Abstract
Objective To evaluate the risk and protective factors of serious infection (SI) in patients with systemic lupus erythematosus (SLE) within 180 days of rituximab (RTX) treatment.
Methods Patients with SLE treated with RTX were analyzed. SI was defined as any infectious disease requiring hospitalization. The clinical characteristics, laboratory profiles, medications, and incidence rate (IR) are presented. Multivariate Cox proportional hazards models and Kaplan-Meier analysis for risk factors of SI were performed.
Results A total of 174 patients with SLE receiving RTX treatment were enrolled. The overall IR of SIs was 51.0/100 patient-years (PYs). Pneumonia (30.4/100 PYs), followed by soft tissue infections, intra-abdominal infections, and Pneumocystis jiroveci pneumonia (all 6.1/100 PYs) were the leading types of SIs. Twelve patients died during the 180-day follow-up (crude mortality rate: 14.6/100 PYs). Chronic kidney disease (CKD), defined as an estimated glomerular filtration rate < 60 mL/min/1.73 m2 (hazard ratio [HR] 2.88, 95% CI 1.30-6.38), and a background prednisolone (PSL) equivalent dosage ≥ 15 mg/day (HR 3.50, 95% CI 1.57-7.78) were risk factors for SIs among all patients with SLE. Kaplan-Meier analysis confirmed the risk of SI for patients with SLE with CKD and a background PSL equivalent dosage ≥ 15 mg/day (log-rank P = 0.001 and 0.02, respectively). Hydroxychloroquine (HCQ) reduced the risk of SIs in patients with SLE (HR 0.35, 95% CI 0.15-0.82; log-rank P = 0.003).
Conclusion SI was prevalent in patients with SLE after RTX treatment. Patients with SLE with CKD and high-dose glucocorticoid use required constant vigilance. HCQ may reduce the risk of SI among patients with SLE administered RTX.
Systemic lupus erythematosus (SLE) is an autoimmune disease caused by dysregulation of innate and adaptive immunity that involves almost all organs and systems. In the past few decades, emerging biologic agents have been shown to play crucial roles in the management of autoimmune rheumatic diseases including SLE. Although rituximab (RTX), a chimeric monoclonal antibody targeting CD20, was approved for the treatment of B-cell lymphoma, rheumatoid arthritis, and antineutrophil cytoplasmic antibody–associated vasculitis (AAV),1 RTX did not succeed in patients with SLE with or without renal involvement as demonstrated in the Lupus Nephritis Assessment with Rituximab (LUNAR) and Exploratory Phase II/III SLE Evaluation of Rituximab (EXPLORER) trials, respectively.2,3 However, many retrospective studies have suggested the favorable efficacy of RTX in patients with refractory SLE despite standard-of-care therapy, especially for autoimmune hemolysis, immune thrombocytopenia, and patients with lupus nephritis.4 Therefore, RTX is recommended in guidelines as an alternative medication for refractory SLE.5
Nevertheless, comorbid infectious disease is a major drawback of RTX treatment in patients with SLE and is associated with the disease activity of SLE as well as concomitant treatment with glucocorticoids (GCs) and immunosuppressants.6 Patients with SLE frequently suffer from serious infections (SIs), which can cause hospitalization or mortality.7 Thus, it is crucial to prevent RTX-associated SIs among patients with SLE. A few studies have mentioned that the incidence rate (IR) of SIs in patients with SLE after RTX treatment ranged from 1.2 to 18.7/100 patient-years (PYs).2,8-12 Most SI events occur within 6 months after initiating RTX treatment, and respiratory and urinary tract infections are the major types.12-17 Although numerous studies presented the efficacy and safety of RTX in refractory SLE, the risk factors of SIs are seldom discussed specifically for SLE after RTX treatment in the literature. Previous studies have shown that hydroxychloroquine (HCQ) reduces the risk of infections in patients with SLE.6,18-20 However, whether HCQ can also reduce the risk of SIs in patients with SLE after RTX treatment remains unknown. Therefore, the aim of this study was to analyze the IR of SIs and to identify the risk and protective factors of SIs in patients with SLE after RTX treatment during a 180-day follow-up period.
METHODS
Study population. The case selection procedure is shown in Figure 1. The discharge notes from Taipei Veterans General Hospital from January 1, 2007, to June 30, 2016, were reviewed and 12,274 patients with diagnostic codes of SLE were identified. The diagnostic codes of SLE included M32.0, M32.10, M32.11, M32.12, M32.13, M32.14, M32.15, M32.19, M32.8, and M32.9, according to International Classification of Diseases (ICD), 10th revision, Clinical Modification codes, and 710.0 for the ICD, 9th revision, Clinical Modification code. Among them, 200 patients who received RTX treatment were identified. Then, the indications for RTX infusion and the diagnosis of these patients were checked according to the 1997 American College of Rheumatology (ACR) SLE Classification Criteria or the 2012 Systemic Lupus International Collaborating Clinics Revised Classification Criteria.21,22 Twenty-four patients were excluded, including 2 patients not fulfilling the SLE classification criteria, 7 undergoing lymphoma treatment, and 15 for prevention of rejections after organ transplantations. Among the remaining 176 patients, 2 patients were excluded because of concurrent pneumonia with respiratory failure while receiving RTX treatment. Finally, 174 patients with SLE who received RTX treatment without concomitant infection were analyzed. This study was approved by the institutional review board of Taipei Veterans General Hospital (2017-03-010C and 2020-07-026CC).
Flow chart of case selection. SLE: systemic lupus erythematosus.
Baseline characteristics and laboratory data of the patients. All patients received RTX treatment during hospitalization. The background characteristics, medications, the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K), age on the admission date, and the date of RTX treatment were recorded for all patients. The index date was defined as the day that the patient received their first RTX infusion. Laboratory data were logged according to the most recent data before the index date. The estimated glomerular filtration rate (eGFR) was calculated according to the Chronic Kidney Disease Epidemiology Collaboration equation.23 The doses of GCs (equivalent to prednisolone [PSL]), disease-modifying antirheumatic drugs (DMARDs), and immunosuppressants administered to the patients were recorded according to the most recent medical records from the outpatient department or during hospitalization before RTX treatment.
RTX infusion regimens. The RTX infusion protocol from the LUNAR study was followed at our hospital, with minor modifications.2 In brief, each dose of RTX was preceded by intravenous methylprednisolone (100 mg), oral fexofenadine (120 mg), and acetaminophen (1000 mg) to prevent allergic reactions. The dosing of RTX is summarized in Table 1. The majority of patients with SLE (n = 120) received 2 RTX infusions at a dose of 500-1000 mg, 2 weeks apart. The time period between the 2 doses was individualized according to the judgment of the physicians and the insurance coverage of the patients. In addition, 16 patients (9.2%) received a single dose of RTX and 5 (2.9%) received concomitant cyclophosphamide (CYC; 200 mg) and RTX treatment. Only 1 pediatric patient (0.6%) received 5 infusions of RTX at a dose of 375-500 mg every other week, which was the modified regimen for those with B-cell lymphoma. No patient with SLE had any clinical evidence of infection while receiving RTX treatment.
Detailed dosing of RTX in the 174 patients with SLE.
Primary endpoint and secondary endpoint. The primary endpoint was the first episode of a SI during the 180-day follow-up period after the first RTX infusion. The definition of SIs was any bacterial, fungal, or viral infectious disease that required hospitalization. The secondary endpoint was any subsequent SI episode or mortality during the 180-day follow-up period. The types of SIs were also recorded.
Risk factor analyses for SIs. The risk factors of SIs in patients with SLE treated with RTX, including SLEDAI-2K, laboratory profiles, systemic GC treatment, and DMARD or immunosuppressant use, were analyzed. CKD was defined as an eGFR < 60 mL/min/1.73 m2.24
Statistical analyses. Data were presented as mean (SD) or n (%) as appropriate. Cases with missing data were excluded from the further inferential statistics. Fisher exact tests were used to compare categorical variables and Mann-Whitney U tests were used to compare numerical data. Multivariate Cox proportional hazards models were conducted to evaluate the risk of SIs in patients with SLE after RTX treatment. Factors with P < 0.1 according to univariate analysis were selected for multivariate analysis using the enter method. The Kaplan-Meier method and the log-rank test were used to compare the risk of SIs between groups. A P value < 0.05 was considered statistically significant. All statistical analyses and plots were generated by SPSS (version 23 for Windows) or StataSE (version 11).
RESULTS
Clinical and laboratory profiles. The mean age of all 174 patients with SLE was 38.8 years old, and 156 patients (89.7%) were female. The mean SLEDAI-2K score of these patients was 10.5. The baseline laboratory data of the patients are shown in Table 2. A total of 110 (63.2%) patients received RTX treatment because of renal involvement, 26 (40.6%) patients were treated for hematologic involvement, 9 (5.2%) patients for arthritis, another 9 (5.2%) patients for neurologic involvement, and 20 (11.5%) patients for other organ system involvement of SLE. The mean urinary protein was 3.6 g/day and 28.7% patients had the status of CKD. To treat the high disease activity of SLE, the average dose of background PSL equivalent dosage was 15 mg/day. Multiple immunosuppressive agents such as HCQ (n = 81), mycophenolate (MMF; n = 64), calcineurin inhibitors (n = 37), azathioprine (AZA; n = 39), and CYC (n = 8) were also frequently used in these patients with SLE. The rate of SIs was documented in 7/81 (8.6%) patients on HCQ, 12/64 (18.8%) on MMF, 11/37 (29.7%) on cyclosporine (CsA), and 4/39 (10.3%) on AZA treatment.
Baseline clinical and laboratory characteristics of patients with SLE receiving RTX treatment.
Compared to the patients with SLE without SIs, patients with SLE with SIs had higher SLEDAI-2K scores (12.6 vs 10.0, P = 0.03), lower hemoglobin levels (96 g/dL vs 118 g/dL, P = 0.01), and lower eGFR levels (69 mL/min/1.73 m2 vs 88 mL/min/1.73 m2; P = 0.03). Patients with SLE with SIs were treated with a higher baseline PSL equivalent dose (23 mg/day vs 13 mg/day; P = 0.01). The patients with SIs were treated less frequently with HCQ (22.6% vs 51.7%; P = 0.01).
SIs and mortality. Among the 174 patients with SLE, 42 episodes of SIs were found in 31 patients with SLE during the 180-day follow-up period (prevalence rate 24.1%; IR 51.0 episodes/100 PYs; Table 3). The first episode of SIs occurred at a median time of 41 (IQR 22-58) days after the first RTX infusion. Pneumonia was the most common SI (25 episodes; IR 30.4/100 PYs), followed by soft tissue infections and intra-abdominal infections (both with IR 6.1/100 PYs) in all patients with SLE. Opportunistic infections were also presented at an IR of 6.1/100 PYs for Pneumocystis jiroveci pneumonia (PCP) and 2.4/100 PYs for cytomegalovirus (CMV) pneumonia. None of the patients with SLE in our study received pneumococcal vaccines or prophylactic sulfamethoxazole-trimethoprim (SMX-TMP) for PCP.
Episodes of severe infectious and mortality outcome in patients with SLE after RTX treatment (n = 31).
Twelve patients (6.9%) died during the 180-day follow-up period (crude mortality IR 14.6/100 PYs); 11 patients died because of pneumonia and 1 patient died because of intra-abdominal bleeding with hypovolemic shock.
Risk factors for SIs. The risk factors for the first episode of SIs in all patients with SLE were analyzed by using univariate Cox proportional hazards models. SLEDAI-2K > 12 (HR 2.52, 95% CI 1.25-5.10), hemoglobin < 8 g/dL (HR 3.53, 95% CI 1.52-8.21), CKD (HR 3.51, 95% CI 1.51-6.19), PSL equivalent dosage ≥ 15 mg/day (HR 2.36, 95% CI 1.13-4.97), and calcineurin inhibitors (HR 2.24, 95% CI 1.07-4.67) were associated with the first SI event within 180 days after RTX treatment. HCQ use was found to be associated with a decreased risk of SIs (HR 0.30, 95% CI 0.13-0.71). The presence of CKD (HR 2.88, 95% CI 1.30-6.38), a PSL equivalent dosage ≥ 15 mg/day (HR 3.50, 95% CI 1.57-7.78), and HCQ use (HR 0.35, 95% CI 0.15-0.82) remained statistically significant, according to the multivariate Cox proportional hazards models (Table 4).
Risk factors analysis of the first severe infections within 180 days in patients with SLE after RTX treatment.
Further, Kaplan-Meier analyses confirmed that the risk factors of SIs among all patients with SLE within 180 days of RTX treatment were existing CKD and PSL equivalent dosage ≥ 15 mg/day (Figure 2A,B). Similarly, HCQ use reduced the risk of SIs (log-rank P = 0.003, among all patients with SLE; Figure 2C).
Kaplan-Meier survival estimates of patients with SLE after rituximab treatment according to (A) eGFR cut-off of 60 mL/min/1.73 m2; (B) prednisolone equivalent dosage cut-off of 15 mg/day; and (C) HCQ use in all patients with SLE. eGFR: estimated glomerular filtration rate; HCQ: hydroxychloroquine; Pred: prednisolone; SLE: systemic lupus erythematosus.
DISCUSSION
The present investigation demonstrated that SIs were not uncommon in patients with SLE after RTX treatment. The IR of SIs was 51.0/100 PYs, and pneumonia was the most common type of SI. The major risk factors of SIs included CKD and PSL equivalent dosage ≥ 15 mg/day. HCQ use seemed to be a protective factor of SIs for patients with SLE. To the best of our knowledge, this study is the largest single-center study evaluating the IR of SIs in patients with SLE after RTX treatment that also identifies the risk factors of SIs. These findings suggest that physicians should be aware of SIs in patients with SLE after RTX treatment, especially if they have such risk factors. According to our data analysis, the use of HCQ is recommended in those preparing for RTX treatment.
The rates of SIs were high in patients with SLE treated with different immunosuppressants, including high-dose GCs (10.4-53.3%), CYC (9.2-24%), MMF (7.2-18.2%), AZA (9.5-32.9%), tacrolimus (5.2-14.1%), and CsA (9.1-28.3%), as demonstrated by a systematic review and network metaanalysis study.25 Our study also reported a similar finding. This study also reported an infection rate of approximately 8.1% to 25.3% with the combination of RTX and MMF treatment, which was compatible with the SI rate of 17.8% (31/174) in the present study. This suggests that we should be aware of SIs following RTX and other immunosuppressive treatment in patients with SLE.
Regarding the episodes of SIs in patients with SLE receiving RTX treatment, the prevalence rate has been reported to be 4.2% to 35.5%,4,8,12,13,15,16,26-30 and the IR ranges from 1.2 to 18.7/100 PYs in different studies.2,8-12 The organ involvement of SLE, cumulative organ damage, geographic regions, ethnic differences, background immunosuppressive medication, experience of RTX treatment, follow-up durations, and the age and year of patient enrollment contributed to data heterogeneity. The incidence of SIs seemed to be lower in patients with SLE with immune cytopenia (prevalence rate 4.2%; IR 1.2/100 PYs)8 and higher in patients with refractory/relapsing SLE and lupus nephritis (prevalence rate 35.5%; IR 18.7/100 PYs).2,9,28 The prevalence rate of SIs in the current study was 24.1%, which is comparable to the previous studies. However, the IR of SIs in our study was 51.0/100 PYs, which is much higher than that reported in previous investigations. Our patients were followed after the first RTX infusion (the first 180 days), which is the hot zone of most infectious episodes according to previous studies.9,12,13,15 A longer follow-up period is supposed to further reduce the IR. In addition, a high proportion of our patients had CKD (28.7%), which may also increase the susceptibility to infections.
A significant incidence of pneumonia episodes, especially PCP, was found among patients with SLE treated with RTX in our study. The 2022 ACR guideline for vaccinations in patients with rheumatic and musculoskeletal diseases recommends that patients aged < 65 years with rheumatic and musculoskeletal diseases on immunosuppressive medications should be administered pneumococcal vaccines.31 However, because of a lack of insurance coverage, our patients with SLE did not receive pneumococcal vaccines. Despite this, none developed pneumococcal pneumonia. Further, none of the patients with SLE in our study received SMX-TMP for PCP prophylaxis. The 2022 European Alliance of Associations for Rheumatology guidelines for AAV endorse the use of SMX-TMP for patients with AAV on treatments like RTX, CYC, or high-dose GCs.32 However, the recommendation for PCP prophylaxis in patients with SLE remains a topic of debate. Only 1 study has addressed PCP prophylaxis in patients with SLE on CYC, and it did not advocate for routine PCP prophylaxis except in cases with severe leukopenia, lymphopenia, high-dose corticosteroids, hypocomplementemia, active renal disease, and a higher mean SLEDAI-2K score.33 Another study advised against PCP prophylaxis in patients with SLE because of the low prevalence of PCP and the high rate of adverse reactions related to SMX-TMP.34 However, given the increased risk of PCP observed in our study, routine PCP prophylaxis might be worth considering for patients with SLE undergoing RTX treatment. Further prospective studies are needed to evaluate the efficacy and safety of PCP prophylaxis in patients with SLE after RTX treatment.
The 180-day mortality rate led by SIs in our study was 6.9% (crude mortality IR 14.6/100 PYs), which is much higher than that reported in previous studies, ranging from 0% to2.2%.4,10-12,16,27 One study has revealed that 5 of 69 patients (7.2%) died because of SIs after a 12-month follow-up, and their infection-related mortality rate is comparable to that of our study.9 The high mortality in our study might be because of the high proportion of patients with CKD and the high prevalence of pneumonia as well as opportunistic infections.
Consistent with many previous studies, respiratory infection was the most common type of SI in our investigation.2,4,9,11,12,14-16,27-29 Only 1 systematic review has shown cutaneous infection to be the most common cause.30 The second most common infections described previously are urinary tract infection and herpes zoster.2,4,11,15,16,29,35,36 However, soft tissue and intra-abdominal infection were the second most common infections in our study, followed by urinary tract infections and herpes zoster infections. Opportunistic infections including PCP and CMV pneumonia were not rare in our study. The LUNAR study showed that 3 of 73 patients (4.1%) in the RTX group had opportunistic infections, including colitis, histoplasmosis, and cryptococcal pneumonia superimposed with fungal sepsis.2 Moreover, Ramos-Casals et al have identified 3 patients with opportunistic infections (1.5%) out of 196 patients with systemic autoimmune diseases treated with RTX, which were caused by tuberculosis, systemic cytomegalovirus infection, and aspergillosis.29 In addition, Md Yusof et al found 8 opportunistic infections (1.1%) in 700 patients with rheumatic and musculoskeletal diseases, of whom 2 had varicella zoster virus, 2 had mycobacterium infection, 2 had cytomegalovirus, 1 had disseminated candidiasis, and 1 had PCP infection.35 The incidence of opportunistic infection in our study was similar to that of the above studies.
The present study revealed 2 risk factors of SIs in patients with SLE after RTX treatment, which were CKD and background PSL equivalent dosage ≥ 15 mg/day. Renal involvement has been shown to predict SIs in patients with SLE undergoing immunosuppressive treatment in a Latin American cohort study18; however, RTX was not included in the study as one of the therapeutic protocols. Anemia caused by autoimmune hemolytic anemia or anemia of chronic disease reflects the severity of SLE. The severity of organ involvement, chronic lung diseases, and heart failure has been found to be associated with SIs in patients with SLE receiving B-lymphocyte depletion therapy.11,35 Concomitant use of PSL at a dosage ≥ 15 mg/day is also considered as a risk factor of SIs in patients with SLE or other systemic autoimmune diseases, with or without RTX treatment.9,18 Our study confirmed that a PSL equivalent dosage of ≥ 15 mg/day is a risk factor of SIs in patients with SLE treated with RTX. In another respect, it has been reported that hypogammaglobulinemia is associated with RTX treatment and is also a risk factor of SIs in patients with SLE undergoing RTX treatment.9,35 However, we found baseline hypogammaglobulinemia was not significantly related to SIs in our study. Some patients in our study lacked baseline serum immunoglobulin data before RTX treatment, which might lessen the statistical power. On the other hand, the use of HCQ has been shown to reduce the risk of SIs in patients with SLE.6,18-20 Further, our study also demonstrated that HCQ use had a protective effect in all patients with SLE after RTX treatment. A further prospective study is needed to evaluate the efficacy of HCQ in preventing SIs in patients with SLE undergoing RTX treatment.
Several limitations of this study must be addressed. First, our study was a historical design. Thus, the protective effect of HCQ in patients with SLE after RTX treatment needs to be confirmed by further prospective studies. Second, the patients included in our study did not receive a standardized regimen of RTX treatment. This would inevitably cause some bias. RTX for SLE and allied diseases is not covered by the National Health Insurance System of Taiwan, so individualized treatment did exist in a real-world setting. Third, we did not routinely check the baseline serum immunoglobulin level before RTX treatment. Therefore, hypogammaglobulinemia could not be analyzed as a risk factor for SIs in our study. Fourth, a high proportion of our patients with SLE had renal involvement. This could lead to a relatively high SI rate in our study. As a result, our findings may not be generalizable to the broader SLE population receiving RTX. Despite these limitations, our study revealed that SIs in patients with SLE after RTX treatment might be quite common after the first round of infusions.
In conclusion, CKD and baseline PSL equivalent dosage ≥ 15 mg/day were the risk factors of SIs. In addition, HCQ use presented as a protective factor for all patients with SLE after RTX treatment.
ACKNOWLEDGMENT
The authors acknowledge the Department of Information Management of Taipei Veterans General Hospital for assistance with data acquisition.
Footnotes
This work was supported in part by the Ministry of Science and Technology, Taiwan (MOST 110-2628-B-075-019, MOST 109-2628-B-075-014, MOST 111-2314-B-075-005, MOST 109-2314-B-010-053-MY3, MOST 109-2811-B-010-532, MOST 107-2314-B-075-051-MY3, MOST 110-2811-B-010-510, and MOST 110-2321-B-A49-003); grants from Taipei Veterans General Hospital, Taipei, Taiwan (V111C-153, V110B-014, V110B-036, V110C-046, V110C-194, and V112B-018); “Yin Yen-Liang Foundation Development and Construction Plan” of the School of Medicine, National Yang Ming University, Taipei, Taiwan (107F-M01-0504); and the “Center for Intelligent Drug Systems and Smart Bio-devices (IDS2B)” from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan. The funders had no role in study design, data collection, analysis, interpretation, or writing of the manuscript.
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication September 14, 2023.
- Copyright © 2024 by the Journal of Rheumatology
