154 Transplant Immunology Summer School Conclusions Approximately 5000 allogeneic transplant are currently per- formed each year world-wide and reported to the Inter- national Bone Marrow Transplant Registry. Indications are changing with time and now the haemoglobinopathy, tha- lassaemia, has become a more common indication than aplastic anaemia for the procedure world-wide. Predicting graft-versus-host disease in allogeneic bone marrow transplant recipients Anne Dickinson Department of Haematology, Royal Victoria Infirmary, Newcastle upon Tyne Graft-versus-host disease (GVHD) remains a major compli- cation of allogeneic bone marrow transplantation and can occur in 30--60% of cases with HLA-DR matched donors) Prevention of GVHD in man can be achieved by T cell depleting donor marrow; however, this leads to other severe complications which include graft failure, rejection and an increased incidence of leukaemic relapse. 2 Over the last five to ten years, therefore, in several bone marrow transplant centres, including our own, assays to predict those patients at greatest risk of GVHD have been developed. Predicting GVHD would aid donor selection, and predict those patients who may benefit from T cell depletion of bone marrow and/or altemative GVHD prophylaxis or therapy. The assays used clinically to predict GVHD have included a mixed epidermal lymphocyte reaction, 3 the skin explant assay4,5 and, latterly, cytotoxic T lymphocyte precursor 6 (CTLp) and helper T lymphocyte precursor (HTLp) frequency assays. 7"s All the assays are based on a mixed lymphocyte or mixed lympho- cyte/epidermal reaction and the CTLp and HTLp assays utilize limiting dilution techniques and in the latter case measurement of interleukin 2 (IL-2). In recent years we have evaluated the skin explant assay, originally described by Vogelsang et al. who significantly predicted the outcome of GVHD in 27/32 patients with HLA identical bone marrow transplants. 4 The assay involves sensi- tizing donor lymphocytes in vitro in a primary mixed lympho- cyte reaction (MLR) and then evaluating the secondary response on patient skin biopsies by grading, histopatho- logically, the graft-versus-host reactivity. Using this assay we initially studied 13 allogeneic bone marrow transplant patients and the results correlated well with the clinical stage of skin GVHD post-transplant (correla- tion coefficient 0.86, p<0.001). 3 Recent studies incorporating a cohort of 34 patient-donor pairs have further demonstrated the usefulness of the assay in predicting acute GVHD. 9 Studies using the skin explant assay for both predicting GVHD and investigating the aetiology of the disease have demonstrated the role of the cytokines tumour necrosis factor alpha and interferon-gamma as important mediators of cel- lular damage in GVHR. 9"1° High levels of both cytokines in MLR supernatants correlated with the degree of GVHD in the patient or GVHR (caused by supernatant alone) in the skin explant assay. MLR supernatants from GVHD positive patient-donor pairs were also shown to induce HLA-DR Address for correspondence: Anne Mary Dickinson, Department of Haematology, The Royal Victoria Infirmary, Queen Victoria Road, Newcastle upon Tyne NEt 4LP, UK. expression on third party skin biopsies. Measurement of the production of cytokines in HLA matched MLC supernatants may therefore be a valuable indicator of incurring GVHD post transplant. Furthermore, recent data suggest that measurement, in limiting dilution assays, of antirecipient helper (IL-2 product- ing) HTLp frequency is of predictive value of GVHD in HLA-identical bone marrow transplant recipients. A similar assay based on the measurement of CTLp has been useful in predicting GVHD in matched unrelated donor transplants. This assay is not useful for predicting GVHD in allogeneic matched recipients due to the very low frequency of cytotoxic T cells pretransplant, but has been used to demonstrate cyotoxic T cell responses post-transplant in patients with GVHD against minor histocompatibility antigens. 11 In conclusion, it is now possible using several techniques to predict acute GVHD occurrence in 80% of HLA and DR matched allogeneic transplant recipients; these assays may, therefore, aid the development of new therapeutic strategies, possibly aimed specifically at anticytokine therapy, for those patients at high risk of developing GVHD. References 1 Glucksberg RP, Storb R, Fefer A et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HLA-matched sibling donors. Transplantation 1974; 18: 295-304. 2 Butturini A. New strategies for T cell depletion. Bone Marrow Transplant 1990; 6: 225-27. 3 Bagot M, Mary J-Y, Heslan Met al. The mixed epidermal cell lymphocyte-reaction is the most predictive factor of acute graft- versus-host disease in bone marrow graft recipients. Br J Haema- tol 1988; 70: 403--409. 4 Vogelsang GB, Hess AD, Berkman AW et aL An in vitro predictive test for graft versus host disease in patients with genotypic HLA-identical bone marrow transplants. N EnglJ Med 1985; 313: 645-50. 5 Sviland L, Dickinson AM, Carey PJ, Pearson ADJ, Proctor SJ. An in vitro predictive test for clinical graft versus host disease in allogeneic bone marrow transplant recipients. Bone Marrow Transplant 1990; 5: 105-109. 6 Kaminski E, Hows J, Man Set al. Prediction of graft versus host disease by frequency analysis of cytotoxic T cells after unrelated bone marrow transplantation. Transplantation 1989; 48: 608--13. 7 Theobald M, Nierle T, Bunjes D, Arnold R, Heimpel H. Host- specific interleukin-2-secreting donor T-cell precursors as pre- dictors of acute graft-versus-host disease in bone marrow transplantation between HLA-identical siblings. N Engl J Med 1992; 327: 1613. 8 Schwarer AP, Jiang YZ, Brookes PA et al. Frequency of anti- recipient alloreactive helper T-cell precursors in donor blood and Transplant Immunology 1994; 2:154-155