T-cell depletion of donor marrow decreases graft-versus-host disease resulting from transplants from unrelated and human leukocyte antigen (HLA)-mismatched related donors. However, there are diverse strategies for T-cell-depleted transplantation, and it is uncertain whether any improve leukemia-free survival (LFS). To compare strategies for T-cell-depleted alternative donor transplants and to compare T-cell depleted with non-T-cell-depleted transplants, we studied 870 patients with leukemia who received T-cell-depleted transplants from unrelated or HLA-mismatched related donors from 1982 to 1994. Outcomes were compared with those of 998 non-T-cell-depleted transplants. We compared LFS using different strategies for T-cell-depleted transplantation considering T-cell depletion technique, intensity of pretransplant conditioning, and posttransplant immune suppression using proportional hazards regression to adjust for other prognostic variables. Five categories of T-cell depletion techniques were considered: narrow-specificity antibodies, broad-specificity antibodies, Campath antibodies, elutriation, and lectins. Strategies resulting in similar LFS were pooled to compare T-cell-depleted with non-T-cell-depleted transplants. Recipients of transplants T-cell depleted by narrow-specificity antibodies had lower treatment failure risk (higher LFS) than recipients of transplants T-cell depleted by other techniques. Compared with non-T-cell-depleted transplants (5-year probability +/- 95% confidence interval CI of LFS, 31% +/- 4%), 5-year LFS was 29% +/- 5% (P = NS) after transplants T-cell depleted by narrow-specificity antibodies and 16% +/- 4% (P <.0001) after transplants T-cell depleted by other techniques. After alternative donor transplantation, T-cell depletion of donor marrow by narrow-specificity antibodies resulted in LFS rates that were higher than those for transplants T-cell depleted using other techniques but similar to those for non-T-cell-depleted transplants. (Blood. 2000;95:3996-4003)