CYTOTOXIC T-LYMPHOCYTE PRECURSOR FREQUENCY ANALYSES IN BONE-MARROW TRANSPLANTATION WITH VOLUNTEER UNRELATED DONORS - VALUE IN DONOR SELECTION

Between May 1989 and February 1994, we performed 48 volunteer unrelate d donor BMTs for first chronic phase chronic myeloid leukemia using in vivo T cell depletion for acute graft-versus-host disease (aGvHD) pro phylaxis. In 40 cases, adequate material was available to measure the frequency of antirecipient MRC cytotoxic T lymphocyte precursor (CTLp) cells in the blood of potential donors. This supplemented standard se rological typing, one-dimensional isoelectric focusing for class I pro teins, and allogenotyping for DR and DQ alleles using DNA RFLP analysi s in the donor selection process. All recipients were conditioned with cyclophosphamide 120 mg/kg, TBI 1320 cGy, and intravenous Campath 1G. GvHD prophylaxis consisted of CsA, short-course methotrexate, and int ravenous Campath 1G. Minimum follow-up in all surviving recipients was 100 days. The development of aGvHD and the probability of leukemia-fr ee survival were compared between the high frequency group (CTLp>1 in 100,000) (n=15) and the low frequency group (CTLp<1 in 100,000) (n=25) . There was a trend for increasing grade of aGvHD, which was statistic ally significant in the high frequency group when compared with the lo w frequency group (P=0.003), Both a high frequency of CTLp (relative r isk [RR] = 9.0, P=0.016) and HLA mismatch (RR=6.7, P=0.023) mere predi ctors of severe aGvHD (grade III or IV). Multivariate analysis showed that CTLp group (RR=3.4, P=0.015) and CMV status (RR=3.9, P=0.008) mer e predictors of leukemia-free survival. Further investigation showed a n interaction between the two, such that CMV seropositive recipients i n the high frequency group had a relative risk of 9.4 (P=0.0001) of tr eatment failure (death or relapse) when compared with other combinatio ns. We conclude that with our present GvHD prophylaxis regimen, CTLp f requency analysis predicts post-BMT outcome and is a valuable aid in d onor selection.