BACKGROUND: Graft-versus-host disease (GVHD) is a major and sometimes fatal complication of allogeneic bone marrow transplantation (BMT). The prediction of GVHD remains an important issue in preventing morbidity and mortality after allogeneic BMT. In the past 10 years, there has been great interest in using the frequency analysis of alloreactive helper and cytotoxic T lymphocyte precursors (HTLp and CTLp) to detect recipient-specific alloreactivity and thus predict GVHD in HLA-matched related and unrelated BMT. However, the results remain controversial. The intention of the present study was to investigate whether alloreactive HTLp and CTLp frequencies measured in donor peripheral blood before BMT would be a useful predictor for the occurrence of acute GVHD after HLA-matched sibling BMT. METHOD: A combined limiting dilution assay was used to determine alloreactive HTLp and CTLp frequencies for 42 HLA-matched sibling patient/donor pairs. The pretransplantation host-reactive HTLp and CTLp frequencies were then correlated with post-transplantation clinical outcomes of acute GVHD. The association between HTLp/CTLp frequencies and the incidence of acute GVHD was determined using the Fisher's exact test. RESULTS: The mean values of HTLp and CTLp frequencies for this cohort of HLA-matched sibling patient/donor pairs were 1:321,322 (range 1:71,000 to 1:1000,000) and 1:195,260 (range 1:3,717 to 1:1000,000), respectively. Acute GVHD (> or =II) was observed in one of four patients with high (>1:100,000) HTLp frequencies and 20 of 36 patients with low (<1:100,000) HTLp frequencies. Similarly, 6 of 10 patients with high (>1: 100,000) CTLp frequencies and 14 of 29 patients with low (<1:100,000) CTLp frequencies developed acute GVHD (> or =II). The overall correlation between hostreactive HTLp/CTLp frequencies and the incidence of acute GVHD in this cohort of patients was 42.5% and 53.9%, respectively. There was no significant difference in the incidence of acute GVHD between the patients with either high or low host-reactive HTLp/ CTLp frequencies (P=0.331 and 0.716, respectively). The data were also analyzed separately for the adult patient group based on GVHD prophylaxis with either cyclosporine alone or the combination of cyclosporine and methotrexate. Within these two prophylaxis groups, neither HTLp nor CTLp frequencies correlated with acute GVHD. CONCLUSION: Host-reactive HTLp and CTLp frequency analysis did not provide informative prediction for the occurrence of acute GVHD after HLA-matched sibling BMT.
BACKGROUND:Graft-versus-host disease (GVHD) is a major and sometimes fatal complication of allogeneic bone marrow transplantation (BMT). The prediction of GVHD remains an important issue in preventing morbidity and mortality after allogeneic BMT. In the past 10 years, there has been great interest in using the frequency analysis of alloreactive helper and cytotoxic T lymphocyte precursors (HTLp and CTLp) to detect recipient-specific alloreactivity and thus predict GVHD in HLA-matched related and unrelated BMT. However, the results remain controversial. The intention of the present study was to investigate whether alloreactive HTLp and CTLp frequencies measured in donor peripheral blood before BMT would be a useful predictor for the occurrence of acute GVHD after HLA-matched sibling BMT. METHOD: A combined limiting dilution assay was used to determine alloreactive HTLp and CTLp frequencies for 42 HLA-matched sibling patient/donor pairs. The pretransplantation host-reactive HTLp and CTLp frequencies were then correlated with post-transplantation clinical outcomes of acute GVHD. The association between HTLp/CTLp frequencies and the incidence of acute GVHD was determined using the Fisher's exact test. RESULTS: The mean values of HTLp and CTLp frequencies for this cohort of HLA-matched sibling patient/donor pairs were 1:321,322 (range 1:71,000 to 1:1000,000) and 1:195,260 (range 1:3,717 to 1:1000,000), respectively. Acute GVHD (> or =II) was observed in one of four patients with high (>1:100,000) HTLp frequencies and 20 of 36 patients with low (<1:100,000) HTLp frequencies. Similarly, 6 of 10 patients with high (>1: 100,000) CTLp frequencies and 14 of 29 patients with low (<1:100,000) CTLp frequencies developed acute GVHD (> or =II). The overall correlation between hostreactive HTLp/CTLp frequencies and the incidence of acute GVHD in this cohort of patients was 42.5% and 53.9%, respectively. There was no significant difference in the incidence of acute GVHD between the patients with either high or low host-reactive HTLp/ CTLp frequencies (P=0.331 and 0.716, respectively). The data were also analyzed separately for the adult patient group based on GVHD prophylaxis with either cyclosporine alone or the combination of cyclosporine and methotrexate. Within these two prophylaxis groups, neither HTLp nor CTLp frequencies correlated with acute GVHD. CONCLUSION: Host-reactive HTLp and CTLp frequency analysis did not provide informative prediction for the occurrence of acute GVHD after HLA-matched sibling BMT.