Long-term humoral immunity is mediated by short-lived plasma cells (replenished by memory B cells) and long-lived plasma cells. Their relative contributions are uncertain for immunity to SARS-CoV-2, especially given the widespread use of novel mRNA vaccines. Yet, this has far-reaching implications in terms of the need for regular booster doses in the general population and perhaps even revaccination in patients receiving B cell-depleting therapy. We aimed to characterise anti-SARS-CoV-2 antibody titres in patients receiving Rituximab following previous SARS-CoV-2 vaccination. We recruited 10 fully vaccinated patients (age: 16.9 ± 2.52 years) with childhood-onset nephrotic syndrome, not in relapse, receiving Rituximab for their steroid/calcineurin-inhibitor sparing effect. Antibodies to SARS-CoV-2 spike (S) and nucleocapsid (N) proteins were measured immediately prior to Rituximab and again ~6 months later, using the Roche Elecys® Anti-SARS-CoV-2 (S) assay. All ten patients were positive for anti-S antibodies prior to Rituximab, with six patients (60%) having titres above the upper limit of detection (>12,500 U/mL). Following Rituximab therapy, there was a reduction in anti-S titres (p = 0.043), but all patients remained positive for anti-S antibodies, with five patients (50%) continuing to have titres >12,500 U/mL. Six patients (60%) were positive for anti-N antibodies prior to Rituximab. Following Rituximab therapy, only three of these six patients remained positive for anti-N antibodies (p = 0.036 compared to anti-S seroreversion). Humoral immunity to SARS-CoV-2 is likely to be mediated in part by long-lived plasma cells.