Abstract
Background
The expanding use of biologic and targeted synthetic disease-modifying antirheumatic drugs (bDMARDs and tsDMARDs) has substantially improved outcomes in autoimmune diseases but is accompanied by complex safety concerns. Risk management plans (RMPs) have been introduced to mitigate treatment-related risks; however, real-world adherence to these strategies and their broader clinical impact remain incompletely characterized.
Methods
We conducted a retrospective observational cohort study of adult patients with autoimmune diseases who received bDMARDs, tsDMARDs, or biosimilars at a tertiary medical center in southern Taiwan between October 2014 and December 2023. Patients were classified into RMP and non-RMP groups based on completion of predefined pre-treatment safety assessments within 6 months prior to therapy initiation, including pulmonary and tuberculosis evaluation, viral hepatitis screening, and documentation of cardiovascular and malignancy risk factors. Clinical outcomes included Pneumocystis jirovecii pneumonia (PJP), major adverse cardiovascular events (MACE), treatment-related respiratory adverse events, and all-cause mortality. Multivariable logistic regression and time-to-event analyses were performed to evaluate associations between RMP implementation and clinical outcomes.
Results
Among 1,078 patients included, only 348 (32.3%) fulfilled the predefined RMP criteria prior to treatment initiation. Compared with the RMP group, patients without RMP exhibited significantly higher incidences of PJP (11.9% vs. 2.3%), MACE (8.5% vs. 2.6%), treatment-related respiratory adverse events (50.4% vs. 42.8%), and all-cause mortality (7.3% vs. 1.7%) (all p < 0.05). After multivariable adjustment, RMP implementation remained independently associated with lower risks of PJP (adjusted odds ratio [aOR] 0.18, 95% CI 0.15–0.84), MACE (aOR 0.30, 95% CI 0.13–0.71), and all-cause mortality (aOR 0.21, 95% CI 0.07–0.61). Subgroup and time-to-event analyses demonstrated that anti-CD20 therapy was associated with the highest risk of early-onset PJP, MACE, and mortality, with most events occurring within the first three to six months following treatment.
Discussion
In real-world clinical practice, adherence to predefined RMPs prior to targeted therapy initiation is suboptimal and is strongly associated with clinically meaningful differences in infectious, cardiovascular, and survival outcomes. These findings suggest that RMP implementation is associated with benefits extending beyond infection prevention and underscore the importance of standardized, consistently applied risk management strategies, particularly for high-risk targeted therapies such as anti-CD20 agents.
|
Key Points • Adherence to predefined risk management plans (RMPs) before initiating biologic or targeted synthetic DMARDs remains suboptimal in real-world practice. • Implementation of RMPs is associated with lower risks of Pneumocystis jirovecii pneumonia, major adverse cardiovascular events, and all-cause mortality. • Lack of structured pre-treatment safety assessment may contribute to early severe complications after therapy initiation. • Anti-CD20 therapy is associated with a higher risk of early adverse outcomes, underscoring the need for careful risk stratification and monitoring. |
Supplementary Information
The online version contains supplementary material available at 10.1007/s10067-026-08152-5.
Keywords: Autoimmune diseases, Biologic DMARDs, Pharmacoepidemiology, Risk management plan, Targeted synthetic DMARDs
Introduction
The therapeutic landscape of autoimmune diseases has evolved substantially with the widespread adoption of targeted therapies, including biologic disease-modifying antirheumatic drugs (bDMARDs) and targeted synthetic DMARDs (tsDMARDs). Agents targeting tumor necrosis factor (TNF), interleukin-6 (IL-6), interleukin-17/23 pathways, cytotoxic T-lymphocyte-associated protein 4-Ig (CTLA4-Ig), B-cell depletion or survival (e.g., anti-CD20, anti-BLyS), and intracellular signaling pathways such as Janus kinase (JAK) have significantly improved disease control and long-term outcomes across a broad spectrum of inflammatory rheumatic and systemic autoimmune diseases [1]. More recently, highly specialized cellular therapies, particularly chimeric antigen receptor T-cell (CAR-T) therapy, have emerged as experimental options for refractory autoimmune conditions [2], further expanding the concept of precision immunomodulation.
Despite these therapeutic advances, targeted immune interventions are accompanied by distinct and evolving safety concerns. Established risks include serious infections, reactivation of latent tuberculosis and hepatitis B virus, cytopenias, hepatic.
dysfunction, dyslipidemia, malignancy, and paradoxical immune-mediated phenomena [3]. In addition, accumulating post-marketing and real-world data have raised concern regarding cardiovascular events, thromboembolic complications, and mortality signals associated with certain JAK inhibitors in high-risk populations, underscoring the importance of individualized risk assessment [4, 5]. The infectious risk spectrum has also broadened beyond classical concerns to include herpes zoster, opportunistic infections such as Pneumocystis jirovecii pneumonia (PJP) and cytomegalovirus (CMV) disease, and viral respiratory infections in the post-COVID-19 era [6]. CAR-T therapy introduces a distinct safety profile characterized by cytokine release syndrome, immune effector cell associated neurotoxicity syndrome, prolonged cytopenias, and sustained immune dysregulation, highlighting the need for structured monitoring frameworks when applying advanced immunotherapies outside oncology [7, 8].
To mitigate these risks, risk management programs (RMPs) emphasizing infection screening, prophylaxis, and longitudinal surveillance have been incorporated into clinical practice and reimbursement policies in many regions, including Taiwan [9, 10]. However, real-world adherence to RMPs may vary over time, particularly after treatment initiation, during off-label use, or following transitions to biosimilar agents. Whether inconsistencies in RMP implementation translate into differential infectious outcomes in contemporary clinical practice remains insufficiently defined. Therefore, we conducted a real-world observational study at a tertiary medical center in southern Taiwan to evaluate RMP adherence during targeted therapy use and to examine its association with infection-related outcomes in patients with autoimmune diseases.
Accordingly, to determine whether RMP adherence remains consistent in real-world practice during targeted therapy use and whether variations in RMP implementation translate into differential infectious outcomes in the modern era, we conducted an observational study at a tertiary medical center in southern Taiwan (Kaohsiung Veterans General Hospital).
Methods
Study design and population
This study was approved by the Institutional Review Board of Kaohsiung Veterans General Hospital (IRB No. KSVGH23-CT12-14). The requirement for informed consent was waived due to the retrospective nature of the study and the use of anonymized data. The study was conducted in accordance with the Declaration of Helsinki.
We conducted a retrospective observational cohort study at a tertiary medical center in southern Taiwan. Adult patients (≥ 18 years) with autoimmune diseases who received biologic disease-modifying antirheumatic drugs (bDMARDs), targeted synthetic DMARDs (tsDMARDs), or biosimilars between October 1, 2014, and December 31, 2023, were eligible for inclusion.
Autoimmune diseases include autoimmune inflammatory joint diseases, connective tissue diseases, autoimmune dermatologic diseases, immune-mediated gastrointestinal diseases, immune-mediated thrombocytopenia, and other immune-mediated conditions. Patients receiving targeted therapies exclusively for oncologic indications were excluded.
The index date was defined as the date of first exposure to a targeted therapy during the study period.
Targeted therapies were categorized according to their principal mechanism of action, including TNF-α inhibitors, IL-6 inhibitors, IL-12/23 inhibitors, IL-17 inhibitors, IL-23 inhibitors, T-cell costimulation inhibitors, B-cell targeted therapies (anti-CD20 monoclonal antibodies and anti-BLyS agents), JAK inhibitors, and biosimilars.
A predefined RMP was operationally defined based on completion of standardized pre-treatment safety assessments within 6 months prior to targeted therapy initiation.
Patients were classified into the RMP group if all the following components were documented:
1. Pulmonary and Tuberculosis Assessment.
2. Chest radiography or chest computed tomography.
3. Latent tuberculosis screening using interferon-gamma release assay (IGRA) or tuberculin skin test, with or without prophylactic treatment as indicated.
4. Viral Hepatitis Screening.
5. Hepatitis B surface antigen, hepatitis B core antibody, and hepatitis C antibody testing.
6. Antiviral prophylaxis or treatment when clinically indicated.
7. Cardiovascular and Malignancy Risk Assessment.
8. Documentation of baseline cardiovascular risk factors, including history of hypertension, diabetes mellitus, dyslipidemia, chronic kidney disease, prior stroke, heart failure, or ischemic heart disease.
9. Review of baseline laboratory parameters relevant to cardiovascular risk (e.g., lipid profile when available).
10. Assessment of prior malignancy history or active cancer status.
11. Patients who did not complete all predefined components within the specified time window were classified into the non-RMP group.
12. Baseline Covariates.
Baseline demographic and clinical variables collected included age, sex, body mass index, smoking and alcohol history, underlying autoimmune disease category, comorbidities, prior exposure to systemic immunosuppressive therapies (including cyclophosphamide and intravenous corticosteroids), baseline corticosteroid dose, and concomitant conventional DMARD use.
Clinical outcomes
The primary outcomes of interest included: Pneumocystis jirovecii pneumonia (PJP), defined by compatible clinical features, radiologic findings, and microbiological confirmation. Major adverse cardiovascular events (MACE), defined as a composite of acute myocardial infarction, ischemic stroke, hospitalization for heart failure, or cardiovascular death. All-cause mortality. Secondary outcomes included hepatitis B virus reactivation, tuberculosis reactivation, cytomegalovirus infection, and treatment-related respiratory adverse events. Outcomes were assessed within the first 6 months following treatment initiation and during extended follow-up up to 12 months.
Statistical analysis
Baseline characteristics were compared between the RMP and non-RMP groups using Student’s t test or Mann–Whitney U test for continuous variables and chi-square or Fisher’s exact test for categorical variables, as appropriate. Multivariable logistic regression models were constructed to estimate adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for associations between RMP implementation and clinical outcomes. Covariates included age, sex, comorbidities (including cardiovascular risk factors), disease category, baseline corticosteroid use, concomitant immunosuppressive therapy, and class of targeted agent. Time-to-event analyses were performed using Kaplan–Meier curves and Cox proportional hazards models to evaluate the temporal association between anti-CD20 therapy and the occurrence of PJP, MACE, and mortality. All statistical analyses were conducted using SPSS (version 29). A two-sided p value < 0.05 was considered statistically significant.
Results
Study population
Between October 1, 2014, and December 31, 2023, a total of 1,078 patients who received targeted therapies for autoimmune diseases were included in this retrospective analysis. Patients were classified according to the implementation of a predefined risk management plan (RMP) prior to treatment initiation.
Baseline characteristics and RMP implementation
Of the 1,078 patients, 348 (32.3%) fulfilled the predefined RMP criteria, whereas 730 (67.7%) did not, resulting in an approximate 1:2 distribution between the RMP and non-RMP groups. Patients in the RMP group were significantly older than those in the non-RMP group (mean age 50.6 ± 14.6 vs. 48.2 ± 17.0 years, p = 0.028). Autoimmune inflammatory joint diseases were more prevalent in the RMP group compared with the non-RMP group (69.9% vs. 49.3%, p < 0.001). Despite guideline recommendations, RMP implementation remained limited (32.3%). This gap reflects real-world constraints, including treatment urgency, reimbursement variability, and lack of standardized implementation systems, which are further explored in the discussion.
In contrast, patients in the non-RMP group had a significantly higher proportion of treatment with anti-CD20 agents (24.9% vs. 0.9%, p < 0.001) and biosimilars (10.1% vs. 3.2%, p < 0.001). Other agents such as anti-TNF alpha (47.7% vs. 37.9%, p = 0.002) and tsDMARDs/Janus kinase inhibitors (JAKi) (18.1% vs. 11.4%, p = 0.003) were more frequently used in the RMP group. Baseline demographic and clinical characteristics are summarized in Table 1 and drug distributions in Table 2.
Table 1.
Baseline Characteristics of Patients
| All patients | With RMP | Without RMP | ||
|---|---|---|---|---|
| Variable | n = 1078 (%) | n = 348 (%) | n = 730 (%) | p-value |
| Demographic characteristics | ||||
| Age, y | 49.0 ± 16.3 | 50.6 ± 14.6 | 48.2 ± 17.0 | 0.028 |
| Female | 591 (54.8) | 204 (58.6) | 387 (53.0) | 0.084 |
| BMI | 24.4 ± 6.9 | 24.5 ± 5.1 | 24.4 ± 7.7 | 0.680 |
| Smoke | 153 (14.2) | 38 (10.9) | 115 (15.8) | 0.033 |
| Drink | 107 (9.9) | 30 (8.6) | 77 (10.5) | 0.322 |
| Comorbidities at baseline | ||||
| HF | 36 (3.3) | 9 (2.6) | 27 (3.7) | 0.342 |
| COPD | 37 (3.4) | 10 (2.9) | 27 (3.7) | 0.487 |
| CKD | 31 (2.9) | 5 (1.4) | 26 (3.6) | 0.051 |
| Liver cirrhosis | 62 (5.8) | 29 (8.3) | 33 (4.5) | 0.012 |
| Stroke | 31 (2.9) | 6 (1.7) | 25 (3.4) | 0.118 |
| Hypertension | 233 (21.6) | 66 (19.0) | 167 (22.9) | 0.145 |
| Diabetes mellitus | 90 (8.3) | 30 (8.6) | 60 (8.2) | 0.824 |
| Hyperlipidemia | 116 (10.8) | 40 (11.5) | 76 (10.4) | 0.592 |
| Disease, n (%) | < 0.001 | |||
|
Autoimmune inflammatory joint diseases |
601 (55.8) | 241 (69.9) | 360 (49.3) | |
| Connective tissue diseases | 129 (12.0) | 6 (1.7) | 123 (16.8) | |
| Autoimmune skin disease | 272 (25.2) | 76 (21.8) | 196 (26.8) | |
|
Immune-mediated gastrointestinal diseases |
5 (0.5) | 0 (0) | 5 (0.7) | |
| Immune-mediated thrombocytopenia | 31 (2.9) | 20 (5.7) | 11 (1.5) | |
| Others | 40 (3.7) | 5 (1.4) | 35 (4.8) | |
| Previous systemic therapy, n (%) | ||||
| Cyclophosphamide (dose) | 32 (3.0) | 1 (0.3) | 31 (4.2) | < 0.001 |
| IV steroid (dose) | 195 (18.1) | 15 (4.3) | 180 (24.7) | < 0.001 |
| Concomitant DMARDs, n (%) | 794 (73.7) | 277 (79.6) | 517 (70.8) | 0.002 |
| HBV Treatment | 36 (3.3) | 16 (4.6) | 20 (2.7) | 0.112 |
| HBV Prophylaxis | 32 (3.0) | 12 (3.4) | 20 (2.7) | 0.522 |
| TB Treatment | 3 (0.3) | 2 (0.6) | 1 (0.1) | 0.202 |
| TB prophylaxis (INH) | 25 (2.3) | 7 (2.0) | 18 (2.5) | 0.643 |
Table 2.
Comparison of drug types between the RMP group and the non-RMP group
| Total | With RMP | Without RMP | p-value | |
|---|---|---|---|---|
| Anti-TNF alpha | 443 (41.1) | 166 (47.7) | 277 (37.9) | 0.002 |
| Anti-IL6 | 113 (10.5) | 35 (10.1) | 78 (10.7) | 0.753 |
| CTLA4-Ig | 41 (3.8) | 12 (3.4) | 29 (4.0) | 0.674 |
| Anti-CD20 | 185 (17.2) | 3 (0.9) | 182 (24.9) | < 0.001 |
| Anti-IL17 | 91 (8.4) | 43 (12.4) | 48 (6.6) | 0.001 |
| Anti-IL23 | 21 (1.9) | 15 (4.3) | 6 (0.8) | < 0.001 |
| Anti-IL12/23 | 30 (2.8) | 11 (3.2) | 19 (2.6) | 0.602 |
| Anti-Blys | 8 (0.7) | 0 (0.0) | 8 (1.1) | 0.050 |
| tsDMARDs/JAKi | 146 (13.5) | 63 (18.1) | 83 (11.4) | 0.003 |
| Biosimilars | 85 (7.9) | 11 (3.2) | 74 (10.1) | < 0.001 |
1.Anti-TNF alpha (Etanercept, Erelzi, Adalimumab, Idacio, Yuflyma, Golimumab, Infliximab, Certolizumab) 2. Anti-IL6 (Tocilizumab) 3. CTLA4-Ig (Abatacept) 4. Anti-CD20 (Rituximab, Truxima, Rixathon) 5. Anti-IL17 (Secukinumab, Brodalumab, Ixekinumab) 6. Anti-IL23 (Guselkumab) 7.Anti-IL12/23 (Ustekinumab) 8. Anti-Blys (Belimumab) 9. tsDMARDs: JAKi (Tofacitinib, Upadacitinib, Baricitinib) 10. biosimilar: Erelzi, Idacio, Yuflyma, Remsima,Truxima,Rixathon)
Adverse events and clinical outcomes according to RMP status
Infectious complications
The incidence of Pneumocystis jirovecii pneumonia (PJP) was significantly higher in the non-RMP group than in the RMP group (11.9% vs. 2.3%, p < 0.001).
Among patients identified as being at risk for opportunistic infections, 129 patients (12.1%) received Pneumocystis jirovecii pneumonia prophylaxis with sulfamethoxazole/trimethoprim. Notably, all PJP cases occurred in patients who did not receive prophylaxis. The results are presented in Supplementary Table 1.
Mortality and cardiovascular outcomes
During follow-up, the non-RMP group exhibited significantly higher rates of all-cause mortality (7.3% vs. 1.7%, p < 0.001) and major adverse cardiovascular events (MACE) (8.5% vs. 2.6%, p < 0.001) compared with the RMP group. Regarding cardiovascular risk management, 164 patients (15.3%) received antiplatelet or lipid-lowering therapy (aspirin and/or statins). Among patients who developed MACE, 21.4% were receiving preventive therapy compared with 14.9% among those without MACE. Only 2 patients (0.2%) were diagnosed with vasculitis-related cardiovascular events, both of whom were classified within the MACE group, indicating that most cardiovascular events were not attributable to primary vasculitis processes. These results are presented in Supplementary Table 2 and Table 3.
Table 3.
Outcomes
| All patients | With RMP | Without RMP | ||
|---|---|---|---|---|
| Variable | n = 1078 (%) | n = 348 (%) | n = 730 (%) | p-value |
| Adverse Events | ||||
| Hepatitis B reactivation | 50 (4.6) | 20 (5.7) | 30 (4.1) | 0.232 |
| TB reactivation | 44 (4.1) | 18 (5.2) | 26 (3.6) | 0.211 |
| PJP infection | 95 (8.8) | 8 (2.3) | 87 (11.9) | < 0.001 |
| CMV infection | 17 (1.6) | 5 (1.4) | 12 (1.6) | 0.799 |
| Treatment related respiratory events | 517 (48.0) | 149 (42.8) | 368 (50.4) | 0.020 |
| Death | 59 (5.5) | 6 (1.7) | 53 (7.3) | < 0.001 |
| MACE | 71 (6.6) | 9 (2.6) | 62 (8.5) | < 0.001 |
Treatment-related respiratory adverse events
Treatment-related respiratory adverse events were more frequently observed in the non-RMP group than in the RMP group (50.4% vs. 42.8%, p = 0.020). Clinical outcomes stratified by RMP status are presented in Table 3.
Multivariable analysis of clinical outcomes
After adjustment for age, sex, comorbidities, and class of targeted agent, RMP implementation remained independently associated with reduced risks of multiple adverse outcomes. Specifically, RMP was associated with: A significantly lower risk of PJP infection (adjusted odds ratio [aOR] 0.18, 95% CI 0.15–0.84). A reduced risk of MACE (aOR 0.30, 95% CI 0.13–0.71). A decreased risk of all-cause mortality (aOR 0.21, 95% CI 0.07–0.61). These primary outcomes and adjusted estimates are summarized in Table 4 (Multivariate analysis section).
Table 4.
Comparison of Clinical Outcomes and Risk Reduction Between Patients With and Without Risk Management Plan (RMP) Implementation
| Univariate analysis | Multivariate analysis | |||
|---|---|---|---|---|
| Patients no. (%) | With RMP | Without RMP | Crude OR (95% CI) | Adjusted OR (95% CI) |
| PJP | 8/348 | 87/730 | 0.17 (0.08–0.36) | 0.18 (0.15–0.84) |
| MACE | 9/348 | 62/730 | 0.29 (0.14–0.58) | 0.30 (0.13–0.71) |
| Death | 6/348 | 53/730 | 0.22 (0.10–0.53) | 0.21 (0.07–0.61) |
OR = odds ratio, CI = confidence interval
*Adjusted variables: sex, age, hyperlipidemia, renal failure, mental disorders
Subgroup analysis: anti-CD20 therapy and PJP risk
In a prespecified analysis of different targeted agents, anti-CD20 therapy demonstrated the most substantial association with PJP risk, with an odds ratio exceeding 10 compared to other therapies (as illustrated in Fig. 1).
Fig. 1.
Risk of PJP, MACE and Death among different targeted therapies in patients with autoimmune diseases. Data are presented as odds ratios (ORs) with 95% confidence intervals (CIs). The vertical line represents an OR of 1.0 (no difference in risk)
Furthermore, patients receiving anti-CD20 therapy experienced significantly earlier infectious and mortality events: The median time to PJP onset was 3.2 months in anti-CD20 users compared with 6.5 months in non-users. The median time to death occurred earlier among anti-CD20–treated patients (3.1 vs. 3.3 months). The median time to MACE was 5.6 months for anti-CD20 users vs. 10.3 months for non-users. Time-to-event analyses for anti-CD20 therapy are detailed in Table 5.
Table 5.
Time-to-event analyses for anti-CD20 therapy
| Survival months | MACE months | PJP months | |
|---|---|---|---|
| With anti CD-20 | 3.1 (2.3–11.5) | 5.6 (2.5–14.9) | 3.2 (1.8–11.2) |
| Without anti CD-20 | 3.3 (0.6–61.0) | 10.3 (1.2–31.7) | 6.5 (1.4–8.1) |
In subgroup analyses, IL-6 inhibitors were associated with a higher observed risk of MACE and all-cause mortality compared with several other treatment classes (Fig. 1B and C). However, these findings should be interpreted cautiously given potential confounding by indication.
Discussion
This retrospective real-world study demonstrates that adherence to a predefined RMP prior to targeted therapy initiation remains suboptimal in routine clinical practice, with nearly two-thirds of patients initiating biologic or targeted synthetic DMARDs without comprehensive pre-treatment safety evaluation. Although previous guidelines, such as those from the ACR and EULAR, have emphasized the importance of conducting RMP assessments [3, 5, 11, 12], real-world data on actual implementation rates remain lacking. Despite guideline recommendations, RMP implementation remained suboptimal, with only 32.3% of our patients completing all predefined assessments prior to treatment initiation. Several factors may explain this gap. First, in real-world practice, treatment urgency, particularly in patients with active or refractory disease, may lead clinicians to prioritize rapid initiation of targeted therapy over comprehensive pre-treatment evaluation. Second, variability in reimbursement policies and off-label use of biologics or biosimilars may reduce adherence to standardized safety protocols, especially among self-paying patients. Third, fragmented care and lack of integrated checklists or electronic reminders may contribute to incomplete documentation of RMP components. Finally, clinician perception of “low risk” patients, particularly younger individuals without overt comorbidities, may lead to underestimation of infection or cardiovascular risks, resulting in omission of RMP procedures. These findings highlight a critical implementation gap between guideline recommendations and routine clinical practice and underscore the need for system-level interventions, such as standardized electronic order sets or mandatory pre-biologic safety checklists.
Our findings suggest that RMP implementation is associated with a broad safety benefit that extends beyond traditional infection prevention. Specifically, patients who did not undergo RMP evaluation had significantly higher rates of PJP, all-cause mortality, and MACE. Multivariable analysis further confirmed this association, with adjusted odds ratios showing that RMP implementation was associated with a substantially lower odds of mortality. This suggests that the baseline assessment within the RMP implementation may facilitate the identification of high-risk individuals and the consideration of timely preventive strategies, which may contribute to improved clinical outcomes. It should also be acknowledged that RMP adherence may partly reflect unmeasured aspects of clinical judgment, healthcare access, or physician risk tolerance, and therefore may function as a surrogate marker of overall safety-oriented practice rather than a discrete therapeutic intervention. PJP remains a severe and potentially fatal opportunistic infection in patients with systemic autoimmune diseases, yet its risk is frequently underestimated in routine clinical practice. Recent population-based data from Taiwan’s National Health Insurance Research Database have demonstrated a significantly increased incidence of PJP among patients with polymyositis/dermatomyositis, systemic sclerosis, systemic lupus erythematosus, and Sjögren’s syndrome, underscoring the vulnerability of these populations [13]. This excess risk has been consistently attributed to cumulative exposure to systemic glucocorticoids and the concomitant use of immunosuppressive agents, highlighting the importance of comprehensive infection risk assessment in this setting. Our findings extend these observations by demonstrating that, among patients receiving anti-CD20 therapy, the absence of structured RMP evaluation is associated with a markedly increased risk of early-onset PJP, with most events occurring within the first three months after treatment initiation. Although host defense against PJP has traditionally been attributed primarily to T-cell–mediated immunity [14–16], emerging evidence indicates that B cells also play a non-negligible role in pulmonary immune responses against PJP [17–19]. B-cell depletion induced by anti-CD20 therapy may therefore synergize with glucocorticoid-induced T-cell dysfunction, creating a permissive immunologic milieu for rapid disease progression. Prior studies have consistently reported a strong association between rituximab exposure and severe PJP, often characterized by abrupt clinical deterioration [20–22]. In the study by Martin-Garrido et al., 88% of rituximab treated patients with PJP developed acute respiratory failure, more than half required intensive care unit admission, and the overall mortality rate reached 30% [23]. Notably, the mean interval between rituximab administration and PJP onset was approximately 76.8 ± 21.6 days, a finding that closely parallels our observation that the incidence of PJP peaks within the first three months of anti-CD20 therapy. Together, these findings support the need for heightened vigilance and early preventive strategies when initiating anti-CD20 therapy, particularly in patients with systemic autoimmune diseases and concurrent immunosuppressive exposure. The observation that all PJP cases occurred in patients without prophylaxis further supports the effectiveness of targeted preventive strategies within the RMP framework.
Beyond infectious complications, our study demonstrates a strong association between RMP implementation and reduced risks of MACE and all-cause mortality. Although RMPs were originally designed to mitigate infection-related complications of biologic and targeted synthetic DMARDs, our findings suggest that their impact extends to broader safety outcomes. Cardiovascular events and mortality in patients with autoimmune diseases are often precipitated by acute inflammatory insults, severe infections, or treatment-related complications rather than by slowly progressive atherosclerosis alone [24]. Accordingly, systematic pre-treatment risk assessment and early safety optimization may indirectly translate into improved cardiovascular and survival outcomes. Importantly, coronary vasculitis accounted for only a negligible proportion of cardiovascular events, suggesting that the observed MACE signal is more likely related to systemic inflammation, infection, or treatment-related factors rather than primary vasculitis mechanisms.
Notably, cardiovascular events in our cohort occurred relatively early after treatment initiation, suggesting that acute immune perturbation, infection-related systemic inflammation, or treatment-associated hemodynamic stress (e.g., infusion reactions) may play a more prominent role than cumulative cardiovascular risk.
The observed association between IL-6 inhibition and increased risks of MACE and mortality warrants careful interpretation. Patients receiving IL-6 inhibitors often represent a population with higher baseline inflammatory burden, refractory disease, or multiple comorbidities, which may confound this association. Therefore, this signal is more likely reflective of underlying patient risk rather than a direct adverse pharmacologic effect.
It is important to emphasize that RMPs should not be interpreted as direct cardiovascular interventions. Instead, they function as upstream risk stratification tools that enable appropriate therapy selection, heightened clinical vigilance, and timely preventive measures. The observed reductions in MACE and mortality associated with RMP adherence likely reflect a composite effect of improved patient selection, avoidance of high-risk therapeutic combinations, reduced early infectious complications, and closer follow-up during the early treatment period. This interpretation aligns with emerging evidence that cardiovascular safety signals associated with targeted therapies particularly JAK inhibitors are strongly modified by baseline patient risk profiles rather than representing uniform class effects [25].
Additionally, we observed that most patients receiving biosimilars did not receive a thorough RMP evaluation. These patients were typically self-paying individuals with autoimmune or dermatologic conditions, often excluded from insurance coverage, especially in the context of off-label use. This lack of assessment may contribute to a distorted perception among clinicians that biosimilars are associated with a higher incidence of adverse events compared to originator biologics. This increased risk is likely a consequence of inadequate risk management rather than the drug's safety itself, suggesting a potential gap in the implementation of safety oversight for non-reimbursed biologics.
Our study also noted that patients with liver cirrhosis were more likely to be included in the RMP group (8.3% vs. 4.5%, p = 0.012), indicating that clinicians maintain higher vigilance for patients with overt organ dysfunction. However, a significant gap remains among many "low-risk" patients who are younger or lack significant comorbidities often bypass RMP despite being prescribed high-potency immunosuppressants like anti-CD20 agents or JAK inhibitors. These findings support the notion that RMP implementation should be guided primarily by the intrinsic risk profile of the targeted agent rather than by patient age or apparent baseline health status alone.
Several limitations should be acknowledged. First, the retrospective design precludes causal inference, and residual confounding particularly confounding by indication cannot be fully excluded, especially for cardiovascular and mortality outcomes. Second, formal cardiovascular risk scores were not systematically available, and cardiovascular risk assessment was based on documented clinical risk factors. Finally, as a single-center study, the generalizability of our findings may be influenced by local practice patterns and reimbursement policies. RMP adherence may also reflect unmeasured aspects of clinical judgment and healthcare access. In addition, RMP implementation was treated as a binary exposure, and the relative contribution of individual RMP components (e.g., infection screening versus cardiovascular risk documentation) could not be disentangled.
Furthermore, subgroup analyses involving anti-CD20 therapy should be interpreted with caution, as the small number of patients receiving anti-CD20 agents within the RMP group may limit the precision and stability of effect estimates.
Conclusions
Our findings suggest that adherence to predefined RMPs prior to targeted therapy initiation is strongly associated with reduced risks of severe infection, cardiovascular events, and mortality in real-world practice. These results support the integration of standardized, consistently applied RMP strategies as essential components of precision safety management, particularly for high-risk targeted therapies such as anti-CD20 agents. Importantly, these findings suggest that suboptimal RMP implementation is not solely a clinician-level issue but reflects broader system-level barriers, including workflow fragmentation and lack of enforceable pre-treatment safety protocols.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors thank the Department of Medical Education and Research, Kaohsiung Veterans General Hospital, for their support. Editorial assistance was funded by UCB. The sponsor had no role in study design, data collection, analysis, interpretation, or the decision to submit the manuscript.
Authors’ contributions
LFW designed the study and performed data analysis. HLL and YHH contributed to study design and critically revised the manuscript. CHY assisted with data collection and analysis. All authors approved the final version of the manuscript.
Funding
This study received no external funding.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Compliance with ethical standards
Ethics approval and consent to participate
The study was approved by the Institutional Review Board of Kaohsiung Veterans General Hospital (IRB No. KSVGH23-CT12-14). The need for informed consent was waived owing to the retrospective design and use of anonymized data. The study was conducted in accordance with the Declaration of Helsinki.
Consent for publication
Not applicable.
Disclosures
None.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Smolen JS, Aletaha D, McInnes IB (2016) Rheumatoid arthritis. Lancet 388:2023–2038 [DOI] [PubMed] [Google Scholar]
- 2.Winthrop KL, Bathon J, Kerschbaumer A, Isaacs JD, Mease P, Gottenberg JE et al (2025) Chasing the target: reports from the Advances in Targeted Therapies meeting, 2024. Ann Rheum Dis 84:927–936 [DOI] [PubMed] [Google Scholar]
- 3.Fragoulis GE, Nikiphorou E, Dey M, Zhao SS, Courvoisier DS, Arnaud L et al (2023) 2022 EULAR recommendations for screening and prophylaxis of chronic and opportunistic infections in adults with autoimmune inflammatory rheumatic diseases. Ann Rheum Dis 82:742–753 [DOI] [PubMed] [Google Scholar]
- 4.Ytterberg SR, Bhatt DL, Mikuls TR, Koch GG, Fleischmann R, Rivas JL et al (2022) Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis. N Engl J Med 386:316–326 [DOI] [PubMed] [Google Scholar]
- 5.Smolen JS, Landewe RBM, Bergstra SA, Kerschbaumer A, Sepriano A, Aletaha D et al (2023) EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2022 update. Ann Rheum Dis 82:3–18 [DOI] [PubMed] [Google Scholar]
- 6.Takeuchi T, Kameda H (2010) The Japanese experience with biologic therapies for rheumatoid arthritis. Nat Rev Rheumatol 6:644–652 [DOI] [PubMed] [Google Scholar]
- 7.Testa U, Castelli G, Pelosi E, Galli E, Chiusolo P (2025) Toxicities associated with CAR-T cell therapies. Mediterr J Hematol Infect Dis 17:e2025039 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhu Y, Liu K, Rosen ST, Liu W, Zhu H (2025) Treatment-related adverse events of chimeric antigen receptor-T therapies for cancers in clinical trials: a systematic review and meta-analysis. EClinicalMedicine 84:103267 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Singh JA, Saag KG, Bridges SL Jr, Akl EA, Bannuru RR, Sullivan MC et al (2016) 2015 American College of Rheumatology guideline for the treatment of rheumatoid arthritis. Arthritis Rheumatol 68:1–26 [DOI] [PubMed] [Google Scholar]
- 10.Lin CT, Huang WN, Hsieh CW, Chen YM, Chen DY, Hsieh TY et al (2019) Safety and effectiveness of tocilizumab in treating patients with rheumatoid arthritis: a three-year study in Taiwan. J Microbiol Immunol Infect 52:141–150 [DOI] [PubMed] [Google Scholar]
- 11.Fraenkel L, Bathon JM, England BR, St Clair EW, Arayssi T, Carandang K et al (2021) 2021 American College of Rheumatology guideline for the treatment of rheumatoid arthritis. Arthritis Rheumatol 73:1108–1123 [DOI] [PubMed] [Google Scholar]
- 12.Bass AR, Chakravarty E, Akl EA, Bingham CO, Calabrese L, Cappelli LC et al (2023) 2022 American College of Rheumatology guideline for vaccinations in patients with rheumatic and musculoskeletal diseases. Arthritis Care Res (Hoboken) 75:449–464 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hsu HC, Chang YS, Hou TY, Chen LF, Hu LF, Lin TM et al (2021) Pneumocystis jirovecii pneumonia in autoimmune rheumatic diseases: a nationwide population-based study. Clin Rheumatol 40:3755–3763 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Thomas CF Jr, Limper AH (2004) Pneumocystis pneumonia. N Engl J Med 350:2487–2498 [DOI] [PubMed]
- 15.Gingerich AD, Norris KA, Mousa JJ (2021) Pneumocystis pneumonia: immunity, vaccines, and treatments. Pathogens 10:2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chen Y, Rong H, Li T, Zhang C, Yang H, Sun H et al (2025) T-cell subsets in Pneumocystis pneumonia. Clin Transl Immunol 14:e70055 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lund FE, Hollifield M, Schuer K, Lines JL, Randall TD, Garvy BA (2006) B cells are required for generation of protective effector and memory CD4 cells in response to Pneumocystis lung infection. J Immunol 176:6147–6154 [DOI] [PubMed] [Google Scholar]
- 18.Noell K, Dai G, Pungan D, Ebacher A, McCombs JE, Landry SJ et al (2022) Germline IgM predicts T cell immunity to Pneumocystis. JCI Insight 7:e17 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kolls JK (2017) An emerging role of B cell immunity in susceptibility to Pneumocystis pneumonia. Am J Respir Cell Mol Biol 56:279–280 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Martino EA, Caserta S, Vigna E, Bruzzese A, Amodio N, Lucia E et al (2025) Infections in chronic lymphocytic leukemia: evolving risks and prevention strategies. Eur J Haematol. 10.1111/ejh.70065 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Nixon A, Ogden L, Woywodt A, Dhaygude A (2017) Infectious complications of rituximab therapy in renal disease. Clin Kidney J 10:455–460 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Elsegeiny W, Eddens T, Chen K, Kolls JK (2015) Anti-CD20 antibody therapy and susceptibility to Pneumocystis pneumonia. Infect Immun 83:2043–2052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Martin-Garrido I, Carmona EM, Specks U, Limper AH (2013) Pneumocystis pneumonia in patients treated with rituximab. Chest 144:258–265 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Conrad N, Verbeke G, Molenberghs G, Goetschalckx L, Callender T, Cambridge G et al (2022) Autoimmune diseases and cardiovascular risk: a population-based study on 19 autoimmune diseases and 12 cardiovascular diseases in 22 million individuals in the UK. Lancet 400:733–743 [DOI] [PubMed] [Google Scholar]
- 25.Silvagni E, Bortoluzzi A, Occhino G, Bellelli G, Garaffoni C, Delvino P et al (2025) Frailty and cardiovascular safety of JAK inhibitors versus TNF inhibitors in rheumatoid arthritis: a real-world comparative study of drug effects and patient profiles. Front Pharmacol 16:1565909 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.