Positioning rituximab in the therapeutic strategies for SLE

Introduction

Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease, potentially affecting any organ or system with a variant clinical phenotype. B cells are thought to play a significant role in disease pathogenesis. B-cell auto- and over-reactivity in patients with lupus have various functions, such as autoantibody production, antigen (and autoantigen) presentation to T cells, cytokine secretion, and expression of toll-like receptors (TLRs). Therefore, targeting B cells in patients with SLE represents a reasonable therapeutic approach. In this review, we discuss when, how, and in which patient subgroups rituximab treatment may represent a reasonable approach.

Pathophysiological rationale and mechanism of action

The central event in SLE pathogenesis is the loss of self-tolerance with polyclonal activation of B cells (1–4). B-cell differentiation into plasmablasts and plasma cells results in the production of different autoantibodies, which in turn bind to self-antigens, forming immune complexes (ICs). These ICs deposit in tissues, such as the kidneys, joints, and skin, and activate the complement system, causing progressive organ damage. Several quantitative and qualitative B-cell defects have been described. B cells from patients with lupus exhibit increased synthesis of pro-inflammatory cytokines, such as tumor necrosis factor, interleukin-6 (IL-6), and interferon-γ (IFNγ), and regulatory cytokines, such as IL-10, thereby further amplifying immune dysregulation. Autoreactive B cells present autoantigens to T cells, perpetuating the cycle of B-cell activation. In addition, B-cell activation is further enhanced by toll-like receptor (TLR) engagement with nucleic acid–protein complexes, which represent the most prominent autoantigens in patients with SLE (5).

Rituximab (RTX) is a chimeric monoclonal antibody (mAb) against CD20, a cell-surface marker on pre-B and mature B cells (6). Among the different potential mechanisms of action leading to B-cell destruction, complement-dependent cytotoxicity (CDC) classifies RTX as a type I monoclonal antibody (mAb). RTX consists of a murine variable region and human constant regions, forming an immunoglobulin G1 (IgG1) molecule designed to enhance precision targeting and immune effector function. RTX is also characterized by the ability to cluster CD20 into lipid rafts. Rouge et al. obtained a structure of CD20 in complex with RTX, illustrating CD20 as a compact double-barrel dimer bound by two RTX antigen-binding fragments (Fabs) and engaging a different composite epitope and an extensive homotypic Fab: Fab interface (7). It has been speculated that RTX cross-links CD20 into circular assemblies, resulting in the activation of the complement cascade.

Evidence from randomized, controlled trials, and observational and real-world evidence studies

A couple of double-blind, randomized, placebo-controlled trials, namely, EXPLORER in patients with SLE and LUNAR in patients with lupus nephritis (LN), did not meet their primary endpoints (8, 9). These results raised concerns regarding the study designs (Table 1). The EXPLORER trial included patients with moderately to severely active extra-renal SLE; the inclusion of different disease manifestations made assessment of a potential uniform improvement difficult. Additionally, in both trials, the patients received intensive background immunosuppressive treatment. In the EXPLORER trial, patients received background immunosuppressants, along with high doses of oral prednisone (0.5–1.0 mg/kg/day). In the LUNAR trial, patients were treated with high-dose corticosteroids and 3 g/day of mycophenolate mofetil (MMF). Such robust background treatment may have masked the potential therapeutic benefit of RTX.

Table 1. Studies on RTX in patients with lupus.

CS, corticosteroids; HCQ, hydroxychloroquine; AM, antimalarials; AZA, azathioprine; MTX, methotrexate; MMF, mycophenolate mofetil; CYC, cyclophosphamide; MZR, minoribine; ABT, abatacept; IVIg, intravenous immunoglobulin; SLEDAI-2K, SLE Disease Activity Index 2000; SELENA SLEDAI, Safety of Estrogens in Lupus Erythematosus National Assessment-SLE Disease Activity Index; BILAG, British Isles Lupus Assessment Group; CR, complete response; PR, partial response; AM, antimalarials; LEF, leflunomide; MCR, major clinical response; PR, partial clinical response; CNS, central nervous system. ECLAM, european consensus lupus activity measurement; HZV, herpes zoster virus; BCDT, B cell depletion therapy.

Both trials utilized stringent and complex endpoints. The EXPLORER trial required that a British Isles Lupus Assessment Group (BILAG) score of C or better at week 24 and no severe flare-ups up to week 52 were achieved to be classified as a major clinical response. The LUNAR trial focused on renal remission, defining a complete renal response as a urine protein-to-creatinine ratio (UPCR) <0.5, normal serum creatinine levels if it was abnormal at baseline, or a serum creatinine level of ≤115% of baseline if it was normal at baseline and there was inactive urinary sediment. In both trials, the choice of the above endpoints was partially inappropriate for the assessment of partial or near-complete responses. Furthermore, both trials assessed their primary efficacy at 52 weeks, a rather early timepoint, since clinical responses in SLE may require a longer time to be achieved. Both trials underscored that RTX effectively depletes peripheral B cells, but this was not always accompanied by clinical improvement.

Despite the negative results seen in both trials, RTX remained a widely used, off-label therapy (25). A substantial number of studies have assessed the efficacy and safety of RTX when administered as an add-on therapy to standard-of-care treatment in patients with SLE, demonstrating high rates of successful outcomes between 60% and 85% (Table 1). These studies examined the short-term efficacy and safety of RTX, with follow-ups ranging from 6 months to up to 5 years. There is only one study thus far that has focused on the satisfactory effects of RTX treatment in patients with SLE with a follow-up of at least 10 years or more, but it suffers from a small sample size and a lack of randomization. The study population samples were also different, ranging from 18 to 261 patients. Efficacy evaluation also varied, as it was either based on overall clinical judgment and disease activity tools such as BILAG and Safety of Estrogens in Systemic Lupus Erythematosus National Assessment (SELENA)-Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) scores or renal responses only. In addition, the RTX treatment schemes were different across studies or even within the same study.

The toxicity profile of RTX is generally considered acceptable. Infusion reaction rates are highly variable, often ranging between 3.5% and 16.4%, whereas some lupus cohorts report a rate of up to 44%. Infections have been recorded in 40%–57% of patients, while the rate of serious infections necessitating hospitalization ranges from 7% to 20%. Respiratory and urinary tract infections, followed by herpes zoster, are the most frequently reported. The mortality rate is reportedly low, ranging from 0% to almost 3%. The vast majority of the deaths are due to severe infections. Based on the positive results in the majority of the studies, RTX can be administered in patients with refractory SLE, according to the guidelines of scientific organizations (26–28).

RTX in LN

Among patients with SLE, roughly 60% develop LN, which may lead to the development of end-stage kidney disease (ESKD). A Bayesian meta-analysis showed that the risk of ESKD in patients with LN decreased between the 1970s and the mid-1990s and then remained stable, but it increased in the late 2000s (29). Despite the extensive administration of immunosuppressive treatments, ESKD risk has not declined.

The first sizeable study that highlighted the beneficial effects of RTX treatment for remission induction in patients with LN applied the steroid-sparing “RITUXILUP protocol” (10). More specifically, 50 patients with LN were administered two doses of RTX 1 g, along with methylprednisolone 500 mg on days 1 and 5, and maintenance treatment with MMF, but no oral steroids. Complete and partial responses were achieved in 52% and 34% at 52 weeks, respectively. Twelve relapses occurred in 11 patients during a median time from remission induction of 65 weeks and 6 out of 50 patients had systemic flares. Among the responders, only two required a course of oral steroid treatment for more than 2 weeks.

Porter et al. examined the effects of the RITUXILUP regimen over ≥5 years (30). Apart from remission maintenance following RTX treatment, a preservation of renal function was also found. Predictors of poor outcomes were no response and a partial response following RTX treatment.

In a real-world evidence study among eight patients with refractory LN, RTX treatment resulted in a remarkable renal response (75% at 1 year and 62.5% at the ≥10 year timepoints) (16). There was stabilization of renal function for more than 10 year after RTX treatment introduction. However, the RTX treatment could not prevent the development of end-stage renal disease in those patients who had an eGFR < 60 mL/min at baseline.

In another study, RTX was administered in 164 patients with biopsy-proven LN, along with glucocorticoids in 162 patients, cyclophosphamide (CYC) in 58, and MMF in 55 (21). At 6 and 12 months after treatment, the response rates were 27% and 30% for a complete response, 40% and 37% for a partial response, and 3% and 33% for no response, respectively. The highest complete response rates were observed in patients with class III LN and mixed types (62% and 50%, respectively), whereas patients with class IV and pure class V exhibited the lowest rates (22% and 18%, respectively). Nephrotic syndrome and renal failure at the time of RTX administration were associated with a worse outcome.

A meta-analysis compared the efficacy and safety profiles of RTX, MMF, tacrolimus (TAC), and CYC (31). It demonstrated that RTX and TAC were the most efficacious treatments for the induction of remission in LN.

The assessment of treatment efficacy in patients with LN is mainly based on proteinuria levels. Treatment should aim for the preservation or improvement of kidney function within 3 months, along with a reduction in proteinuria of at least 25% by 3 months and 50% by 6 months and a UPCR <700 mg/g by 12 months and as low as possible afterwards (26). It has been shown that the renal survival rate at 10 year was 95% for patients who achieve a complete renal response, 45% for those with a partial renal response, and 19% for those with no response (32). Another study showed that at 15 years, the chronic kidney disease (CKD)-free survival rates were 95.2%, 87.6%, and 55.4% in patients who had achieved a complete, partial, or no response at 12 months, respectively (33). Of note, CKD-free survival rates did not differ between patients with a complete renal response and those with a partial renal response. An early proteinuria response of <0.8 g/24 h at 12 months was the best predictor of 7-year renal survival (34). Each renal flare contributes to the accrual of kidney damage, resulting in an increased risk of ESKD (35). With regard to residual low-level proteinuria, despite its association with better long-term survival compared to high levels of proteinuria, it may still not prevent the development of CKD in all patients with LN (36).

A study addressed the differences in the intensity of resorption of sub-epithelial ICs in patients with membranous LN between those treated with RTX (n = 7) and those who received conventional immunosuppressive treatment [n = 17, CYC, MMF, and azathioprine (AZA)] (37). The study demonstrated that IC resorption was significantly increased after RTX treatment but not after conventional treatment; repeated renal biopsies were performed after a median of 10 months (range 6–18). All the patients in the RTX group were responders compared with 53% of the patients who received conventional treatment. Interestingly, resorption of the immune complexes was more pronounced in clinical responders compared to non-responders. Although the number of patients in the study was small and the definitions of clinical response were loose [≥50% reduction in proteinuria to ≤2 g/day and normal renal function (or increased by ≥25% if previously abnormal)], RTX therapy exhibited clinical and histological efficacy.

The resulting increase in circulating B-cell activating factor (BAFF) serum concentrations after B-cell depletion may enhance the repopulation of autoreactive B cells and hence attenuate the efficacy of RTX. This mechanism has been postulated to be targeted by the anti-BAFF mAb, belimumab (BEL), which has already been approved for the treatment of patients with SLE. Neutrophil extracellular traps (NETs) have been proposed as prominent autoantigens that play an important role in the immunopathogenesis of SLE. A sequential treatment with RTX and BEL in 16 patients with severe and refractory SLE achieved significant clinical responses (38). At weeks 12 and 24, RTX and BEL treatment significantly reduced excessive NET formation. Clinical efficacy was also observed. At week 24, the median SLEDAI score decreased from 18 to 2, a low lupus disease activity state was achieved in 10 patients, and renal responses were observed in 11 (5/11 were complete responses). The clinical benefits were observed despite the tapering of MMF and glucocorticoids.

The synergistic effect of BEL in 15 RTX-treated patients with SLE was also investigated in a phase 2 proof-of-concept study including 12 patients with LN (39). A clinical response was observed in 67% and a renal response in 75%, with a total of 67% recorded as complete renal responders, while 33% were recorded as non-responders due to persistent LN or major or repetitive minor flares during a follow-up of 2 years. CD20+ B cells were significantly depleted in responders compared to non-responders.

In a multicenter, randomized, open-label clinical trial, 43 patients with refractory or recurrent LN received RTX, CYC, and glucocorticoids followed by weekly infusions of BEL or not (40). At week 48, a renal response was recorded in 52% of the 21 patients treated with BEL and in 41% of the 22 patients who were not treated with BEL. In contrast to the autoreactive transitional B cells, the percentages of total and autoreactive naive B cells decreased from baseline to week 48 in the BEL group compared to the group that did not receive BEL. The authors concluded that clinical efficacy was not enhanced when BEL was administered sequentially to RTX and CYC. However, the introduction of BEL after RTX and CYC may be a reasonable choice as a maintenance treatment, since there was a trend of better responses in patients with nephrotic levels of proteinuria and in those treated with BEL after RTX and CYC. In contrast, there was an increased frequency of ESKD in the patients treated with RTX and CYC only.

RTX is frequently administered in patients with active proliferative LN with inadequate response to CYC or MMF, showing remarkable efficacy in reducing levels of proteinuria and preserving renal function. However, RTX alone may be insufficient for patients with LN, suggesting that RTX may be best administered as an add-on treatment.

The prognostic impact of RTX

The holistic “treat-to-target” approach for chronic autoimmune diseases such as SLE focuses on the achievement of a sustained, long-term remission, elimination of flares, and, finally, the prevention of irreversible organ damage. Intending to explore which subgroup of patients requires repeated cycles of RTX treatment, researchers reviewed the data of 131 patients with SLE; among them, 44 received one cycle of RTX and 87 received two or more cycles (41). The former group of patients had more than one organ/system involved and had been treated with significantly fewer immunosuppressants before RTX introduction compared to those who required multiple RTX cycles. Responses to RTX treatment were more frequent in patients who did not have previous immunosuppressive treatments. Of note, a statistically higher risk of repeated RTX cycles was observed in patients who had been previously treated with AZA and methotrexate (MTX) compared to those who had received CYC and MMF before RTX treatment initiation.

Thus, patients who required only one cycle of RTX had active, non-refractory SLE and had been exposed to fewer immunosuppressive drugs. Therefore, the sooner a patient with SLE is treated with RTX, the more likely they are to benefit.

Focusing on relapse prevention, an observational multicenter study in 147 patients with SLE reported that treatment failure was evident in 27% at 6 months after RTX treatment (42). Notably, a low number of previous immunosuppressive treatments and low complement C4 levels reduced the risk of treatment failure at 6 months.

Moreover, 80 patients received RTX as a maintenance treatment over a median of 24 months with a median follow-up of 38 months, while the median follow-up after the last RTX treatment was 13 months. During the maintenance treatment, 85 relapses were recorded, mainly involving joints and skin, whereas 28 out of 80 patients (35%) never flared. The rate of major flares during the RTX maintenance treatment was 38%, but this increased to 67% when RTX was discontinued. It was found that active articular disease at the time of the first RTX administration was strongly associated with relapses during RTX maintenance treatment. The relapse-free survival after RTX maintenance treatment was similar to that seen in patients after a single course of RTX.

It may thus be considered that RTX could be administered not only as a rescue therapy but also as a remission maintenance therapy in patients with SLE. Prevention of mild or severe disease flares may reduce the corticosteroid dosage and its subsequent side effects, as well as organ damage in the long-term. However, the optimal interval between RTX courses has not been established so far.

The potential impact of RTX on long-term damage acquisition in patients with SLE has also been examined. For this purpose, a comparison of the difference in the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SLICC/ACR DI) from the first visit until the end of the follow-up was conducted (43). It was shown that the mean SLICC/ACR DI of the group receiving conventional treatment at the beginning of the follow-up was 0.3 compared with 0.9 in the RTX group and at the end of the follow-up, this mean was 0.7 in the control group and 1.5 in the RTX group. Although RTX fell short of expectations in reducing damage, there was a statistically significant difference between the two groups regarding the proportion of patients who progressed to a higher damage index. Thus, at the end of the follow-up, higher SLICC/ACR DI scores were observed in 58.2% of the patients in the control group compared to 44.2% in the RTX group, despite their higher damage baseline scores.

Organ damage resulting from the SLE course itself or from treatment is also a predictor of mortality. Damage accumulation increases progressively among patients with SLE, especially damage related to corticosteroid treatment. Therefore, despite the fact that RTX does not directly reduce long-term damage, it may be useful for maintaining patients on low doses of corticosteroids.

Early use vs. refractory disease

In an effort to maximize sustainable remission rates, researchers examined the effects of the early introduction of RTX in the treatment of patients with SLE. Eight patients with predominantly non-renal SLE were treated with a combination of RTX and CYC at the time of their SLE diagnosis (44). Post-treatment, the patients were treated with AZA. The median decrease in of BILAG score at 6 months was −12.00 vs. 13.22 in the SLE patients treated conventionally. Additionally, the RTX/CYC regimen appeared safe and also resulted in a reduction of the cumulative prednisolone dose.

The long-term, i.e., up to 7 years, results of the RTX/CYC regimen in 16 newly diagnosed patients, the majority with non-renal SLE, as a first-line treatment have also been reported (45).

The mean number of flares during follow-up was 2.63 in the RTX/CYC group and 4 in the controls. The mean cumulative prednisolone dose was 4,745.6 mg in the RTX/CYC group and 12,553.92 mg in the controls.

A retrospective study included treatment of newly diagnosed patients, with a median time interval of 7 days between diagnosis and enrollment (46). The primary outcomes were defined as the attainment of a modified lupus low disease activity state and remission by 12 months; these were achieved in 76% and 50.5% of patients treated with RTX as first line (n = 104), respectively, and 45.8% and 9.7% of the patients treated with conventional immunosuppressants as first line (n = 154), respectively. The study supported the concept that high exposure to previous immunosuppressive treatments is associated with a high likelihood of a poor response to RTX treatment, but was underpowered to evaluate disease subsets such as LN or neuropsychiatric systemic lupus erythematosus. Thus, early and prompt introduction of RTX in the treatment of patients with SLE may alter the natural course of the disease, resulting in not only low disease activity in the short term but also in less organ damage in the long term. For this purpose, a clinical trial in 128 patients aged 5 years or older with a new moderate to severe SLE diagnosis will be conducted (47). All participants will receive standard treatment and steroids. Half of them will be treated with RTX infusions at study entry and at 6 months and the other half will receive a placebo. Patients will be evaluated for a total of 12 months and the main study outcome will be the amount of time that it takes for their disease activity and steroid dosage to be at an acceptable target state.

Options after RTX failure

It has been suggested that patients who fail to respond to RTX treatment due to insufficient depletion may benefit from another anti-CD20 mAb rather than BEL (48). More specifically, 14 out of 77 patients with SLE who required multiple RTX cycles and had no B-cell depletion and non-response were switched to either BEL or to another anti-CD20 agent, such as ocrelizumab (OCR), ofatumumab (OFA), and obinutuzumab (OBZ) (n = 8, n = 3, n = 2, and n = 1, respectively). In the group who were switched to BEL, the SLEDAI-2K score was not reduced and only one patient achieved a Systemic Lupus Erythematosus Responder Index 4 (SRI-4), whereas two out of eight showed new or worsened BILAG-2004 grade A for LN. In the group who were switched to another anti-CD20 agent, the median SLEDAI-2K score improved from 16 at baseline to 5 at 6 months and all patients achieved an SRI-4. However, no conclusions can be drawn from such very small patient populations.

OBZ was administered in nine patients with secondary failure to RTX (49). At 6 months, median SLEDAI-2K and total BILAG-2004 scores were significantly decreased. One patient did not respond to OBZ and required treatment with CYC. Regarding safety, one patient who was not previously vaccinated died from COVID-19 infection.

Alternative anti-CD20 mAbs may increase responses in patients with SLE who do not benefit from RTX treatment. Furthermore, higher response rates are achieved with B-cell killing strategies compared to B-cell silencing strategies in patients who exhibit inadequate responses to RTX treatment.

Combination strategies with BEL

Combination strategies with BEL have also been evaluated. In a phase 2, randomized, double-blind study, 52 patients with SLE were treated with RTX and, 4–8 weeks later, were randomly assigned to receive IV BEL or placebo for a total of 52 weeks (50). The levels of IgG anti-double-stranded DNA (anti-dsDNA) antibodies were lower in the patients who received BEL compared to placebo at 52 weeks. BEL significantly decreased the risk of severe flares compared with placebo, with 3 in the BEL group and 10 in the placebo group. Treatment with BEL did not increase the incidence of serious adverse events.

An exploratory analysis of a randomized controlled 52-week trial analyzed the immune profiles of patients with SLE who received the combination of BEL after RTX versus placebo after RTX (51). Ten out of 21 patients in the BEL group and 8 out of 23 in the placebo group showed a major clinical response at 52 weeks. There was a predictive association between baseline serum IgA2 anti-dsDNA antibody concentrations and clinical response to BEL after RTX. More specifically, patients with high serum IgA2 anti-dsDNA antibody concentrations at baseline were more likely to show a major clinical response when treated with BEL after RTX than those who received a placebo, with a difference of 48% between the two groups compared to the difference of 13% between those in the two groups without this biomarker. Furthermore, among those who had a major clinical response, serum IgA2 anti-dsDNA antibody concentrations were only significantly reduced from baseline in the BEL group. It was also found that serum IgA2 anti-dsDNA antibody concentrations were associated with active LN, whereas serum IgA1 anti-dsDNA antibody and IFN-α concentrations were associated with mucocutaneous involvement. However, these parameters did not predict a response to BEL treatment. Patients with an increased baseline serum IL-6 concentration were less likely to exhibit a major clinical response, irrespective of the treatment. The authors suggested that the incorporation of serum IgA2 anti-dsDNA antibody concentration into everyday clinical practice may guide more individualized therapeutic management of patients suffering from SLE.

In conclusion, BEL administration following RTX successfully eliminates the risk of severe flares. Indeed, this double B-cell targeting approach does not increase the risk of severe adverse events. However, the strategy “BEL after RTX” may only be associated with better outcomes in specific patient subsets, namely those with high baseline serum IgA2 anti-dsDNA antibody concentrations, but this finding needs to be confirmed in a larger-scale trial, in which patients should be stratified according to serum IgA2 anti-dsDNA antibody levels.

Safety, dosing, and monitoring

RTX administration may cause infusion reactions, such as cutaneous rash, flushing, pruritus, back pain, fever, chills, headache, dizziness, urticaria, nausea, vomiting, throat tightness, bronchospasm, chest tightness, dyspnea, hypoxia, wheezing, anaphylaxis, hypotension, and, in rare cases, serum sickness (52).

Additionally, RTX treatment is associated with susceptibility to common infections, but also with reactivation of hepatitis B virus and opportunistic infections such as Pneumocystis jirovecii pneumonia and John Cunningham (JC) virus-induced progressive multifocal leukoencephalopathy (53). RTX has been associated with a higher risk of infections in patients with serum IgG levels <6 g/L and measurement of immunoglobulin levels before the initiation and periodically or before each RTX cycle has been recommended. However, it has been reported that the incidence of the development of hypogammaglobulinemia after repeated cycles of RTX is low (21% for IgM and 7% for IgG) (54). Indeed, RTX has been associated with a higher risk of COVID-19 infection. Patients receiving RTX treatment experienced more severe COVID-19 disease and a higher mortality rate, while hospitalizations may be more frequent within the first 6 months following RTX treatment (55). In terms of vaccinations, RTX administration should be avoided for 2–4 weeks following vaccinations.

A meta-analysis revealed that RTX had the lowest rate of serious adverse events compared to BEL, OBZ, standard-of-care treatment, anifrolumab, and voclosporin. This meta-analysis, using surface under the cumulative ranking curve-based ranking probabilities, showed that anifrolumab was associated with the lowest likelihood of an infection (78.7%), followed by RTX (72.6%), OBZ (67.5%), BEL (34.4%), OCR (33.4%), and the standard treatment (13.5%) (56). The risk of severe adverse events overall was significantly lower with RTX, BEL, and OBZ compared to the standard treatment.

Aiming to minimize the side effects of RTX treatment, such as infections, allergies, and infusion reactions, as well as to save costs, researchers examined the effects of a low-dose regimen, which exhibited efficacy similar to that obtained with standard RTX regimens. Thus far, low-dose RTX regimens have been administered to patients with SLE-related thrombocytopenia and patients with lupus enteritis.

Low-dose RTX (two infusions of 200 mg on days 1 and 15) was administered to 8 patients with SLE-associated immune thrombocytopenia (57). The overall response rate, complete or partial, was 25.0%, 87.5%, 75.0%, and 75.0% at 1, 3, 6, and 12 months, respectively. One patient withdrew from the study due to Herpes Zoster Virus infection and an allergic reaction 2 weeks after the first dose of RTX.

In a prospective pilot study, 10 patients with SLE with refractory thrombocytopenia received RTX at a dose of 100 mg once weekly for 4 weeks (58). An overall response [PLT > 50 × 10(9)/L] was observed in 5/10, 6/10, 7/10, and 5/10 patients at weeks 4, 12, 24, and 36, respectively. Peripheral CD19(+) B cells were depleted in all patients at week 4, and gradually increased at weeks 24 and 36.

Low-dose RTX (either 375 mg/m² of RTX every 2 weeks or 2 doses or 500 mg every 2 weeks) plus glucocorticoids has been suggested as an effective and safe induction treatment for lupus enteritis, with clinical efficacy comparable to conventional immunosuppressants (CYC 0.5 g/m² IV monthly, or MMF 1–2 g/day along with glucocorticoids), as it significantly improved gastrointestinal symptoms and imaging findings, and decreased SLEDAI-2K and BILAG-2004 scores, serum IL-6 levels, and fecal IgA levels at 6 months after treatment (59). In the RTX group, only 1 patient out of 16 relapsed during a median follow-up of 12 months.

Therefore, the high/standard-dose RTX regimens should be minimized for many patients with SLE; low-dose regimens may represent an appropriate therapeutic option.

Emerging B-cell-targeted therapies

To manage difficult-to-treat SLE, the therapeutic landscape has evolved to target B cells with varying mechanisms, depth, and precision. The current therapeutic options range from traditional mAbs to next-generation glycoengineered biological agents and cellular treatments. RTX is associated with a high rate of anti-drug antibody development, frequently correlated with secondary non-response. However, it offers a well-characterized long-term safety profile, extensive clinician familiarity, and a predictable dosing schedule. RTX is easily accessible and globally available. The introduction of various biosimilars makes it the most cost-effective B-cell depleting option.

Other mAb-mediated B-cell targeting treatments include OBZ, an approved treatment for active LN. OBZ is a humanized, glycoengineered type II anti-CD20 mAb that causes direct non-apoptotic cell death primarily via antibody-dependent cellular phagocytosis and antibody-dependent cellular cytotoxicity, and not complement-dependent cytotoxicity. It causes deep and sustained B-cell depletion. Although approved, it remains restricted, resulting in high costs and intensive authorization requirements.

Other anti-CD20 agents, namely, a fully human OFA and a humanized OCR mAb, target distinct epitopes of the CD20 molecule to increase complement-dependent cytotoxicity and antibody-dependent cellular cytotoxicity. They lead to more efficient B-cell depletion than RTX due to low immunogenicity and modified binding kinetics. They can be administered as alternative treatment options in patients with secondary non-response to RTX. A phase III study evaluating OCR in patients with active LN was prematurely terminated because patients treated with this agent along with MMF experienced severe infections (60). A phase III trial investigating OCR in patients with SLE without renal involvement was similarly halted due to a lack of significant therapeutic response. Studies have demonstrated that OFA may be effective in patients with LN (61). Their structure minimizes the risk of anti-drug antibody production and, therefore, they are appropriate treatment options for patients with severe allergic reactions to chimeric antibodies. Clinical data regarding optimal dose strategies and long-term efficacy remain scarce. Similar class-wide risks, including opportunistic infections, hepatitis B reactivation, hypogammaglobulinemia and late-onset neutropenia, have been reported. OFA and OCR remain off-label options for SLE, leading to high costs and significant insurance coverage barriers.

BEL is a human mAb that specifically inhibits soluble B-lymphocyte stimulator (BLyS/BAFF). It promotes the apoptosis of transitional and naive B cells, while completely sparing memory B cells and long-lived plasma cells, causing a modest reduction in B cells. It is an approved biologic that is indicated for both active autoantibody-positive SLE and active LN. BEL possesses an acceptable safety profile with no significant increase in severe opportunistic infections over standard therapy. It preserves protective humoral immunity and memory B-cell pools. It takes several months for clinical benefits to be noticed. BEL is ineffective as a monotherapy for acute, life-threatening, or highly refractory disease flares, and it is associated with a slight increase in psychiatric adverse events and mild hypersensitivity reactions. It has widespread availability and favorable insurance coverage.

Chimeric antigen receptor (CAR)-T cell therapy consists of autologous or allogeneic T cells genetically engineered to express a CAR targeting CD19 (or dual CD19/BCMA) that directly executes major histocompatibility antigen (MHC)-independent, perforin- and granzyme-mediated cytolysis of target cells, causing ultra-deep and complete B-cell depletion. The therapy has been highlighted by impressive pooled data from early-phase clinical trials (62). It aims towards resetting the immune system, enabling long-term, drug-free remission without the need for ongoing maintenance immunosuppression. The procedure requires intensive lymphodepleting chemotherapy conditioning. CAR-T cell therapy carries significant risks of cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome. Other major concerns include prolonged cytopenias, severe hypogammaglobulinemia, and a heightened vulnerability to severe infections during the B-cell aplasia window. However, the therapy has extremely low accessibility, as it is restricted to specialized tertiary cell-therapy centers running clinical trials. CAR-T cell treatment is one of the most expensive available medical therapies, as the estimated cost is hundreds of thousands of dollars per single-intervention dose. In addition, the CAR-T cell era of therapeutics is currently evolving, with many different “versions” of effector cells and target cells. Such differences currently preclude the generation of a more conceptualized statement.

Discussion

The negative results of randomized controlled clinical trials on RTX did not stop its use in patients with refractory SLE. RTX has been administered in the treatment of SLE for more than 20 years and reportedly exhibits a response rate of approximately 70%.

Not all patients with SLE benefit from RTX treatment. It has been well-documented that in patients with SLE, inadequate peripheral B-cell depletion is associated with a poor response to RTX treatment (63). However, peripheral B-cell depletion may not accurately predict clinical responses. For example, renal biopsies performed at a median of 12 weeks from the administration of only one or two doses of RTX showed that patients with lower levels of renal B cells, irrespective of the peripheral B-cell count, had better clinical outcomes (64). Moreover, B cells were detectable in renal biopsy specimens of patients with poor outcomes, while their peripheral B cells remained low or undetectable.

Aiming to delve into the immune response to RTX treatment, single-cell profiling was performed in nine patients with SLE from pretreatment to 15 months after treatment (65). Naive, memory, and age-associated B cells were significantly decreased early (1–6 months) following treatment, followed by later repopulation by mainly transitional B cells. RTX non-responders exhibited a proportion of antigen-experienced B cells that were not eliminated through the depletion. RTX responders had enhanced expression of cytotoxicity-related genes, MHC class II antigen presentation, and T-cell activation in CD4T central memory and double negative T cells, whereas their repopulated naive B cells were transcriptionally distinct from pretreatment naive B cells. In contrast to the dynamic alterations in B cell subpopulations, the B cell receptor repertoire changes were only transient.

The available evidence suggests a potential benefit of RTX in selected patient subgroups, and if this treatment fails, another anti-CD20 mAb can be administered (Figure 1). RTX seems to be highly effective in patients with refractory SLE, hematological abnormalities and central nervous system (CNS) lupus, refractory LN, and recurrent flares, and as a glucocorticoid-sparing regimen. However, RTX has not demonstrated a favorable benefit for mild disease. In contrast, there are limited data regarding its effectiveness when administered to newly diagnosed patients with SLE. Suitable candidates for early treatment may be patients with severe disease and at a high risk for organ damage, those with organ or life-threatening manifestations, or those requiring rapid glucocorticoid reduction. Moreover, there is no global consensus on whether fixed-scheduled dosing or B-cell driven pretreatment yields better responses and long-term outcomes.

Therefore, there is a need for therapeutic strategies that cause more efficient B-cell depletion for some patients with SLE. OBZ, which causes potent and sustained B-cell depletion, has already been approved by the FDA for the treatment of adults with LN. Administration courses longer than the 52 weeks evaluated in the REGENCY trial require further assessment. Additionally, allogeneic CD19-targeted chimeric antigen receptor T-cell therapy may induce sustained clinical remission in patients with severe SLE refractory to standard treatment (62). Moreover, the absence of classic CAR-T-associated toxicities in the patients of the study, although of limited number, raises hopes for safely inducing long-term remission of previously refractory SLE.

While the above therapeutic approaches demonstrate superiority, they cannot completely replace RTX due to high costs and their associated risk profiles. Thus, RTX remains a viable alternative therapeutic approach, although large-scale controlled trials are required in order to overcome the limitations of clinical heterogeneity and lack of optimal dosing protocols.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Clio Mavragani, National and Kapodistrian University of Athens, Greece

Reviewed by: Yusuke Miyazaki,University of Occupational and Environmental Health, Japan

Celia Rodríguez Tudero, University Hospital of Salamanca, Spain