Donor non-specific IFN-γ production by primed alloreactive cells as a potential screening test to predict the alloimmune response

In order to devise an in vitro experimental system that predicts the in vivo generation of T lymphocytes capable of initiating the rejection process and thereby to individualize the immunosuppressive strategy in a rational way, we studied the in vitro strength of alloresponses to non-specific donors as a surrogate tool to identify patients with heightened alloimmunity. We measured interferon gamma. (IFN-gamma) produced by primed alloactivated peripheral blood lymphocytes (p-allo-PBL) against third party stimulator PBLs in four groups by the enzyme linked immunospot (Elispot) assay: 16 with excellent renal transplant function (group 1); nine with chronic rejection (group 2); 11 allo-sensitized (PRA>60%) by graft loss on dialysis (group 3) and 36 normal controls (group 4). The Elispot assay was performed using 10(6) irradiated stimulator PBLs and 10(5) responder PBLs for 24 and 48 h. Each responder was challenged by 2-4 independent stimulators. Results: At 24 h, mean+/-S.D. and 95% confidence intervals (CI) of spots/10(5) responder cells were 12.8+/-8.7 (10-15.5); 57.8+/-116 (11.9-103.7); 77.5+/-91.3 (39.8-115.2); and 20.7+/-17.9 (17.4-24.1) in groups 1-4, respectively. P<0.01 between groups I or 4 vs. 2 or 3. An arbitrary spot level of greater than or equal to30 has positive and negative predictive values of 58% and 95%, respectively, sensitivity of 94% and specificity of 65% to identify patients with enhanced immunity. Conclusion: Chronic allogeneic stimulation is associated with enhanced p-allo-PBL. IFN-gamma producing frequencies against third party stimulators. Significant variation in IFN-gamma spots produced by p-allo-PBL may be useful to choose less allogeneic donors. Diminished p-allo-PBL alloresponse to third party stimulators may predict transplant patients with decreased alloresponses who may benefit from lesser immunosuppressive drugs. (C) 2003 Elsevier B.V. All rights reserved.