Predictors of relapse and long-term outcome in children with steroid-dependent nephrotic syndrome after rituximab treatment.

Background: In patients with complicated steroid-dependent nephrotic syndrome (SDNS), rituximab (RTX) followed by immunosuppressive agent (IS) can maintain remission without the use of prednisolone (PSL). However, available data on the predictive factors for relapse and the long-term outcome after this protocol are few.
Methods: We retrospectively analyzed 43 SDNS patients who were followed-up for a long time (>2 years, mean 5.4 years) after a single dose of RTX (375 mg/m ² ) from September 2007. After RTX, PSL was tapered off within 6 months; monotherapy with IS, such as cyclosporine or mycophenolate mofetil, was continued to prevent post-RTX relapse. For patients who achieved >12 months of PSL-free remission, IS was also tapered off.
Results: Thirty-nine patients (91 %) could discontinue PSL without relapses at a median of 154 days after the initial RTX. The first relapse of NS occurred in 39 patients (91 %) at a median of 586 days; additional RTX doses were administered in 28 patients (65 %). Kaplan-Meier analysis showed that shorter CD19 cell depletion (<150 days) and younger age at RTX initiation (<12.5 years) were significantly associated with high risk for first relapse after RTX (log rank p < 0.05). In multivariate analysis, mycophenolate mofetil therapy as maintenance IS after RTX was the only predicted risk factor for first relapse (hazard ratio 2.75; p = 0.027). At the last follow-up, IS was still used in 33 patients (77 %); treatment-free remission (>12 months) was achieved in only five patients (12 %).
Conclusions: The introduction of RTX may not be necessarily associated with improved long-term outcome.