Rituximab (RTX) is a chimeric monoclonal antibody engineered through recombinant DNA technology that specifically targets the transmembrane protein CD20 on B cells, binding to it and inducing cytotoxicity mediated by complement and antibodies, thus depleting peripheral B cells. The CD20 antigen is not expressed on hematopoietic stem cells, normal plasma cells, or other normal tissues, allowing RTX to selectively reduce B lymphocytes, inhibiting the production of autoantibodies without the non-specific immunosuppressive toxicity (1). As understanding of RTX has deepened, targeted therapy focusing on B cells has gained attention and is increasingly applied in clinical settings. The 2021 Kidney Disease Improving Global Outcomes (KDIGO) guidelines recommend RTX for treating various immune-mediated glomerular diseases, such as idiopathic membranous nephropathy (IMN), minimal change disease (MCD), and lupus nephritis (LN) (2). The consensus among experts in China also supports the use of RTX for various pathological types of glomerular nephritis, including IMN, MCD, focal segmental glomerulosclerosis (FSGS), Anti-Neutrophil Cytoplasmic Antibodies-associated vasculitis, and LN (3), and also in studies suggesting effectiveness in IgA nephropathy (IgAN) (4) and IgA vasculitis (5). In terms of safety, RTX is closely associated with severe infections, including Pneumocystis jirovecii pneumonia (PJP) and the reactivation of hepatitis B virus (HBV) and tuberculosis (TB). The mechanisms of infection risk involve B-cell depletion, B-cell–T-cell crosstalk, hypogammaglobulinemia, delayed neutropenia, and diminished immune response after vaccination (6). Studies have shown that RTX poses a significantly higher risk of infection in secondary membranous nephropathy (SMN) compared to primary glomerular diseases (PGDs) (7), particularly in older individuals, those with low baseline IgG levels, and a recent history of infections. In our center, RTX is more frequently used in PGDs; hence, this study aimed to explore the incidence and risk factors for pulmonary infections in PGDs treated with RTX to assist in clinical assessment and preventive measures.

With the increasing use of biologics in immunologically mediated glomerular diseases, their safety in terms of infection risk has become increasingly critical. The RI-CYCLO study posited that there is no difference in efficacy and safety between rituximab and a cyclical regimen of GC and CTX in treating IMN (10). A cohort study of 135 IMN patients with a median age of 67 years followed up for 4 years demonstrated that age is a reliable predictor of infection risk (11), likely due to immunosenescence associated with aging, which impairs resistance to infections (12). In patients over 70 years old treated with RTX, while a 70% complete remission rate could still be achieved, the rate of fatal pulmonary infections was 8.1% (13). Our study in the PGD population similarly confirmed that advanced age is a risk factor for pulmonary infections.

The Trivin study (14) identified diabetes and cumulative RTX dose as independent risk factors for infections among patients with various types of glomerulopathies. Even among 90% of patients maintaining normal sIgG levels, a higher cumulative dose of RTX was associated with increased infection risk (15). However, this phenomenon was not observed in our study, possibly because the target population was PGD patients, and those with diabetes were relatively few, only 14.2%. Grouping by an RTX dose threshold of 2,000 mg did not reveal any impact of cumulative RTX dose on infection risk, possibly because all patients underwent thorough infection risk assessments before RTX therapy, allowing adequate dosing of 2,000 mg RTX in low-risk patients, whereas high-risk patients, including those of advanced age, with baseline pulmonary disease, or with low lymphocyte counts, received intermittent low doses based on B-cell depletion. The study of Odler et al. (16) suggested that factors closely associated with infection occurrence included low BMI and high baseline Scr, and GC treatment concurrently. Despite a 58.5% incidence of hypogammaglobulinemia, it was not associated with infection risk.

In our study, infections resolved after the reduction or discontinuation of immunosuppressants and antibiotic treatment, except for one patient who died from respiratory failure after being transferred to the Intensive Care Unit (ICU), and one patient discontinued treatment and was discharged against medical advice.

In addition to pulmonary infections, rare cases of other infections were not included in our study, such as herpes zoster, gastrointestinal infections, and cellulitis.

Our analysis identified sALB as an independent protective factor. Complete innate and adaptive immune responses depend on sALB; oxidative degradation of sALB can impact interactions with bioactive lipid mediators crucial for antimicrobial defense and repair, leading to impaired immune function, pulmonary edema, and fluid retention, which in turn promote infection development (17). Hypoalbuminemia is also an independent prognostic factor for various cardiovascular diseases (18, 19). Hypoalbuminemia is associated with the acquisition and severity of infectious diseases. There is bio-mechanistic plausibility for a causal link between hypoalbuminemia and increased risks of primary and secondary infections. Serum albumin levels have a prognostic value for complications in viral, bacterial, and fungal infections. Hypoalbuminemia predicts the development of healthcare-associated infections, particularly with Clostridium difficile (17).

Neutrophil percentage-to-albumin ratio (NPAR) emerged as a new predictive marker for poor infection prognosis, which proved to be more accurate than albumin or neutrophil percentage alone (20). NPAR, as a novel biomarker for systemic inflammation and infection, is receiving increasing attention. Studies have identified the optimal threshold of NPAR for predicting stroke-related infection incidence as 1.64 (21). Various studies have demonstrated that NPAR can serve as a prognostic indicator for patients with cardiogenic shock (22), myocardial infarction (23), acute kidney injury (24), and pancreatic cancer (25). In our study, the infection group had significantly higher NPAR values than the non-infected group, but further large-scale clinical studies are needed to confirm this as a risk factor.

Our study included 68.7% of patients with IMN, which is the most common primary renal syndrome in middle-aged and elderly patients, characterized by a higher incidence of hypertension, more severe glomerulosclerosis and interstitial fibrosis, and lower eGFR, leading to a higher risk of progressing to end-stage renal disease (ESRD) and higher infection risk, especially under immunosuppressive treatment (13). Multiple studies have shown that baseline creatinine is a high-risk factor for infection (16, 26), aligning with our findings.

In autoimmune responses, underlying infections are autoimmunity against three immunological targets: neutrophils, complement, and cytokines. The complement system is involved in both innate and adaptive immunity. It comprises three major pathways: the classical pathway (the activation of which is dependent on antibodies) and the lectin and alternative pathways (which can be directly activated by microorganisms, without the involvement of adaptive immunity) (27). Renal diseases are frequently mediated by immune complexes via the classical pathway, whose activation is antibody-dependent. While the pathogenic role of autoantibodies against complement molecules is often unclear, specific autoantibodies targeting the C3 convertase can enhance its activity and reduce complement levels, thereby predisposing individuals to severe bacterial infections. Our study demonstrated that decreased C3 is an independent risk factor for pulmonary infections following RTX treatment.

Pulmonary infections, which progress rapidly and can easily lead to respiratory failure, lack specific clinical and CT features, sometimes requiring Next-Generation Sequencing (NGS) of bronchoalveolar lavage fluid for definitive diagnosis, highlighting the importance of preventing pulmonary infections. Therefore, it is advised that special attention and careful risk assessment be given to patients over 50, especially those with diabetes, chronic pulmonary diseases, malnutrition, low serum IgG, and renal impairment, possibly administering Sulfamethoxazole prophylactically. Thus, individualized prevention and treatment strategies, including screening for tuberculosis, are of critical importance. Recommendations from the National Clinical Research Center for Infectious Diseases (28) suggest routine prophylactic antituberculosis treatment for 3 to 9 months for patients requiring long-term use of immunosuppressants with LTB. Prophylactic regimens included isoniazid monotherapy (300 mg once daily for 6 or 9 months), isoniazid plus rifapentine (once weekly for 3 months), isoniazid plus rifampin (once daily for 3 months), isoniazid plus rifapentine (once daily for 1 month), and rifampin monotherapy (600 mg once daily for 4 months). In adaptive immune responses, immunity against tuberculosis is primarily mediated by T cells. Prior to the initiation of a CD20 antagonist, neither LTB screening nor prophylactic antituberculosis treatment is required. However, prophylactic treatment for LTB is recommended in patients receiving GC at doses exceeding 15 mg, CTX, MMF, or TAC. No cases of TB infection were observed among all pulmonary infections reported in our study.

The lack of effect of concomitant immunosuppressant use on pulmonary infection outcomes could be influenced by underlying diseases, economic conditions, and compliance, among other factors, making it difficult to standardize treatment protocols. The variability in RTX dosing and duration of use among study participants, as well as potential recall bias, may have influenced these results. The sample sizes for FSGS and IgAN were relatively limited. In these patient populations, RTX was administered only after rigorous pathological selection. For instance, in IgAN, RTX was considered only when the rate of foot process effacement exceeded 70%. In FSGS, pathological changes were minimal and more consistent with those in MCD. Both groups consisted predominantly of young patients and exhibited a relatively low incidence of infection. Future studies should aim to expand the sample size within these pathological subtypes.

In conclusion, the study primarily explored the occurrence and influencing factors of pulmonary infections in PGD patients treated with RTX, providing a theoretical basis and clinical experience for the treatment and prevention of pulmonary infections. The study has certain limitations, including its single-center, retrospective design, a relatively short follow-up duration, and a lack of external validation. Therefore, further confirmation through subsequent multicenter prospective studies is warranted.