Graft Versus Host Disease - Pathophysiology & Management

Amarin Alexander, M.D. & Shirley Wong, Pharm.D.

 
Introduction

First described in irradiated mice infused with donor cells in the 1950's, graft versus host disease has been the bane of scientists' existence from the beginning of allogeneic bone marrow transplantation (BMT). The presence of immunocompetent donor cells in an immunocompromised host may cause a number of symptoms and cellular changes often leading to acute or chronic rejection, termed Graft Versus Host Disease (GvHD). In the intervening years, as our understanding of the immune system has progressed, so has our ability to control, or limit the process of GvHD. As well, over the last 50 years, we have unearthed a beneficial phenomenon termed "Graft versus Tumor" (GvT), a consequence of GvHD that may lead us down new paths of cancer therapy.1,2

GvHD

Billingham in 1966 proposed the requirements under which GvHD can occur, later summarized by Ferrara and Deeg: `the graft must contain immunologically competent cells, the recipient must express tissue antigens that are not present in the transplant donor, and the recipient must be incapable of mounting an effective response to destroy the transplanted cells.'3,4 Today we know that the mediators of GvHD are T cells, human leukocyte antigens (HLAs) in the host, and the host being typically immunocompromised either chemically or physiologically (as in the case of neonates). While there are cases of GvH-like syndromes in autologous and syngeneic transplants, these are typically mild.10 Thus, allogeneic BMT is overwhelmingly the most common setting in which clinically significant GvHD is seen.

Extensive work on HLA antigens over the years has demonstrated that the better the HLA class I and II match, the lower the risk and intensity of the GvHD. Recently the minor HLA antigens have also been identified as contributing to GvHD. Prophylaxis of the host also has been associated with less severe GvHD. Risk factors for developing GvHD are indicated in Table 1.

In discussing GvHD, it is important to delineate between acute and chronic GvHD. The cutoff is arbitrarily set at 100 days post-BMT in clinical trials, but in practice, manifestations of both acute and chronic GvHD may be found later or earlier, respectively.5 Clinically the symptoms of both can be mild, moderate, or severe, with 100% mortality associated with the latter.

Acute GvHD (aGvHD)

Incidence of rejection varies, but is inversely correlated with the number of HLA matches. For example, in genotypically identical sibling transplants or with HLA-matched transplants, aGvHD ranged from 40% to 59%.6 However, in the case of 2 or 3 mismatched HLA antigens, the incidence of GvHD increases to 80%.7 Onset may begin within days in hyper-acute GvHD, but the median is 19-25 days after transplantation.

As described graphically by Ferrara et al,8 the pathophysiology of acute GvHD is easier to understand in the setting of a 'profoundly damaged' host. He reminds us that the typical recipient—the host—of BMT is usually immunocompromised with substantial endothelial and epithelial damage from infection, and preparative chemotherapy and/or radiation therapy. This milieu of increased cytokines, adhesion molecules, and other molecular activators creates a stimulating broth for immunocompetent donor T-cells. Donor cells find themselves in a foreign, inhospitable environment from the moment they are infused. It is no surprise that these cells react "normally," by initiating an inflammatory response.

In fact, one can extrapolate from the organs involved in GvHD, that the immune reaction does indeed mimic that of an anti-infectious inflammatory response. The target organs of GvHD, namely the skin, liver, and gut, are all involved in first-line defense from bacteria. Because of this role, these organs boast plentiful supplies of antigen-presenting cells (APC's)—the very cells that activate the immune response when paired with competent lymphocytes. This distinguishes these organs from the heart and kidneys, which are rarely, the targets of aGvHD.

Acute GvHD has been described in three phases. The time prior to actual donor infusion, when the patient is becoming immunocompromised by the transplant conditioning therapy, is considered Phase I. During this time, the host is undergoing cytokine release (such as TNF-a and IL-1) due to inflammation and cellular damage secondary to the pre-transplant regimen. As mentioned, this inflammatory state has been linked in GvHD. In fact, as the host inflammatory reaction abates, the risk of GvHD also drops. The second phase occurs after transfusion.

During Phase II, the donor T-cells are activated by host APC's and begin to proliferate. Importantly, if host APC's are eliminated prior to the BMT, the incidence of GvHD drops dramatically. Clearly, the host participates in the propagation of the GvHD by its APC « donor T-cell interaction.

Explanations for the third or inflammatory phase have moved beyond the purely cytotoxic T lymphocyte model. Other mediators of aGvHD are now known include natural killer cells and monocytes reacting to IL-1, IL-2, IFN-g, TNF-a and other cytokines, as well as direct cytokine toxicity itself. Viral and bacterial infections have also been hypothesized as precipitants of cGvHD, with host organs as `innocent bystanders' of the immune response.8

In addition to the three phases of aGvHD described above, clinical grading of aGvHD is often performed. Staging ranges from grade I with milder symptoms to stage III with severe multi-organ involvement, and grade IV being life-threatening.9,10 Most patients with aGvHD present with rash and diarrhea which can progress to life-threatening desquamation and liver failure, respectively. The mortality associated with grade IV aGvHD approaches 100%.4Initial symptoms are typically rash, diarrhea, and abdominal pain. Liver function abnormalities are often noted, but appear later and are correlated with the severity of the reaction. Skin changes are the most common manifestation of both acute and chronic GvHD. Other symptoms are noted in table 2. Acute GvHD leads to a delay in return to immunocompetence of the host. This, plus the iatrogenic immunosuppression of the host makes infection, not organ failure, the most common etiology of death in aGvHD.6

Chronic GvHD (cGvHD)

Chronic GvHD is loosely defined as GvHD occurring 100 days after transplantation. The incidence varies with HLA-matching and the source of the donor cells (bone marrow versus peripheral stem cell) but ranges from 27% to 72%.12 The symptoms may overlap those of the acute phase, and may begin to appear as late as two years after transplantation. Organ fibrosis with collagen deposition and atrophy are the hallmarks of cGvHD and thus, in its effects, cGvHD is often likened to autoimmune diseases such as lupus or scleroderma. Diagnosis may be made clinically or by biopsy and is graded either "limited or extensive," though some groups use "mild, moderate, and severe."4,12

cGvHD is generally subdivided into three categories:

  • Progressive onset: Continuation of aGvHD
  • Quiescent: Presence of cGvHD with previously resolved aGvHD
  • De novo: New onset cGvHD with no prior aGvHD

Prognosis is generally better when the cGvHD is de novo, that is, without any prior acute phase. Poor prognosis is associated with older patients and resistance to GvHD treatment.6 Other factors associated with poor prognosis in cGvHD are listed in table 3.

Clearly, the incidence of cGvHD varies according to many of the same factors affecting aGvHD. Additionally, CMV infection in the host or donor, and previous aGvHD are associated with increased incidence of cGvHD.6 Interestingly, donor age and recipient radiation source or dose do not appear to influence the incidence of cGvHD. Abnormalities of cellular immunity and selective epithelial attack are contributors to cGvHD. Decreased thymic function of the host due to age or pre-transfusion chemoradiotherapy has lead some to propose extra-thymic production of T lymphocytes with decreased negative selection and subsequent autoreactivity as another link in cGvHD.12,13

As in autoimmune disease, skin changes can be diffuse and devastating or localized and clinically insignificant. Over 80% of patients with cGvHD will have at least skin and glandular changes. The former can develop into severe contractures, ulcers, and atrophy while the latter usually involves drying of the mucous membranes of the eyes, mouth, and urethra.6

Much work has centered around prevention of GvHD. Immunocompetent donor T cells have been identified as the chief culprit in cGvHD, with severity of the disease actually linked to the number of donor immunocompetent T cells present in the host.14 Since we know that T cells are one of the major players, T lymphocyte-depleted BMT has been studied. While these have shown promise in decreasing the incidence of GvHD, the coincident tumor relapse rates have been unsatisfactorily high. One explanation is the suppressive effect the donor T-cells have on the host's residual tumor load. It is now generally thought that the infused donor T cells may have a tumor-suppressive effect, referred to as Graft versus Tumor effect (GvT).

Prevention of GvHD

Preventing aGvHD may be the best way of preventing or limiting cGvHD. Strategies are generally aimed at eliminating donor T cells or blocking their activation during GvH response. The approaches have included in-vitro manipulation of the bone marrow and treatment of the recipient with immunosuppressive therapy after transplantation.15

The removal of T cells from the donor's bone marrow can be accomplished by physical separation via lectin agglutination or by treatment with monoclonal antibodies directed against T cells. These procedures usually achieve more than 90% reduction of T cells in the donor's bone marrow and result in substantial reductions in the incidence and severity of GvHD.16 Although T-cell depletion reduces mortality from GvHD, problems with engraftment and increased relapse rates of leukemia resulted in survival rates similar to those for patients who receive conventional prophylactic immunosuppressive therapy.

Immunosuppressive therapy typically consists of combined regimens of methotrexate (MTX), cyclosporin (CsA), tacrolimus (FK 506), and/or a corticosteriod. In general, combined therapies are more efficacious than monotherapy. Storb et al17 reported a higher rate of aGvHD in patients who received methotrexate alone compared with those treated with the combination of methotrexate and cyclosporin (53% versus 18%). These pharmacologic agents affect different stages of donor T-cell activation providing the rationale for the use of combined therapies. Methotrexate, an antiproliferative agent, prevents the division and clonal expansion of T cells. Cyclosporin and tacrolimus, on the other hand, block the synthesis of IL-2 and thus prevent T-cell activation. Corticosteroids are lympholytic; they also prevent the synthesis of IL-1 in APC's and thereby eliminate one of the stimuli required for antigen presentation.18-23

The most common regimen for GvHD prophylaxis includes the use of long-term cyclosporin therapy starting 1-2 days prior to transplantation and continuing for 6 months after the transplant, plus a short course of intermittent methotrexate boluses. A corticosteroid may be added to increase efficacy, however it may also increase the risk of infection. Methotrexate is myelosuppressive and can cause severe mucositis and diarrhea. The adverse effects from cyclosporin and tacrolimus appear to be similar, with nephrotoxicity and hypertension be the most common. Closed monitoring of plasma drug levels and dosage adjustments based on severity of the side effects are recommended. While many of these combinations may decrease the incidence of cGvHD, some combinations have potentially severe side effects as described above. Others, while preventing cGvHD, may not necessarily decrease long-term mortality. Generally, response to immunosuppressive regimens is good, with 5-year survival up to approximately 60% and up to 50% complete remission.12

Intravenous gamma-globulin preparations administered prophylactically have also been shown to be beneficial for the prevention of GvHD. Although the mechanism of action is not clear, it was hypothesized that binding of the gamma globulins to cells with Fc receptors, such as the natural killer cells, provides a signal for the elimination of the cells or prevents them from recognizing the target tissues.24 Pentoxyfylline, a xanthine derivative capable of down-regulating TNF-a production, when administered with cyclosporin plus either methotrexate or methylprednisolone was associated with a decreased incidence of GvHD compared with historical controls in a phase I/II trial.25 However, more randomized clinical trials are needed to examine its effect on long term survival. Other drugs that are currently under investigation include trimetrexate, fludarabine, and rapamycin (Sirolimus).10

Treatment of GvHD

The immunosuppressive agents used to prevent GvHD also may be used to treat established GvHD. They are generally less efficacious for treatment than prophylaxis. Corticosteroids with or without cyclosporin have been the standard of treatment for cGvHD. Tacrolimus provides an effective alternative in patients with GvHD unresponsive to cyclosporin therapy.26 Antithymocyte globulin (ATG) has also been used, however its efficacy is variable and its usefulness is limited by its toxicity.10 Ursodiol has shown benefit when used on a short-term basis to treat cGvHD of the liver that is refractory to immunosuppressive therapy.27 Notably, none of the present treatments are totally effective in the majority of patients. Clinical trials are under way to define the optimal drug therapy for the management of cGvHD. Thalidomide with immunosuppressive properties has shown promising results in the treatment of cGvHD. In a clinical trial, high-risk patients who received thalidomide as primary treatment showed a 3-year survival rate of 48% which was very similar to the results reported with cyclosporin and prednisone combination therapy. 28 In addition, a phase II study has confirmed the apparent efficacy of thalidomide as salvage therapy in patients with cGvHD unresponsive to conventional treatment.29 A randomized trial of thalidomide versus placebo added to cyclosporin and prednisone therapy is currently being investigated in patients with high-risk cGvHD.10 Similar to thalidomide, clofazimine, used in leprosy and immune-mediated skin disorders, has shown activity in some patients with cutaneous and oral cGvHD as well.30

Mycophenolate, a potent immunosuppressive agent for the prevention of acute rejection in renal transplantation, has also been studied in BMT, but the results were disappointing. The use of Mycophenolate for cGvHD was associated with low response rates and high incidence of leukopenia, anemia, and thrombocytopenia.31-33

Etretinate, a synthetic vitamin A derivative, has been used in the treatment of cGvHD of the skin. Marcellus et al34 found that 74% of the patients with refractory sclerodermatous cGvHD experienced softening of the skin, flattening of the cutaneous lesions, increased range of motion, and performance status while on oral etretinate. The major side effects were skin breakdown and ulceration leading to discontinuation of therapy in 18% of the patients.

Research on the use of monoclonal antibodies against T-cell activation and proliferation during GvHD is increasing. Antibodies directed against cytokines, such as tumor necrosis factor (TNF), and their receptors, such as IL-2 receptor antagonist (Daclizumab), appear promising. Immunotoxins, such as anti-CD5-ricin targeting mature T cells, are now undergoing clinical trials.4 According to some case reports and animal studies, penicillamine, cyclophenile, halofunginone, and nedocromil sodium may have a role in the treatment of cGvHD. 12

Supportive care for patients with GvHD is also important. Topical sunscreen products to prevent worsening of skin involvement, artificial tears for ocular dryness, pilocarpine for xerostomia, and nutritional supplements for patients with severe mucositis are crucial. For patients on long-term corticosteroid therapy, estrogen replacement in women, high calcium and vitamin D intake, antiosteoporosis agents, and daily exercise are highly encouraged to decrease bone loss.35

Future Paths

Early recognition and intervention will hopefully reduce the severity of GvHD when it occurs. Of course, better HLA matching and recognizing the risk factors of GvHD will also help in reducing the incidence. Research into selectively depleting the donor lymphocyte contamination is a priority. Increased understanding in the cellular biology of lymphocytes and antigen-receptor interactions during GvHD is critical to provide new insights into the pathophysiology of the disease and aid in designing and optimizing drug therapy. In addition, host APC or MHC blockade may be another therapeutic option in the future. The role of umbilical cord stem cells, which are theoretically immunologically immature, is another area of investigation, with promising initial results.

Conclusion

Great strides in the last 50 years have led to improved outcomes in BMT recipients. In fact from this research, BMT, once reserved for leukemics, is beginning to be applied to other hematologic disorders like multiple myeloma, aplastic anemia, and sickle cell disease. Currently, immunosuppressive drug therapy is the mainstay of GvHD prophylaxis and treatment but is associated with severe toxicity. T-cell depletion of bone marrow grafts decreases the incidence of GvHD but increases the risk of engraftment failure and leukemia relapse. Unfortunately, work is still needed to prevent and manage GvHD which, together with infection, is still the major cause of morbidity and mortality in the BMT patients. Improved understanding of the immune system and inflammatory response will aid in this effort.

REFERENCES:

  1. Thomas ED. Bone marrow transplantation: a review. Seminars in Hematology 1999;36(S7):95-103.
  2. Weiden PL, Flournoy N, Thomas ED, et al. Antileukemic effect of chronic graft-versus-host disease in human recipients of allogeneic marrow grafts. New England Journal of Medicine 1979;300(19):1068-73.
  3. Billingham RE. The biology of graft-versus-host reactions. Harvey Lect 1966-67;62:21-78.
  4. Ferrara JLM and Deeg HJ. Graft-versus-host Disease. New England Journal of Medicine 1991;324(10): 667-674.
  5. Smith BR. Stem cell transplantation. In: DeVita VT, Hellman S, Rosenberg SA, eds. Cancer: principles and practice of oncology. Philadelphia: Lippincott-Raven, 1997: 2621-2638.
  6. Caudell KA. Graft-versus-host disease. In: Whedon MB and Wujcik D, eds. Blood and Marrow Stem Cell Transplantation. Sudbury, MA: Jones and Bartlett, 1997:177-204.
  7. Przepiorka D, Ippoliti C, Koberda J et al. Interleukin-2 for prevention of graft-versus-host disease after haploidentical marrow transplantation. Transplantation 1994;58(7):858-860.
  8. Ferrara JLM, Levy R, Chao NJ. Pathophysiologic mechanism of acute graft-vs.-host disease. Biology of Blood and Bone Marrow Transplantation. 1999;(5):347-356.
  9. Deeg HJ, and Storb R. Graft-versus-host disease: pathophysiological and clinical aspects. Annual Review of Medicine 1984;35:11-24.
  10. Flowers MED, Kansu E, Sullivan KM. Hematopoietic stem cell therapy. Pathophysiology and treatment of graft-versus-host disease. Hematology/Oncology Clinics of North America 1999;13(5):1091-1107.
  11. Deeg HJ, Storb R, Thomas ED. Bone marrow transplantation: a review of delayed complications. British Journal of Haematology 1984;57:185.
  12. Gaziev D, Galimberti M, Lucarelli G, et al. Chronic graft-versus-host disease: is there an alternative to the conventional treatment? Bone Marrow Transplantation 2000;25:689-696.
  13. Weinberg K, Annet G, Kashyap A, et al. The effect of thymic function on immunocompetence following bone marrow transplantation. Biology of Blood Marrow Transplantation 1995;1:18-25.
  14. Kernan NA, Collins NH, Juliano L, et al. Clonable T lymphocytes in T cell-depleted bone marrow transplants correlate with development of graft-v-host disease. Blood 1986;68:770-3.
  15. Martin PJ. Pharmacologic approach for prevention and treatment of acute graft-versus-host disease. Clin Aspects Autoimmunity 1990;4(2):8-
  16. Vallera DA, Blazar BR. T cell depletion for graft-versus-host disease prophylaxis: a perspective on engraftment in mice and humans. Transplantation 1989;47:751-60.
  17. Storb R, Deeg HJ, Pepe M, et al: Graft-versus-host disease prevention by methotrexate combined with cyclosporin compared to methotrexate alone in patients given marrow grafts for severe aplastic anaemia: Long-term follow-up of a controlled trial. Br J Haematol 1989;72:567-572.
  18. Culpepper J, Lee F. Glucocorticoid regulation of lymphokine-production by murine T lymphocytes. In: Webb DR, Goeddel DV, eds. Lymphokines: molecular cloning and analysis of lymphokines. New York: Academic Press, 1987:275-89.
  19. Kahan BD. Cyclosporin. N Eng J Med 1983;309:1094-104.
  20. Jolivet J, Cowan KH, Curt GA, et al. The pharmacology and clinical use of methotrexate. N Eng J Med 1983;309:1094-104.
  21. Fay JW, Wingard JR, Antin JH, et al. FK506 (tacrolimus) monotherapy for prevention of graft-versus-host disease after histocompatible sibling allogeneic bone marrow transplantation. Blood 1996;87:3514-3519.
  22. Nash RA, Pineiro LA, Storb R, et al. FK506 in combination with methotrexate for the prevention of graft-versus-host disease after marrow transplantation from matched unrelated donors. Blood 1996;88:3634-3641.
  23. Venkalaramanan R, Swaminathan A, Prasad T, et al. Clinical Pharmacokinetics of tacrolimus. Clin Pharmacokinet 1995;29(6):404-430.
  24. Sullivan KM, Kopecky KJ, Jocom J, et al. Immunomodulatory and antimicrobial efficacy of intravenous immunoglobulin in bone marrow transplantation. N Eng J Med 1990;323:705-12.
  25. Bianco JA, Appelbaum FR, Nemunaitis J, et al. Phase I-II trial of pentoxyfylline for the prevention of transplant-related toxicities following bone marrow transplantation. Blood 1991;78:1205-11.
  26. Kanamaru A, Takemoto Y, Kakishita E, et al. FK506 treatment of graft-versus-host disease developing or exacerbating during prophylaxis and therapy with cyclosporin and/or other immunosuppressants. Bone Marrow Transplant 1995;15:885-889.
  27. Fried RH, Murakami CS, Fisher LD, et al. Ursodeoxycholic acid treatment of refractory chronic graft-versus-host disease of the liver. Ann Intern Med 1992;116:624-629.
  28. Vogelsang GB, Farmer ER, Hess AD, et al. Thalidomide for the treatment of chronic graft versus host disease. N Engl J Med 1992;326:1055-1058.
  29. Parker PM, Chao N, Nademanee A, et al. Thalidomide as salvage therapy for chronic graft-versus-host disease. Blood 1995;86:3604-3609.
  30. Lee SJ, Wagner SA, McGarigle CJ, et al. Treatment of chronic graft-versus-host disease with clofazimine. Blood 1997;89:2298-2302.
  31. Basara N, Blau WI, Romer E et al. Mycophenolate mofetil for the treatment of acute and chronic GVHD in bone marrow transplant patients. Bone Marrow Transplant 1998;22:61-65.
  32. Nash RA, Fulong T, Storb R, et al. Mycophenolate mofetil (MMF) as salvage treatment for graft-versus-host disease (GVHD) after allogeneic hematopoietic transplantation (HSCT): safety analysis. Blood 1997;10(suppl 1):105a (Abstr. 559).
  33. Mookerjee B, Althomonte V, Vogelsang G. Salvage therapy with mycophenolate mofetil and FK506 for refractory chronic graft-versus-host disease. Blood 1998;10(suppl 1):457a (Abstr. 559).
  34. Marcellus DC, Altomonte VI, Farmer ER et al. Etretinate therapy for. refractory sclerodermatous chronic graft-versus-host disease. Blood 1999;1:66-70.
  35. Siadak M, Sullivan KM. The management of chronic graft-versus-host disease. Blood Rev 1994;8:154-160.

November, 2000/ Jacksonville Medicine [dcms-footer.htm]