Graft-versus-host disease: allo- and autoimmunity after bone marrow transplantation.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to H M Shulman.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The authors immunostained autopsy liver tissue from 31 marrow transplant recipients, 19 with hepatic veno-occlusive disease (VOD) and 12 without VOD. A panel of monoclonal and polyclonal antibodies was used to characterize the materials within the occluded venous lesions and to define the location and types of injured cells. Most patients with early VOD (survival less than 50 days, n = 11) had dense periadventitial and intramural immunostaining of terminal hepatic and sublobular central venules (CV) with anti-Factor VIII (9/11) and anti-fibrinogen (4/4) but had no immunostaining with antibody to platelet GPIb. Early VOD patients also had marked loss of Zone 3 hepatocyte cytokeratin (8/9) versus late VOD (1/5) or non-VOD (2/7). Patients with late VOD lesions (n = 8, survival greater than 50 days) had increased collagen within occluded VOD lesions. Type III much greater than Type I, and increased sinusoidal collagens Types I, III, and IV. These studies and other data suggest the following events in the genesis of VOD. Initial injury to endothelium of CV and/or sinusoids and possibly also to Zone 3 hepatocytes triggers the coagulation cascade in the periadventitial zone of CV. The late sequela, collagenous CV occlusion, results from activated mural myofibroblasts and/or embolized Ito cells and hepatocytes.
The authors attempted to induce hepatic veno-occlusive disease (VOD) in 64 dogs. Preparative treatments included combinations of total-body irradiation (TBI) or localized hepatic irradiation (LI) or both and chemotherapy consisting of dimethylbusulfan (DMB), L-phenylalanine mustard (L-PAM), methotrexate, or monocrotaline. VOD occurred infrequently in those dogs given 9.2 Gy TBI and DMB (1/10), TBI and/or LI (9.2-27 Gy) with L-PAM (2/36) or high dose methotrexate and LI (0/2). Specifically, VOD occurred in the dogs with a shorter interval between TBI and DMB or in the dog that received the glutathione reductase inhibitor, buthionine sulfoximide (BSO) before L-PAM. In contrast, among 17 dogs given monocrotaline, 8 developed VOD, particularly when used with L-PAM +/- irradiation (7/13). The major cause of death, early gastrointestinal toxicity, was further augmented by higher doses of irradiation, by shortening the interval between LI and L-PAM administration to less than 4 weeks, and administering BSO or monocrotaline before L-PAM. Gastrointestinal toxicity was lessened by giving low dose cyclophosphamide given before L-PAM. VOD can be produced in dogs especially with monocrotaline or BSO given before and L-PAM +/- irradiation. However, gastrointestinal toxicity renders the study of VOD beyond the acute phase difficult. Nevertheless, this approach appears useful for the study of VOD in other animals and for developing agents aimed at preventing VOD.
We determined the incidence of venocclusive disease of the liver (VOD) in 96 recipients of autologous bone marrow transplants (BMT) to be 9.4%, a figure less than that reported for allogeneic transplantation. The development of VOD was compared in a cohort of 21 autologous BMT recipients and in 56 randomly chosen, comparably conditioned, concurrent allogeneic BMT recipients. One of these 21 (4.8%) autologous recipients developed VOD, versus 14 of 56 (25%) allogeneic recipients (P less than 0.05). Logistic regression analysis confirmed pretransplant hepatocellular dysfunction as a risk factor for VOD, and suggested that the use of methotrexate and/or cyclosporine contributes to the development of VOD after chemoradiation therapy. However, a graft-versus-host reaction cannot be excluded as a cause of the higher incidence of VOD in allogeneic recipients.
Fifty-eight patients received an allogeneic or syngeneic marrow transplant following conditioning with high doses of dimethylbusulfan (DMB), cyclophosphamide (CY) and total body irradiation (TBI). Thirty-two patients had either chronic myeloid leukemia (CML) in accelerated phase or blast transformation, or acute leukemia after first relapse. The actuarial survival of these 32 patients at 3 years was 12% compared with 25% for a group of 206 patients with similar diagnoses prepared for transplantation with CY and TBI alone. This reduced survival was associated with a greater incidence of early non-leukemic deaths, in particular as a result of severe hepatic veno-occlusive disease. The incidence of leukemic relapse was not different in the two groups. Of 13 patients with CML in chronic phase who received syngeneic transplants following DMB, CY and TBI, nine are alive in hematologic and cytogenetic remission from 3.9 to 9.4 (median 6.2) years post-transplant.
Thirty patients with advanced preleukemic syndromes were treated with marrow transplantation. Most cases were diagnosed by the presence of peripheral pancytopenia and a diagnostic marrow examination but in 6 of the 30 patients pretransplant chromosome studies were instrumental in establishing the diagnosis. Three patients prepared for transplantation with cyclophosphamide alone recurred with their disease within 6 months of transplantation. The other 27 patients were treated with cyclophosphamide and total body irradiation. Twenty of these 27 patients had preleukemia not associated with prior therapy or severe marrow fibrosis. Thirteen of these 20 are alive and well 9 to 56 months from transplant and 7 died, 4 of interstitial pneumonia, 2 of candida septicemia, and 1 of disseminated zoster. There have been no disease recurrences in this group. The remaining preleukemic patients, which include 3 patients transplanted for preleukemia secondary to prior therapy and 4 patients transplanted for preleukemia associated with severe marrow fibrosis, have all died. Major problems in these patients included disease recurrence (2 cases) and, in those with severe marrow fibrosis, graft failure (2 cases). These results suggest that for patients with life-threatening pancytopenia due to spontaneous preleukemia without severe marrow fibrosis, marrow transplantation can prolong disease-free survival and may result in cure of the disease.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The cases of 13 allogeneic marrow transplant recipients who had undergone laparotomy for manifestations of severe enteritis were reviewed to determine the causes of the severe intestinal disease and to assess the relation between clinical, histologic, and microbiologic findings. Laparotomies were performed a median of 63 days (range, 11 to 273 days) after transplantation for suspected peritonitis, intestinal obstruction, or bleeding. Intestinal tissue was available from small bowel resections in nine patients, intraoperative biopsies in one, and from autopsies in three patients who died shortly after laparotomy. Widespread small bowel ulceration was present in all 13 cases. Four causes of ulceration were identified: chemoradiation toxicity (n = 2), acute graft-versus-host disease (GVHD) (n = 5), opportunistic infections superimposed on either GVHD or toxicity from chemotherapy (n = 4), and Epstein-Barr virus-associated lymphoproliferative disorder (n = 2). Intestinal infections, unrecognized before laparotomy, were due to cytomegalovirus (CMV), herpes simplex virus (HSV), adenovirus, and Torulopsis glabrata. CMV- and HSV-infected cells, often lacking diagnostic inclusions, were identified in the intestine by in situ hybridization with biotinylated DNA probes. Eleven patients died in the perioperative period, and two died 452 and 558 days after surgery of complications of chronic GVHD. Poor outcomes were related to extensive intestinal involvement, which was commonly underestimated before surgery, failure to diagnose intestinal infections early, poor marrow function, impaired immunity, and refractoriness of severe GVHD.
We have described the first case of systemic Fusarium moniliforme infection complicating marrow transplantation. The infection was fulminant despite aggressive therapy. Review of previously reported cases of human Fusarium infection suggests that the skin is a common portal of infection.
Based on studies in multiply transfused dogs showing that cyclosporine (CsA) pregrafting reduced the incidence of rejection, we treated 11 multiply transfused patients with severe aplastic anemia with CsA, 12.5-20 mg/kg/day i.m. on days -6 to -1, and cyclophosphamide, 50 mg/kg/day i.v. on days -5 to -2, followed on day 0 by a marrow graft from an HLA-identical sibling donor. Patients were not given additional infusions of buffy coat cells from their marrow donors. As postgrafting prophylaxis for graft-versus-host disease 10 patients were given CsA, 12.5 mg/kg/day orally, and one patient was given a standard regimen of intermittent methotrexate. Among 10 evaluable patients, 2 rejected their marrow grafts. Eight patients had sustained engraftment. Two of these developed severe veno-occlusive disease of the liver, a complication previously observed only in patients conditioned with total-body irradiation but not in more than 200 patients with aplastic anemia conditioned with cyclophosphamide without the addition of CsA. These data indicate that conditioning of patients with cyclophosphamide and concurrent CsA does not uniformly prevent graft rejection and can result in severe hepatic toxicity.
A murine monoclonal antibody to cytomegalovirus (CMV) was used to identify virus-infected cells in coded frozen tissue sections from 52 consecutive open-lung biopsies obtained from marrow transplant recipients with pneumonia. The diagnostic sensitivity of immunofluorescence (IF) using this antibody exceeded that of standard histology performed on touch imprints and frozen and permanent lung sections and was equal to viral culture and in situ CMV nucleic acid hybridization. In comparison with patients with CMV pneumonia demonstrated histologically and by IF, those with negative histology and positive IF were more likely to have seroconverted before biopsy. Despite this evidence of an immune response to CMV pneumonia, the two groups did not differ in CMV positivity by culture or by hybridization, and their long-term survival was equally poor. The findings demonstrate that this antibody can play an important role in both the rapid diagnosis of CMV infection and the elucidation of CMV pathophysiology.
Two hundred fifty-five patients received chemoradiotherapy and a marrow graft for treatment of malignancy. Fifty-three developed venocclusive disease (VOD) of the liver. The clinical presentation was characterized by jaundice, fluid retention, ascites, upper abdominal pain, and encephalopathy. Insidious weight gain was the first sign of VOD, occurring a mean of 6.2 +/- 5.2 days after transplantation, followed shortly by jaundice. Twenty-four patients (45%) had a serious, progressive liver disease, but the others recovered a mean of 21.6 days after the onset of jaundice. Analysis of pretransplant factors did not disclose a significant association with serious VOD. Patients with serious VOD had significantly higher maximal values for bilirubin and SGOT, gained more weight, and were more likely to have encephalopathy. Supportive treatment did not appear to influence the outcome.
Explore the source record for details and available documents.
Recent data show that stem cells in primate epidermis are concentrated at the bases of rete ridges. Because the early lesions of graft-versus-host disease (GVHD) in the vertebrate skin are in basal epidermis and hair follicle cells, the authors hypothesized that stem cells or their early progeny might be targeted. In this study they used morphometric methods to examine the distribution of the lesions of GVHD in the skin and lip of human bone marrow allograft recipients. They found that rete ridges are the primary sites of attack in early GVHD of the skin. They also found that the concentration of stem cells in ridges for various anatomic sites (body or arm, palm or sole, and lip) is directly proportional to the frequency and concentration of the lesions of GVHD. Osmotic fragility and the expression of early differentiation antigens related to the major histocompatibility complex are discussed as potential explanations for this phenomenon.