Studies on the mechanism of tolerance or graft-versus-host disease in allogeneic bone marrow recipients at the level of cytotoxic T-cell precursor frequencies.
Previous studies demonstrated that the frequency of donor-versus-host-reactive cytotoxic T-cell precursors (CTL-p) before allogeneic bone marrow transplantation (BMT) from matched unrelated donors correlates with the incidence of graft-versus-host disease (GvH-D). We investigated whether clinical manifestations of GvH-D after HLA-identical sibling BMT are accompanied by an increased frequency of minor histocompatibility antigen (HA)-specific CTL-p. We further asked whether changes of third-party-reactive CTL-p as a measure of overall immunocompetence are related to infectious complications frequently seen immediately after BMT. Eighteen patients (16 with an HLA-identical bone marrow graft, two with either one HLA-A or one HLA-DR mismatch) were studied. Limiting dilution analysis (LDA) was used to assess donor CTL-p frequencies against recipient pre-BMT, donor, and third-party targets in a follow-up study. Eight cases receiving HLA-identical marrow grafts never developed signs of GvH-D. Undetectable or very low frequencies (1/131,458) of minor HA-specific CTL-p were demonstrated pre-BMT. Two recipients, one of an HLA-A- and one of an HLA-DR-mismatched graft, exhibited low frequencies of recipient-specific CTL-p (1/66,920 and 1/85,577, respectively) before transplantation, which further decreased despite mild GvH-D grade I, or decreased within 3 months after grafting in the other case. Eight patients receiving HLA-identical grafts developed GvH-D. Recipient-specific CTL-p were less than 1/300,000 in five patients during limited GvH-D (four with grade I and one with grade II disease of the skin), but were detectable in three patients presenting with extensive GvH-D grades II to III and ranged from 1/7,993 to 1/210,108. The differences in post-BMT recipient-specific CTL-p frequencies between patients with GvH-D grades 0 to I (median, less than 1/300,000) and those with GvH-D grades II to III (median, 1/111,970) were statistically significant (P less than .05). Posttransplant lymphocytes from all 18 patients contained less than 1/300,000 CTL-p with specificity for donor targets. Comparison of third-party-reactive CTL-p frequencies between donor and post-BMT recipient lymphocytes showed a severe and long-lasting depletion subsequent to BMT, which was not related to infectious complications. Again, these differences reached the level of statistical significance (median CTL-p before BMT, 1/4,417; after BMT, 1/14,289; P less than .005).(ABSTRACT TRUNCATED AT 400 WORDS)