is Th2-predominant. CD4+ T cells are increased in early disease and activated monocyte/macrophages appear to play an important role in pathogenesis[[32]Zhang Y. McCormick L.L. Desai S.R. Wu C. Gilliam A.C. Murine sclerodermatous graft-versus-host disease, a model for human scleroderma: cutaneous cytokines, chemokines, and immune cell activation.Journal of Immunology. 2002; 168: 3088Crossref PubMed Scopus (204) Google Scholar]. These are areas of active investigation, in which researchers are testing novel immunomodulatory molecules in mouse models for GVHD. The goal is to develop more specific and targeted interventions that will decrease GVHD, preserve graft versus malignancy effect, and decrease the risk of infection that occurs with systemic immunosuppression. Graft versus host disease is a potentially devastating consequence of hematopoietic transplantation that is immunologically mediated. Several variant forms exist (acute, chronic lichenoid, chronic sclerodermoid, transfusion-related, autologous/syngeneic). GVHD results from attack of transplanted donor lymphocytes on host tissues when histocompatibility and immunologic differences exist between donor and recipient (allogeneic transplantation), and when recovery of the immune system occurs in the setting of damaged thymus and cyclosporin administration (syngeneic/autologous transplantation). Skin is a major target organ, along with gut and liver. A plausible hypothesis for GVHD is that conditioning with XRT and chemotherapy before transplantation damages tissue, causing release of cytokines and chemokines and causing nonspecific inflammation in the "barrier epithelia" that interface with the outside environment. Host antigen presenting cells in these sites are activated, and then interact with donor lymphocytes brought to the area by the inflammation. The donor lymphocytes proliferate and trigger downstream events (more cytokine secretion, activation of macrophages and NK cells). Damage to epithelia of skin, gut and liver follows via direct cytotoxic injury and via cytokines that cause keratinocyte apoptosis. Mouse models for GVHD have been invaluable in dissecting out the critical pathways in GVHD and in testing new interventions to prevent and treat GVHD. In the past, therapy has been with potent nonspecific immunosuppressants. More recently, exciting new more focused therapies (immunomodulatory molecules and antibodies to immune cell markers, PUVA, extracorporeal photopheresis) have been developed and are increasingly useful for GVHD.