Determinants of the duration of B-cell depletion after rituximab in a pediatric population

Rituximab is used for a number of indications in pediatric nephrology, including the prevention of relapse in steroid-dependent nephrotic syndrome (SDNS). Clinical efficacy of rituximab is dependent on B-cell depletion, delayed reconstitution of B cells is protective of relapse in SDNS [1], and repeat courses of rituximab for SDNS or calcineurin inhibitor–dependent nephrotic syndrome decreases the risk of relapse [2], but may increase the risk of rituximab-associated adverse effects including hypogammaglobulinemia and infusion reactions [3]. In children, duration of B-cell depletion is highly variable across individuals [4]. It would therefore be helpful to predict the duration of B-cell depletion and, thus, optimize the administration scheme. However, the reasons for prolonged B-cell depletion after rituximab therapy are understudied. In this retrospective study, we included all patients followed in pediatric nephrology at CHU Sainte-Justine who were treated with rituximab between March 2006 and July 2022 for nephrological indication. We used the term “infusion” for each infusion of rituximab, and the term “course” for each course of treatment which can consist of one to four infusions over a period of 1–4 weeks. We collected clinical data from the medical record, including age, sex, body surface area, indication, and doses and dates of rituximab treatment and concomitant immunosuppressive medications. Peripheral blood immunophenotyping was routinely assessed at repeated occasions after rituximab courses, allowing estimation of the duration of B-cell depletion which was defined as the time between the last rituximab infusion of the course and the first detection of CD20+ cells in the peripheral blood. Descriptive characteristics are shown as medians [interquartile range (IQR)] or n (%) and were compared using a Mann–Whitney U test or a Fisher’s exact test. Probabilities of maintaining CD20+ cells depletion were calculated using the Kaplan–Meier method, and hazard ratios (HR) with 95% confidence intervals (95% CI) were calculated using a Cox proportional hazards models with mixed effects, using patient ID as a random effect, which was performed unadjusted and with adjustment for rituximab dose as a fixed effect. The Cox proportional hazards assumption was met. Analyses were performed using the survival and the coxme packages of R version 3.6.3. Ethical approval for chart review was obtained from the Sainte Justine Hospital ethical research committee (approval number 2020-2729). Patient consent was not required for this study. Eighty-nine patients were included, median (IQR) age was 11.0 (6.9, 14.1) years. Sixty-six (75%) were being treated for SDNS, 6 (7%) for humoral rejection after kidney transplantation and the rest were treated for other inflammatory diseases. Nineteen (22%) patients were also treated with mycophenolate mofetil, with missing data for mycophenolate mofetil for two patients. Table 1 shows the clinical characteristics of patients at the time of their first rituximab course, according to mycophenolate mofetil treatment. A total of 246 rituximab courses were analyzed, while courses with no detectable CD20+ cells at rituximab infusion (n = 17) were excluded from the analysis. Supplementary data, Table S1 shows the HR of B-cell depletion over time according to clinical characteristics. Factors significantly associated with longer duration of CD20+ cells depletion were higher dose of rituximab [HR 0.94 (95% CI 0.89, 0.99), P = .022] and concomitant treatment with mycophenolate mofetil [HR 0.52 (95% CI 0.34, 0.80), P = .003] (Fig. 1A). Median CD20+ cells depletion time was 173 (95% CI 162, 185) days in patients with no mycophenolate mofetil and 218 (95% CI 197, 250) days in patients with mycophenolate mofetil (P = .0046). Use of mycophenolate mofetil remained associated with the duration of CD20+ cells depletion after adjustment for cumulative dose of rituximab. The duration of CD20+ cells depletion was similar in patients receiving one or more than one infusion per course (Fig. 1B), between first and additional courses (Fig. 1C), and between males and females (Fig. 1D). Similar results were found when restricting the analysis to patients with SDNS only (Supplementary data, Table S2). A longer duration of CD20+ cells depletion was associated with a non-significant decrease in the risk of clinical relapse in patients with nephrotic syndrome [HR for 10 additional days of B-cell depletion, 0.95 (95% CI 0.90, 1.00), P = .062]. B-cell depletion duration after rituximab course. Kaplan–Meier curves showing the duration of B-cell depletion in patients receiving rituximab (line) with 95% CI (colored area) according to (A) mycophenolate mofetil (MMF) status (no MMF, full orange line; MMF, dashed blue line); (B) number of infusions (single infusion, full pink line; multiple infusions, dashed blue line); (C) number of previous courses (first course, full green line; additional courses, dashed red line); (D) sex (male, full blue line; female, dashed red line). P-values and 95% CI were calculated using a Cox model. Descriptive characteristics at baseline. Descriptive characteristics at baseline according to concomitant use of mycophenolate mofetil. Values are shown as n (%) or median (IQR). P-values were calculated using the Fisher’s exact test (for qualitative variables) or the Mann–Whitney U test (for quantitative variables). Two patients are missing from this table due to unknown mycophenolate mofetil treatment status. Descriptive characteristics at baseline. Descriptive characteristics at baseline according to concomitant use of mycophenolate mofetil. Values are shown as n (%) or median (IQR). P-values were calculated using the Fisher’s exact test (for qualitative variables) or the Mann–Whitney U test (for quantitative variables). Two patients are missing from this table due to unknown mycophenolate mofetil treatment status. There are, to our knowledge, very few data regarding the assessment of the characteristics associated with B-cell depletion duration in pediatric patients receiving rituximab. Delbet et al. found that while B-cell depletion was highly variable across individuals, the B-cell depletion time after the first course predicted depletion time for additional courses [4]. We did not confirm these results in our study, using a statistically different approach. Further, we found that B-cell depletion time after rituximab infusion was similar between the first and additional courses and between single and multiple infusions of rituximab. We found, however, that the use of mycophenolate mofetil was independently associated with the duration of B-cell depletion, in addition to the dose of rituximab as previously published [5]. In line with our findings, Hogan et al. have shown that the initial dose of rituximab was associated with B-cell depletion duration, impacting the risk of relapse [5]. The association with mycophenolate mofetil warrants further investigation, but may be explained by the direct effect of this drug on B cells. B cells are reliant on de novo synthesis of purines. Mycophenolate mofetil inhibits the inducible type II isoform inosine monophosphate dehydrogenase (IMPDH), which is preferentially expressed on activated T and B lymphocytes [6]. Specifically, mycophenolate mofetil has been shown to block the expansion of naïve and memory B cells as well as the terminal differentiation of B cells [7], and to reduce their secretion of interleukin-6, an important cytokine for B cell proliferation and survival [8]. The potent cytostatic effects of mycophenolate mofetil on B cells may therefore explain the delayed reconstitution of B cells following rituximab. Multiple rituximab courses may be time-consuming for the patient and cumbersome for the health system. Prediction of B-cell depletion duration may help avoiding unnecessary rituximab administrations, which may lead to increased cost of treatment and risk of serious adverse events, as we have previously shown that rituximab was associated with a significant number of serious adverse events in children [9]. Use of mycophenolate mofetil after rituximab has in fact been suggested for the prevention of relapse in SDNS [10]. Limitations to our study include the lack of standardized times at which peripheral blood immunophenotyping was performed, which was due to the fact that this determination was performed at each patient’s visits to the clinic. However, our center's policy to routinely perform CD20+ count prior to rituximab administration allowed us to provide repeated CD20+ determinations for each patient and hence a reliable estimate of CD20+ depletion duration. Further, given the relatively low number of patients with SDNS receiving mycophenolate mofetil in our study, we could not assess whether treatment with mycophenolate mofetil was associated with an increased risk of SDNS relapse. Race of study participants was not available, thereby limiting the generalizability of our findings. Last, we did not monitor the pharmacokinetics of mycophenolate mofetil, which does not allow to provide a target mycophenolic acid window in patients receiving rituximab in order to obtain a more prolonged response. In conclusion, we found that only the dose of rituximab and the concomitant use of mycophenolate mofetil were independently associated with B-cell depletion duration. The identification of factors associated with a prolonged duration of B-cell depletion can be useful to optimize administration of rituximab treatment. We would like to thank the nurses and physicians from CHU Sainte-Justine nephrology department. This work was supported by a Fonds de recherche du Québec—Santé (FRQS)/Fondation des Étoiles fellowship award to A.F. R.A., A.-L.L. and A.F. conceived the study. R.A., A.-L.L. and A.F. designed and performed the analysis, and wrote the paper. R.A., C.L. and E.D. collected the data. R.L., A.C. and E.D. participated in the interpretation of the results. All authors revised the manuscript. The authors have no conflicts of interest to disclose. Results presented in this paper have not been published previously in whole or part, except in abstract format. Original data and code available upon request to the corresponding author.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.