CD4 + and CD8 + T-lymphocyte number as predictive marker of relapse after rituximab treatment in childhood-onset refractory nephrotic syndrome

Background Rituximab is a promising option for refractory idiopathic nephrotic syndrome. However, no simple predictive markers for relapse after rituximab have been established. To determine such markers, we investigated the relationship between CD4 + and CD8 + cell counts and relapse after rituximab administration. Methods We retrospectively investigated patients with refractory nephrotic syndrome who received rituximab followed by immunosuppressive as maintenance therapy. Patients were divided into no relapse in 2 years after rituximab treatment or relapse group. After rituximab treatment, CD4 + /CD8 + cell counts were measured monthly, at prednisolone discontinuation, and at B-lymphocyte recovery. To predict relapse, these cell counts were analyzed using receiver operating characteristic (ROC). Additionally, relapse-free survival was reevaluated based on the result of ROC analysis for 2 years. Results Forty-eight patients (18 in the relapse group) were enrolled. At prednisolone discontinuation (52 days after rituximab treatment), the relapse-free group showed significantly lower cell counts than the relapse group (median CD4 + cell count: 686 vs. 942 cells/µL, p = 0.006; CD8 + : 613 vs. 812 cells/µL, p = 0.005). In the ROC analysis, CD4 + cell count > 938 cell/µL and CD8 + cell count > 660 cells/µL could predict relapse in 2 years (sensitivity, 56% and 83%; specificity, 87% and 70%). The patient group with both lower CD4 + and CD8 + cell counts showed significantly longer 50% relapse-free survival (1379 vs. 615 days, p