Previous studies suggest stable renal transplant recipients can have either prednisone (P) or cyclosporine withdrawn; however, 30% of these patients undergo rejection requiring reinstitution of P or CsA. Some patients return to baseline creatinine levels, while others either stabilize at a higher creatinine level or lose their graft. It would be ideal to establish immunologically based criteria for selecting patients who can be successfully withdrawn or tapered from immunosuppression. We have investigated the development of donor antigen-specific hyporeactivity by using donor cells and/or homozygous typing cells defining the HLA-Dw specificities of the donor cells as stimulator cells in the mixed lymphocyte culture (MLC) and comparing the pre- and posttransplant responses of peripheral blood mononuclear cells from 199 kidney transplant recipients. Of these, 27% of the haploidentical living-related donor and 25% of the cadaver recipients developed in vitro donor antigen-specific hyporeactivity. The LRD recipients who did so have lower mean creatinine levels at 6, 12, and 24 months posttransplant (1.3, 1.3, and 1.2, respectively) than those who remained responsive to the donor antigens (1.6, 1.7, and 1.8) (P < 0.05). However, no differences in the mean creatinine levels were observed between CAD recipients who developed donor antigen-specific hyporeactivity and those who remained responsive. Rejection episodes were common in all groups in the first 3 months posttransplant; however, recipients who developed donor antigen-specific hyporeactivity tended to experience fewer rejection episodes after 3 months posttransplant. The percentage of recipients who remained rejection-free after 3 months post-transplant was 95% for CAD recipients who developed donor antigen-specific hyporeactivity vs. 83% for those who remained responsive. Only 1 hyporesponsive recipient (0/15 LRD; 1/20 CAD) developed chronic rejection vs. 12 (5/41, LRD; 7/60, CAD) recipients who remained responsive. For those with graft function at 3 months (when hyporesponsiveness was first determined), the actuarial 36-month graft survival was higher in the hyporesponsive (92%, LRD; 94%, CAD) than in responsive groups (LRD, 76%; CAD, 91%). No differences in the degree of HLA-DR mismatching (MM) were observed for the LRD hyporesponsive (1.20 DR MM) vs. reactive (0.98 DR MM) groups or for the CAD hyporesponsive (1.35 DR MM) vs. reactive (1.30 DR MM) groups. Donor antigen-specific hyporeactivity could not be determined (UND) for 30 of the LRD recipients and 33 of the CAD recipients due to lack of mismatching for DR antigens (0.03 DR MM and 0.25 DR MM, respectively).(ABSTRACT TRUNCATED AT 400 WORDS)