The induction of immunologic unresponsiveness to improve renal allograft survival was attempted in 64 patients by the pretransplant administration of donor-specific whole blood or buffy coat in conjunction with continuous azathioprine immunosuppression. All donor/recipient combinations were at least one-haplotype-disparate. Presensitization, defined as a positive Amos or antiglobulin crossmatch or a high-titer (greater than 1:8) B-cell-positive crossmatch, was present in 6 patients and not present in 58 patients. Attempts at desensitization of the already sensitized group were uniformly unsuccessful. Treatment of the 58 nonpresensitized patients resulted in transient sensitization in 2 patients, permanent sensitization in 1 patient, and no evidence of sensitization in 55 patients. Fifty-three patients underwent renal transplantation from the specific blood donor, and only two have experienced renal allograft rejection loss during a mean follow-up period of 22 months (5-45 months); 57% have never experienced a rejection episode. The two-year renal allograft survival rate was 85%. This is significantly (P less than 0.01) better than our historical experience of 64% with one-haplotype living-related transplants. The low rate of sensitization (5%) has permitted almost all patients to undergo eventual renal transplantation from the specific blood donor. This and the low rate of rejection (4%) argues for a modification of the immunologic response, rather than a selecting-out process as the mechanism for improved allograft survival.