This study is a long-term observational follow-up of a previously published pediatric cohort with INS treated with anti-CD20 therapy (rituximab or ofatumumab) at Ospedale Pediatrico Bambino Gesù. Inclusion criteria were as follows: frequently-relapsing (defined as ≥2 consecutive relapses within 6 months or ≥3 relapses within any 12-month period) or steroid-dependent (defined as ≥2 consecutive relapses during prednisone treatment or within 14 days of its discontinuation) nephrotic syndrome (13); a minimum follow-up >3 years from the last anti-CD20 infusion; available data on circulating B-cell subsets and on serum immunoglobulin levels. Among the 27 patients included in the original study, patients who had transitioned to adult care (n = 9) or lacked updated data on B-cell subsets or serum immunoglobulin levels (n = 5) were excluded. The final analysis comprised 13 patients with complete longitudinal data (Figure 1).

All patients had steroid-dependent nephrotic syndrome. Relapse was defined as proteinuria ≥ 3+ for at least three consecutive days on urine dipstick testing (13).

Patients received a single infusion of anti-CD20 therapy, followed by a second infusion administered seven days later in cases of incomplete B-cell depletion (CD19⁺ B cells > 10 cells/μL of peripheral blood lymphocytes). All patients initially received rituximab (375 mg/m²). Among those who underwent multiple infusions (≥ 2), two patients received ofatumumab (1500 mg/1.73 m²) as their final anti-CD20 treatment.

Anti-CD20 therapy was administered during steroid-induced remission, and repeat courses were given only in the event of relapse, except for two patients who experienced rapid B-cell recovery (within 1 and 3 months, respectively). Demographic and clinical characteristics, number of relapses, infectious episodes, concurrent immunosuppressive therapies (prednisone, mycophenolate mofetil, and calcineurin inhibitors), number and type of anti-CD20 infusions, and use of immunoglobulin replacement therapy were recorded. Concomitant immunosuppressive agents were tapered or discontinued following anti-CD20 administration unless relapse occurred.

Blood samples were collected in accordance with institutional guidelines for informed consent, following approval by the local Ethics Committee and in compliance with the Declaration of Helsinki. Samples were obtained at baseline (time of the first anti-CD20 infusion), 2–7 days post-infusion, at first follow-up (>2 years after the last infusion), and at extended follow-up (>3 years after the last infusion). Peripheral blood mononuclear cells (PBMCs) were isolated using Ficoll-Paque Plus (Amersham Biosciences) density-gradient centrifugation.

PBMCs were stained with fluorochrome-conjugated monoclonal antibodies targeting CD19, CD24, CD27, CD38, IgD, and IgM (BD Biosciences) and analyzed by multicolor flow cytometry (FACS BD LSRFortessa, BD Biosciences) (14). Subsets of gated CD19⁺ B cells were identified based on surface marker expression as follows: transitional (CD38^highCD24^high) and mature-naïve (CD38^intermediateCD24^low) B cells, expressed as absolute counts. Memory B cells were defined as CD19⁺CD27⁺ cells, and memory subsets were classified as IgM memory (IgM⁺IgD^intermediate) or switched memory (IgM^-IgD^-), as previously described (14). Data was analyzed using FACSDiva software, with 50,000 live lymphocyte events gated per sample. Age-related reference ranges were based on previously published data (15).

Total serum IgG, IgA, and IgM concentrations, as well as anti-tetanus and anti-HBV IgG titers, were measured as part of routine diagnostics in the clinical laboratory of the Ospedale Pediatrico Bambino Gesù. Immunoglobulin deficiency detected at the last follow-up but absent at baseline was classified as “de novo”.

Clinical characteristics of the 13 included patients (7 males) are summarized in Table 1. Data collected at baseline (time of first anti-CD20 administration) and at the first long-term follow-up >2 years from last anti-CD20 infusion (mean: 5.4 years; range: 2.1–7.1 years after anti-CD20 infusion) were previously reported in the original study (4). At the first follow-up, most patients had received one (n = 6) or two (n = 6) courses of anti-CD20 therapy, while one patient had received five courses.

At the extended follow-up (mean: 6.6 years after the previous evaluation), two patients had received a second anti-CD20 retreatment, two a third, and another a fourth and fifth retreatment, all administered in response to disease relapse (Table 1). The extended follow-up included data collected >3 years after the last anti-CD20 infusion (range: 3.2–16.0 years) and more than 9 years after the first infusion (range: 9.0–16.7 years).

From a clinical perspective, four patients who were already in sustained remission at the first long-term follow-up remained relapse-free throughout the extended follow-up. In contrast, eight of nine patients who had relapsed by the first follow-up continued to experience relapses despite additional anti-CD20 courses. However, one patient who had a single relapse one year after the first anti-CD20 course subsequently achieved long-term remission, which persisted without further anti-CD20 treatment up to the extended follow-up (6.6 years after the previous evaluation).

Importantly, median duration of concomitant oral immunosuppression after first anti-CD20 infusion was 10 years (range 1–14 years) and a progressive reduction in oral immunosuppressive therapy was observed over time. At the first long-term follow-up, three patients had successfully discontinued all immunosuppressive agents. This improvement became even more pronounced at the extended follow-up, when six patients were completely off immunosuppressive therapy for more than 3 years since the discontinuation of the last immunosuppressive drug (range: 3–15 years). Moreover, none of the remaining patients required triple therapy or the combination of prednisone with calcineurin inhibitors, and only a few continued on single or dual regimens with prednisone (n = 1), mycophenolate mofetil (n = 4), or their combination (n = 2) (Table 1).

Longitudinal changes in circulating B-cell subsets are shown in Figure 2. Normalization of total CD19⁺ and mature naïve B-cell counts was observed at both the first and extended follow-ups (Figures 2A, C). Transitional B cells, which were significantly increased at the first long-term follow-up, returned to baseline levels - within age-related normal ranges - at the extended follow-up (Figure 2B). In contrast, the marked reduction in total memory, IgM memory, and switched memory B-cell counts observed at the first long-term follow-up persisted at the extended follow-up (Figures 2D–F).

Of note, among all B-cell subsets analyzed, patients who remained relapse-free after the last anti-CD20 course showed significantly lower median levels of switched memory B cells compared with relapsing patients (median [IQR]: 0.48 [0.13–2.16] vs. 7.68 [2.07–28.28], p < 0.05).

Longitudinal changes in serum immunoglobulin concentrations are shown in Figures 2G–I. Median serum IgG levels significantly increased at the extended follow-up (Figure 2G), although they remained below the normal range in six patients (Table 1). In contrast, serum IgA and IgM levels did not show significant changes throughout the entire follow-up period (Figures 2H, I). However, four patients developed de novo severe hypogammaglobulinemia (defined as serum IgG < 160 mg/dL), irrespective of the number of anti-CD20 treatment courses (one course in two patients, two in one patient, and five in one patient), detected already at first long-term follow-up. Two of these patients were the youngest and the only below 6 years old at the time of the first anti-CD20 infusion. All four patients required initiation of immunoglobulin replacement therapy (0.2-0.3 g/Kg every 2–4 weeks, adjusted based on clinical efficacy) (Figure 2G). Attempts to discontinue replacement therapy during the extended follow-up period were unsuccessful, due to marked decline in IgG levels and/or the occurrence of infections upon discontinuation of immunoglobulin replacement. Consequently, all four patients remained on immunoglobulin replacement therapy more than four years after its initiation (range: 4.5–10.4 years), which likely masked their true endogenous serum IgG concentrations at the extended follow-up (Figure 2G, Table 1).

Additionally, de novo severe hypo-IgA was detected in four patients at the first long-term follow-up (in association with severe hypo-IgG in two cases) and in one additional patient at the extended follow-up (Figure 2H, Table 1).

Despite vaccination against tetanus and HBV according to national immunization schedules before anti-CD20 administration, and re-immunization in seven patients after the last anti-CD20 infusion, most patients showed impaired vaccine competence based on vaccine-specific IgG titers at baseline and at the first long-term follow-up (Table 1). The significant increase in the number of patients protected against tetanus observed at the extended follow-up could be attributable to immunoglobulin replacement therapy in four patients, as described above for total IgG levels, and to re-immunization in one patient (Table 1).

Infectious events were reported in four patients at the first follow-up and in seven patients at the extended follow-up (Table 1). Among them, frequent (monthly during the winter season) respiratory tract infections were reported in four patients, and pneumonia and an HZV infection requiring hospitalization were reported in two different patients.