Background: It has been reported that preformed donor-specific anti-human leukocyte antigen antibodies (DSAs) were associated with poor graft outcomes in transplant recipients.  To eliminate preformed DSAs, prior to either kidney or liver transplantation, we desensitized DSA-positive patients by injecting rituximab and then performed plasmapheresis in those patients using a protocol for ABO-incompatible (ABO-I) transplant recipients.  B-cell depletion with rituximab may influence T-cell responses to alloantigens since B-cells are effective antigen-presenting cells that are capable of activating donor-specific T-cells.  It is also likely that the cytokine release syndrome, which is caused by rituximab, may enhance T-cell activation.  In this study, we investigated the impact of pre-transplant desensitization with rituximab on the subsequent response of T-cells to alloantigens in DSA-positive kidney and liver transplant recipients. 
Methods: Twenty DSA-positive patients of kidney or liver allografts were employed in this study, while thirty-eight DSA-negative ABO-I patients of kidney or liver allografts formed the control group that received a similar desensitization preconditioning with rituximab and plasmapheresis.  To monitor the patients’ immune status, a mixed lymphocyte reaction assay using a carboxyfluorescein diacetate succinimidyl ester-labeling technique was performed before and after desensitization with the consent of the patinets, donors, and third-party healthy volunteers.
Results: Before desensitization, there were no significant differences in the average stimulation index (SI) values for the responses of the CD4+ and CD8+ T-cells to donor stimuli between the two groups.  However, after desensitization, the average SI values for the CD4+ T-cells were significantly higher in the DSA-positive patients than those in the DSA-negative ABO-I patients.  Consistently, in liver transplant recipients, the incidence of acute cellular rejection (ACR) was higher in DSA-positive patients than that in the DSA-negative ABO-I patients (60% and 16.7%, respectively), and none suffered from antibody-mediated rejection. Treatment with either steroid-bolus or rabbit thymocyte globulin (rATG) completely cured ACR.
Conclusions: These findings demonstrate that B-cell depletion with rituximab exacerbates anti-donor CD4+ T-cell responses in DSA-positive organ transplant recipients, and that it is probably associated with a high rate of ACR events.  It might be possible that treatment with rituximab undesirably depletes specific B-cell subsets that could regulate T-cell responses to alloantigens.  Although this issue is currently under investigation, T-cell responses could possibly be inhibited by rATG before or after desensitization to prevent ACR events.