От эмпирического лечения к точному вмешательству: библиометрический анализ нейропсихиатрической системной волшебной системы с использованием нескольких баз данных (Erithematosus)2006-2025).

Introduction

Neuropsychiatric systemic lupus erythematosus (NPSLE) is among the most severe complications of systemic lupus erythematosus (SLE) ( 1 ). It presents with a highly heterogeneous spectrum of clinical manifestations, ranging from cognitive impairment and psychiatric symptoms to seizures and meningitis or encephalopathy. The clinical burden of NPSLE extends beyond the management of acute episodes, with profound and sustained long-term cognitive deterioration and substantial deterioration in patients’ everyday functioning and well-being. Despite diagnostic and therapeutic advancements over the past two decades, NPSLE management remains hindered by three core challenges: the profound heterogeneity and non-specificity of clinical symptoms; the absence of consensus diagnostic cutoffs and unified interpretive frameworks; and the limited representation of patients with active NPSLE in many pivotal SLE clinical trials. Consequently, there is a critical dearth of high-quality, evidence-based therapeutic strategies tailored specifically to this population ( 2 – 4 ).

To address these gaps, combining bibliometric and topic modeling methodologies can systematically characterize the evolution of NPSLE research, thereby objectively identifying the field’s foundational origins, emerging hotspots, and existing evidence gaps ( 5 , 6 ). Accordingly, based on literature retrieved from the Web of Science Core Collection and Scopus databases between 2006 and 2025, this study integrates BERTopic modeling with multidimensional bibliometric tools to map the knowledge architecture and thematic trajectories of NPSLE research. By consolidating pivotal clinical evidence, this work aims to provide an objective reference framework to guide the standardization of NPSLE diagnostics, the development of mechanism-driven stratified therapies, and the rational design of future clinical trials ( 6 ).

Materials and methods

Data sources and data cleaning

To ensure the robustness of our findings, literature was systematically retrieved from two major academic databases: the Web of Science Core Collection (WoSCC) and Scopus. In addition, PubMed was searched for related clinical trials. For WoSCC and Scopus, the search timeframe was from January 1, 2006, to December 31, 2025, and the final literature search was performed on January 16, 2026. For PubMed, the timeframe was from January 1, 1990, to December 31, 2025, and the final search was performed on March 14, 2026. Inclusion criteria were restricted to English-language publications, and document types were limited to original research articles and reviews ( 7 , 8 ). For PubMed, document types were restricted to clinical trials.

To ensure methodological consistency across different database systems, we clearly distinguished the search fields and terminology standards used in each database: specifically, WoSCC was searched using the Topic Search field (TS), whereas Scopus was searched using the Title/Abstract/Keywords field (ABS), and PubMed was searched using the Medical Subject Headings (MeSH) terms. The search strategy combined Medical Subject Headings (MeSH) terms and supplementary keywords related to neuropsychiatric lupus using Boolean operators. For PubMed, the search was conducted using MeSH terms (with additional keyword terms as needed); the complete database-specific query strings are provided in Supplementary Table 1 : WoSCC: TS = (“NPSLE” OR “neuropsychiatric lupus” OR (“neuropsychiatric manifestation” AND (“systemic lupus erythematosus” OR “SLE”)) OR (“central nervous system lupus” OR “CNS lupus”)), Scopus: TITLE-ABS-KEY (“NPSLE” OR “neuropsychiatric lupus” OR (“neuropsychiatric manifestation” AND (“systemic lupus erythematosus” OR “SLE”)) OR (“central nervous system lupus” OR “CNS lupus”)). The initial search yielded 1,260 records from WoSCC and 1,373 from Scopus. For PubMed, an initial 6,325 records were retrieved, and 73 records remained after restricting to clinical trials (for further screening). Ultimately, 1,569 unique publications were included in the final analysis, consisting of 1,244 research articles and 325 reviews ( Figure 1 ). Clinical trials were screened independently by two investigators, and discrepancies were resolved by a third reviewer. In addition, we included a consistency check, yielding an inter-reviewer agreement of κ = 0.89 (95% CI: 0.86–0.92), which strengthens the reliability and methodological rigor of the screening process. The pre-specified inclusion and exclusion criteria are provided in Supplementary Table 7 . Among the PubMed-retrieved clinical trials, 58 were excluded after full-text screening, and 15 representative trials were incorporated into the discussion ( Supplementary Table 1 ).

Bibliometric analysis

We used bibliometrix (v5.0) in R (v4.4.5) ( 9 ) to compute basic bibliometric statistics, including annual publication volume, core authors, journal distribution, and citation-related indicators. During the import into the bibliometrix R package, we applied the package’s disambiguation/standardization functions to normalize bibliographic records by removing non-alphabetic characters, punctuation marks, and redundant spaces, thereby reducing noise introduced by cross-database formatting differences. Prior to import, duplicate records across WoSCC and Scopus were identified and removed using custom Python scripts during data integration, ensuring that subsequent disambiguation/standardization and bibliometric analyses were conducted on harmonized unique records. Thematic evolution was explored using bibliometrix’s built-in thematic evolution function. For collaboration network construction (authors, countries, and institutions), we used VOSviewer (v1.6.20). Communities were identified using the VOS clustering approach together with modularity optimization. Network thresholds were set as follows: the minimum number of publications was 5 and the minimum number of citations was 0. CiteSpace (v6.2.R1) was used for keyword clustering, keyword burst detection, and citation burst analysis. Time slicing was set to 1-year intervals. Node types included authors, institutions, and keywords. Network pruning was performed using Pathfinder and pruning sliced networks. Key parameters were set as: g-index (k = 25), LRF = 2.5, L/N = 10, LBY = 5, and e = 1.0. Finally, outputs were visualized in Python (v3.14) using pandas, matplotlib, and seaborn ( 10 – 12 ).

Results

Publication trends and geographical distribution

A total of 1,569 publications met the inclusion criteria and were included in the final analysis ( Table 1 ). Analysis of the combined dataset showed an average annual growth rate of 6.04% in publication volume from 2006 to 2025. This growth was more pronounced in the Scopus database (7.25%) compared to the WoSCC database (4.7%), a difference likely attributable to the broader indexing scope of Scopus. Furthermore, curve fitting indicated an accelerated growth trend in overall literature output across the combined databases (R² = 0.82) ( Figure 2 , Table 1 ). Citation volume patterns were broadly comparable across databases, indicating consistent scholarly recognition of NPSLE research regardless of indexing source.

Geographically, publication output was highly concentrated ( Figure 2C ). China (n = 254, 16.2%), the United States (n = 246, 15.7%), and Japan (n = 136, 8.7%) were the top three contributing nations ( Figure 2D ). Collectively, the top ten countries accounted for 65.9% of the total global output ( Figure 2E ). Distinct patterns of international collaboration were observed among these leading nations. For instance, the multiple-country publication (MCP) rate for the United States was 23.6%, whereas Israel exhibited a markedly higher MCP rate of 46.3% ( Table 2 ), highlighting divergent national strategies for international research cooperation. At the institutional level, Tel Aviv University (n = 134) and the University of Campinas (n = 129) showed relatively high publication volumes and citation frequencies ( Figure 2D ). Note that institutional counts use full counting, whereas national counts use de-duplicated counting; therefore, these values are not directly comparable.

Analysis of author, country, and institutional output

To further characterize the research landscape of NPSLE from 2006 to 2025, we analyzed the productivity and output patterns of key authors, countries, and institutions ( Figure 3 , Table 3 ). At the individual author level, Chaim Putterman (n = 36), Simone Appenzeller (n = 32), and Gerda M. Steup-Beekman (n = 25) were the top contributors by publication volume ( Figure 3A , Table 3 ). These authors also demonstrated substantial academic impact, with h-indices of 22, 18, and 16, and total citation counts of 1,666, 907, and 904, respectively ( Table 3 ).

At the national level, China and the United States exhibited the most rapid growth in annual publication output, consistently leading global contributions ( Figure 3B ). As previously noted, China (n = 254), the United States (n = 246), and Japan (n = 136) were the top three countries by cumulative publication volume. Institutionally, Tel Aviv University, Leiden University, and Albert Einstein College of Medicine demonstrated a steady increase in annual publication volume, serving as the primary affiliations for the field’s leading scholars ( Figure 3C ).

Finally, an evaluation of overall author productivity using Lotka’s Law ( Figure 3D ) revealed a highly skewed distribution typical of bibliometric data: a small fraction of core authors contributed the majority of the literature, while the vast majority of authors produced only one or two publications.

Keyword co-occurrence and hotspot evolution

To map the thematic architecture and evolving priorities within the NPSLE domain from 2006 to 2025, we conducted a keyword co-occurrence analysis ( Figure 4 ; Supplementary Table 2 ). The most frequent keywords were “systemic lupus erythematosus” (n = 886), “neuropsychiatric lupus” (n = 396), and “human” (n = 258), forming the foundational lexicon of the field. Based on co-occurrence frequencies, the keyword network was partitioned into five primary thematic clusters: major clinical studies, biomarkers, neuropsychiatric lupus, cerebrospinal fluid analysis, and immunological characteristics. This clustering captures the overarching thematic framework and intrinsic structural linkages of NPSLE research ( Figure 4A ).

Keyword prominence evolved significantly over the study period. Burst analysis of the top 30 keywords ( Figure 4B ) revealed a clear temporal shift in research focus. Early-burst keywords were predominantly associated with fundamental disease characteristics and epidemiology, such as “central nervous system,” “psychiatric manifestations,” and “prevalence.” In contrast, keywords with sustained bursts from 2019 to 2025 shifted toward precision diagnostics and therapeutics, including “neuropsychiatric lupus erythematosus,” “immunosuppressive treatment,” and “hydroxychloroquine.” This transition reflects the field’s progression from foundational disease characterization toward targeted therapeutic intervention.

Annual fluctuations in keyword popularity further illustrate this evolutionary trajectory ( Figure 4C ). Core keywords were categorized into four primary domains: disease definition and diagnosis, clinical manifestations, biomarkers and laboratory indices, and imaging and therapeutics. A detailed temporal analysis of trending terms ( Figure 4D ) confirmed that the 2006–2012 period was dominated by definitional terms, such as “overt neuropsychiatric manifestations,” “definitions,” and “autoimmune disease.” Conversely, from 2013 to 2025, terms related to diagnostic evaluation and pharmacotherapy—such as “diffusion,” “infliximab,” “Montreal Cognitive Assessment,” and “efficacy”—exhibited marked increases in frequency, highlighting the emergence of novel clinical and investigative priorities.

Author, country, and institutional collaboration networks

Analysis of author collaboration networks partitioned the core authors into eight distinct clusters based on co-occurrence linkages ( Figure 5A ). As detailed in Supplementary Table 3 , George K. Bertsias (1,831 citations), Chaim Putterman (1,666 citations), and Yehuda Shoenfeld (1,513 citations) received the highest total citations. In contrast, Gerda M. Steup-Beekman (total link strength [TLS] = 96), Tom W. J. Huizinga (TLS = 86), and Simone Appenzeller (TLS = 76) led in collaborative network breadth. This distinction highlights variations among core authors regarding overall citation impact versus collaborative engagement.

At the national level, the international collaboration network ( Figure 5B ) identified the United States as a primary hub, maintaining extensive transnational collaborations with Italy, the United Kingdom, and Canada. The United States (12,662 citations), Italy (6,548 citations), and the United Kingdom (6,115 citations) ranked highest in total citations ( Supplementary Table 4 ). Furthermore, the United States (TLS = 258), the United Kingdom (TLS = 161), and Canada (TLS = 140) exhibited the highest total link strengths, underscoring their central roles in global research networks.

Institutional collaboration analysis ( Figure 5C ) demonstrated that Tel Aviv University (2,184 citations), Leiden University (1,892 citations), and Albert Einstein College of Medicine (1,657 citations) accrued the most citations ( Supplementary Table 5 ). Conversely, the University of Crete (TLS = 33), Albert Einstein College of Medicine (TLS = 31), and Tel Aviv University (TLS = 18) displayed the strongest collaborative linkages.

Finally, a dual-map overlay of journals ( Figure 5D ) characterized the interdisciplinary knowledge flow in NPSLE research. Citing articles were primarily published in molecular biology, immunology, and rheumatology journals (e.g., Lupus, Arthritis and Rheumatology), whereas the cited references were broadly distributed across general medicine, pharmacology, and neuroscience journals (e.g., Neurology, Brain).

Journal distribution and high-impact literature

The dissemination of research findings in this field demonstrates pronounced journal concentration ( Table 4 ). Lupus ranked first with 218 articles, 5,523 total citations, and an h-index of 41, establishing it as the primary publication venue and the most influential specialized journal in the NPSLE domain. Annals of the Rheumatic Diseases followed closely, garnering 4,470 total citations and an h-index of 18. Furthermore, leading rheumatology and immunology journals, including Autoimmunity Reviews, Rheumatology, and The Journal of Rheumatology, served as core publication platforms, with all top ten journals maintaining an h-index above 14.

High-impact literature was similarly concentrated within these leading journals ( Figure 6B , Table 5 ). The 2019 article by Fanouriakis et al., published in Annals of the Rheumatic Diseases, received the most global citations (n = 1,448) and the highest normalized total citation score (33.57), indicating an academic impact substantially above the field average. The research group led by G. K. Bertsias contributed four articles to the top 20, representing the highest concentration of high-impact research output by a single team. Additionally, highly cited literature frequently appeared in high-impact general medical and review journals. For instance, the 2007 article by D’Cruz et al. was published in The Lancet, which possessed a 2025 impact factor of 88.5.

Citation burst analysis of the referenced literature revealed a distinct temporal stratification ( Figure 6A ). Among the top 50 articles with the highest citation burst strengths, the early phase (2006–2012) was dominated by foundational literature establishing disease definitions and diagnostic criteria, which exhibited prolonged burst durations. In contrast, the recent phase (2019–2025) was characterized by studies detailing immune mechanisms, biomarkers, and therapeutic regimens. Specifically, the 2019 publication by Hanly et al. achieved the highest burst strength (47.88) across the evaluated timeframe. Furthermore, a relative hiatus in strong citation bursts was observed between 2013 and 2018. This period coincides with the transitional phase previously identified, during which the field’s research focus shifted from fundamental disease characterization to targeted therapeutic interventions.

Thematic clustering, attention dynamics, and evolutionary trajectories

To systematically map the thematic architecture and its evolution from 2006 to 2025, multidimensional topic modeling was employed ( Figure 7 ). First, a three-way Sankey diagram illustrated the associative network among core authors, their affiliated countries, and publishing journals ( Figure 7A ). In the Sankey diagram ( Figure 7A ), the thickest flows indicate the predominant linkages: Appenzeller and Putterman are mainly connected to major hubs such as the United States and China, whereas Shoenfeld is predominantly linked to institutions in Israel. These predominant country-to-journal pathways show that the associated publications are largely disseminated through specialized lupus and rheumatology-related journals such as Lupus and Rheumatology. This highlights the established pathways between knowledge generation and dissemination among the field’s core entities.

Regarding thematic structure, Multiple Correspondence Analysis (MCA) of keywords categorized the research into three principal clusters: (1) neuroinflammation and autoimmune mechanisms, (2) cognitive dysfunction and neuroimaging, and (3) antiphospholipid syndrome and biomarkers. These clusters directly correspond to three core investigative trajectories focusing on pathogenesis, clinical evaluation, and disease phenotyping, respectively ( Figure 7B ).

A thematic quadrant map further evaluated the centrality and novelty of these research themes ( Figure 7C ). Foundational terms (e.g., “systemic lupus erythematosus”) were positioned in the high-centrality, low-novelty quadrant. Themes such as “brain” and “autoantibodies” occupied the high-centrality, high-novelty quadrant, indicating well-developed and active research areas. Conversely, terms like “clinical cohort” appeared in the low-centrality, high-novelty quadrant, while “multiple sclerosis” was located in the low-centrality, low-novelty quadrant.

Finally, a thematic evolution timeline delineated the chronological shifts in research focus over the two-decade study period ( Figure 7D ). From 2006 to 2012, core themes focused on foundational disease characteristics and pathogenic mechanisms, such as antiphospholipid syndrome and inflammation. Between 2013 and 2018, the focus transitioned toward target organ and systemic involvement, encompassing the central nervous system. From 2019 to 2022, precision phenotyping of NPSLE emerged as the central theme. Most recently (2023–2025), advanced diagnostic and evaluative modalities—specifically neuroimaging (MRI), fluid biomarkers (cerebrospinal fluid), and psychiatric assessment (depression)—have become the primary research frontiers.

Fine-grained topic identification and temporal trajectories via BERTopic modeling

To complement the macroscopic thematic analysis, we applied the BERTopic framework to conduct fine-grained topic modeling on the included literature. This approach delineated specific thematic classifications, inter-topic associations, and chronological trajectories from 2006 to 2025 ( Figure 8 , Table 6 ; Supplementary Table 6 ). The model successfully identified nine core themes, providing comprehensive coverage of the NPSLE research landscape ( Table 6 ).

The three largest topics by document volume were “clinical manifestations of NPSLE” (n = 256), “autoimmune and clinical characteristics of SLE” (n = 121), and “NPSLE animal models and autoantibodies” (n = 117). The remaining themes encompassed core symptoms and diagnosis, neuroimaging (MRI), pediatric NPSLE, cognitive dysfunction, general SLE characteristics, and cerebrospinal fluid/serum biomarkers. Collectively, these themes represent the primary research domains of the field, spanning clinical phenotypes, fundamental mechanisms, diagnostic evaluations, and biomarker discovery.

The spatial distribution, inter-topic distances, and document affiliations of these nine themes are mapped in Figure 8A , illustrating the clustering patterns and boundaries of each topic. An inter-topic similarity matrix ( Figure 8B ) further quantified the associations among themes. “Autoimmune and clinical characteristics of SLE” and “general characteristics of SLE” exhibited the highest similarity (0.95). The similarity scores between “clinical manifestations of NPSLE” and both “core symptoms and diagnosis of NPSLE” and “cognitive dysfunction in SLE” were 0.85. Additionally, the similarity between “NPSLE neuroimaging (MRI)” and “cognitive dysfunction in SLE” was 0.87. These metrics highlight the strong interconnections between specific clinical and diagnostic research areas.

Analysis of temporal dynamics ( Figure 8C ) revealed the annual frequency trends for the nine core themes from 2006 to 2025. The “clinical manifestations of NPSLE” theme consistently remained the most frequent, demonstrating a sustained upward trajectory. Similarly, the annual frequencies of “NPSLE neuroimaging (MRI),” “cognitive dysfunction in SLE,” and “NPSLE cerebrospinal fluid/serum biomarkers” gradually increased over time. Independent temporal trajectories for each theme ( Figure 8D ) were evaluated using linear regression ( Supplementary Table 6 ). Statistically significant upward trends were confirmed for “clinical manifestations of NPSLE” (Topic 0; R² = 0.498, P = 0.001), “autoimmune and clinical characteristics of SLE” (Topic 1; R² = 0.390, P = 0.004), “cognitive dysfunction in SLE” (Topic 6; R² = 0.468, P = 0.002), and “general characteristics of SLE” (Topic 7; R² = 0.439, P = 0.007). The trends for the remaining themes were not statistically significant ( Figure 8D ; Supplementary Table 6 ).

Discussion

Discussion based on bibliometric results

Within the context of systemic lupus erythematosus (SLE) research, the conceptualization of NPSLE has shifted from a rare, severe subtype to a recognized, prevalent complication. Our findings demonstrate sustained growth in publication volume over the past two decades, reflecting the field’s evolution from a niche area into an interdisciplinary research hotspot ( 13 , 14 ). Early research primarily focused on disease definition and diagnostic criteria, whereas contemporary efforts have increasingly emphasized elucidating pathogenic mechanisms and optimizing targeted therapies. Geographically, China and the United States have emerged as the leading contributors ( 15 ). Parallel to this growing academic interest, the diagnostic approaches to NPSLE have evolved significantly. While early frameworks relied heavily on clinical symptoms, recent advancements have driven the development of a multidimensional diagnostic approach that integrates neuroimaging, fluid biomarkers, and clinical phenotyping. In particular, combining neuroimaging metrics with cognitive assessments has refined objective diagnostic algorithms for NPSLE ( 16 , 17 ).

These diagnostic advancements have driven a parallel shift in therapeutic strategies, moving from broad-spectrum corticosteroids toward targeted immunotherapies ( 18 ). B-cell-targeted therapies remain a major research focus, and the neuroprotective efficacy of hydroxychloroquine has been increasingly recognized, establishing it as an evidence-based recommendation for NPSLE management ( 19 , 20 ). Overall, the field has followed a logical continuum: establishing diagnostic frameworks (dominant in the early years), elucidating underlying pathogenic mechanisms, and most recently developing and optimizing targeted therapies. This data-driven trajectory mirrors the paradigm shift observed in research hotspots over time.

Furthermore, the heterogeneous development of NPSLE research is reflected in global collaboration networks. A distinct contrast exists between the two leading nations: China produces the highest volume of publications, while the United States leads in international collaboration and overall citation impact ( 21 ). Additionally, author productivity is highly concentrated among a small group of core researchers. While this concentration supports continuity in foundational research, it may also pose a barrier to entry for new investigators.

Based on the developmental trajectories and topic modeling results of the past two decades, several key frontiers in NPSLE research can be identified. Current hotspots focus on multimodal neuroimaging-informed assessment together with cognitive performance, multimodal biomarker discovery, and dedicated studies of pediatric NPSLE ( 22 ). Collectively, these trends indicate a broader shift in clinical research from qualitative phenotypic categorization to quantitative assessment, and from reactive management to proactive risk stratification and early intervention.

Integration of core evidence-based paradigms in the clinical management of NPSLE

Building on the bibliometric analysis, we conducted a systematic review and evidence synthesis of core clinical research, including 15 clinical trials, diagnostic accuracy studies, and observational cohorts ( Table 7 ). This section evaluates the evidence base and existing gaps in NPSLE clinical management across three key dimensions: the evolution of diagnostic systems, the characterization of clinical risk factors, and the evidence supporting therapeutic strategies. This synthesis grounds the macroscopic bibliometric trends in clinical data and establishes a foundation for discussing pathogenesis and novel therapeutics.

Current research indicates a shift in NPSLE diagnosis, moving from structural imaging alone to multimodal diagnostic workflows that incorporate functional imaging, reproducible neurocognitive testing, and biomarker readouts ( 24 , 25 ). Regarding structural imaging, a prospective study by Katsumata et al. demonstrated that abnormal signals from large lesions (diameter ≥ 10mm) were exclusively observed in patients with central nervous system (CNS) involvement and significantly correlated with clinical outcomes (RR 3.7, 95% CI 2.9-4.7, P = 0.0002). However, the limited sensitivity of conventional MRI to minute lesions, diffuse alterations, and subclinical changes highlights the limitations of relying solely on structural imaging.

Functional imaging addresses these diagnostic limitations ( 26 ). Studies by Kovacs et al. and Padovan et al. reported that the abnormality detection rate of SPECT in NPSLE reached 85%, significantly outperforming conventional structural MRI. Furthermore, Lee et al. found that the positive detection rate of 18F-FDG PET/CT in MRI-negative patients was 75%, most frequently implicating the temporal, occipital, and frontal lobes. These findings demonstrate that functional imaging effectively mitigates the blind spots of conventional MRI, making it a valuable adjunctive diagnostic tool. Ultimately, combining functional and structural imaging improves diagnostic sensitivity and provides novel insights into the complex pathophysiology of NPSLE.

In terms of neurocognitive assessment, studies by Monastero et al. and Nowicka-Sauer et al. demonstrated a high overall prevalence of cognitive impairment (57%) in patients with SLE, regardless of overt neuropsychiatric manifestations. Notably, severe cognitive deterioration was exclusively observed in patients with overt NPSLE. This indicates that cognitive impairment is both a common complication of SLE and a key indicator of disease severity, providing a strong rationale for routine cognitive screening ( 27 ). Regarding biomarkers, Vogelgesang et al. provided initial evidence that cerebrospinal fluid quinolinic acid levels correlate significantly with SLE-associated neuropsychiatric manifestations. Concurrently, Su et al. characterized the relationship between anti-U1RNP, anti-Sm, and anti-ribosomal P antibodies and oxidative stress, identifying potential targets for the objective diagnosis and etiologic phenotyping of NPSLE ( 28 , 29 ).

Although these investigations establish a foundation for a multidimensional diagnostic approach to NPSLE, the existing evidence base retains notable limitations. Consensus cutoff values and a harmonized interpretive framework across these diverse diagnostic modalities have not yet been established. Furthermore, symptom attribution relies heavily on clinical judgment; current attribution models yield positive predictive values (PPVs) ranging from 70% to 85% and lack an objective gold standard ( 30 ). Finally, specific populations—particularly pediatric NPSLE cohorts—face unique clinical challenges, including the absence of dedicated diagnostic criteria and frequent diagnostic delays.

Pathogenesis of NPSLE

The pathogenesis of neuropsychiatric systemic lupus erythematosus (NPSLE) is highly heterogeneous, involving multiple factors and pathways; consequently, a unifying pathophysiological model has yet to be established ( 20 ). Nevertheless, current evidence indicates that the development of NPSLE involves a sequence of interconnected events that form a complex pathological network ( Figure 9 ).

Central to this process is the disruption of the neuroimmune barrier. As the primary defense mechanism of the central nervous system (CNS), the compromise of the blood-brain barrier (BBB) is a critical prerequisite for the cerebral infiltration of pathogenic autoantibodies ( 31 , 32 ). Antiphospholipid and anti-endothelial cell antibodies can activate vascular endothelial cells, upregulating the expression of adhesion molecules (e.g., ICAM-1 and VCAM-1) and increasing vascular permeability. Concurrently, “second-hit” events, such as infections or physiological stress, can induce transient BBB disruption, enabling circulating pathogenic factors to invade the CNS. Furthermore, the blood-cerebrospinal fluid (CSF) barrier, the meningeal barrier, and the recently characterized central lymphatic system also facilitate the permeation of immune cells and inflammatory mediators. Dysregulation of these barriers provides the structural basis for subsequent autoantibody influx and inflammatory cascades.

Following neuroimmune barrier disruption, brain-reactive autoantibodies enter the CNS and exert direct neurotoxic effects ( 33 ). Anti-NMDAR antibodies bind to the NR2A/B subunits of cerebral glutamate receptors, inducing excessive calcium influx and neuronal apoptosis via mechanisms analogous to glutamate excitotoxicity. Anti-ribosomal P protein antibodies target neuronal surface P antigens (NSPA) and perturb the synaptic modulatory functions of NMDARs; notably, within limbic regions such as the hippocampus and amygdala, this interference contributes to memory impairment and affective dysregulation. Conversely, anti-aquaporin-4 (AQP4) antibodies mediate astrocyte toxicity, which is implicated in NPSLE cohorts with concomitant demyelinating lesions. The pathogenicity of these autoantibodies has been validated in animal models, and adoptive transfer experiments confirm their direct pathogenic roles.

Concurrently, localized glial cells undergo aberrant activation. Type I interferon (IFNα)-mediated microglial activation is a primary pathogenic mechanism in NPSLE ( 34 ). Activated microglia secrete proinflammatory cytokines—including IL-6, IL-8, TNF, and TWEAK—and mediate abnormal synaptic loss via complement C1q-dependent synaptic pruning. This “innate immune memory” phenotype, imprinted on microglia following inflammatory stimulation, can persist long after the systemic inflammatory insult subsides. This enduring activation helps explain why NPSLE symptoms often occur independently of systemic SLE disease activity. Consequently, elevated CSF levels of IFNα, IL-6, CXCL10, and CCL2 have emerged as robust biomarkers of CNS-intrinsic inflammation.

Beyond cytokine-mediated inflammation, the complement cascade plays a pivotal role in NPSLE pathology ( 35 , 36 ). Histopathological studies of NPSLE brain tissue demonstrate localized complement activation within the CNS. Complement components, particularly C5a and C1q, further disrupt BBB integrity, facilitate peripheral leukocyte infiltration, and amplify microglial activation. Although studies on the correlation between CSF and serum complement levels remain inconclusive, the emerging clinical efficacy of complement-targeted therapeutics (e.g., eculizumab) in refractory NPSLE supports the pathogenic role of complement activation in specific endotypes.

Alongside these innate immune mechanisms, adaptive immunity is significantly involved in NPSLE pathophysiology ( 37 ). T cells (including double-negative T cells and T follicular helper cells), B cells, and plasma cells can infiltrate the brain parenchyma via perivascular spaces, predominantly breaching the CNS through the blood-CSF barrier at the choroid plexus. These infiltrating cells can synthesize pathogenic autoantibodies intrathecally, further activating glia and sustaining a localized neuroinflammatory environment that exacerbates neurotoxicity.

Recent advances in NPSLE therapeutics: from broad-spectrum immunosuppression to targeted interventions

Therapeutic research in NPSLE has transitioned from reliance on broad-spectrum corticosteroids toward optimized immunosuppressive regimens and targeted therapies ( 23 ). Regarding conventional immunosuppression, a randomized controlled trial by Barile-Fabris et al. demonstrated that the clinical response to intravenous cyclophosphamide in severe NPSLE significantly exceeded that of intravenous methylprednisolone (RR 2.05, 95% CI 1.13–3.73). This finding provides strong evidence for its use as a first-line treatment in severe cases. Furthermore, studies by Ramos et al. and Stojanovich et al. showed that combining low-dose intravenous cyclophosphamide with prednisone yielded a 96% response rate alongside a favorable tolerability profile. This regimen offers an effective, lower-toxicity alternative for patients intolerant to high-dose immunosuppressive therapy. Collectively, these studies establish cyclophosphamide as a central component in the clinical management of NPSLE ( 38 ).

Regarding biologic agents, a meta-analysis by Xie et al. demonstrated that belimumab does not significantly increase the risk of psychiatric disorders (OR 0.89, 95% CI 0.64–1.23) or all-cause mortality (OR 1.10, 95% CI 0.64–1.89) in SLE patients with concurrent neuropsychiatric symptoms, thereby supporting its safety in clinical practice ( 39 , 40 ). This evidence gap is largely attributable to the frequent exclusion of patients with active NPSLE from major SLE registration studies. Consequently, mainstream targeted therapeutics—such as rituximab and anifrolumab—lack dedicated, high-quality, randomized controlled data specific to the NPSLE population ( 41 , 42 ). The disparity between the clinical burden of NPSLE and the low quality of evidence for its treatment presents a critical gap that must be addressed in the rigorous design of future clinical trials. The evolution of therapeutic strategies and the core evidence are summarized in Table 8 . Despite these findings, the overall quality of evidence for novel targeted NPSLE therapies remains limited ( 43 ).

Management of NPSLE and future perspectives

Through a comprehensive bibliometric analysis of 1,569 NPSLE-related publications from 2006 to 2025, this study delineates the 20-year academic trajectory and evolving research paradigms within the field ( 44 ). Integrating recent advances in pathogenesis, diagnostic technologies, and therapeutics, it is evident that NPSLE clinical management is transitioning from empirical, symptom-directed management toward mechanism-informed, stratified interventions. However, this transition is constrained by incomplete diagnostic agreement, substantial gaps in evidence-based medicine, and the complex heterogeneity of NPSLE pathogenesis. Future research must address these challenges by optimizing diagnostic frameworks, stratifying treatment strategies, standardizing clinical research, advancing translational medicine, and fostering global collaboration. These efforts are essential to improve clinical management, long-term prognoses, and patient quality of life.

Regarding the diagnostic framework, constructing a multidimensional approach with agreed-upon interpretive rules is critical. Specifically, this requires integrating clinical phenotypes, multimodal imaging, fluid biomarkers, and standardized neurocognitive scales. This integration necessitates optimizing symptom attribution models and interpretative thresholds, alongside developing dedicated diagnostic criteria for specific populations, such as pediatric cohorts. Particular emphasis should be placed on establishing interpretable links between neuroimaging metrics and cognitive performance, together with clinically actionable decision rules to provide quantitative benchmarks for clinical practice.

In the clinical trial domain, initiating multicenter, large-scale randomized controlled trials (RCTs) dedicated exclusively to NPSLE is urgently needed. The systematic exclusion of patients with active NPSLE from pivotal SLE trials must be addressed. It is essential to evaluate the efficacy and safety of novel targeted therapeutics—such as CAR-T cell therapy, BTK inhibitors, Type I interferon pathway inhibitors, and CNS-penetrant ACE inhibitors—specifically within NPSLE cohorts ( 45 ). A critical priority is establishing a standardized clinical endpoint evaluation system. For cognitive dysfunction—a highly prevalent yet challenging endpoint to quantify—exploring imaging or fluid biomarkers as surrogate endpoints is essential to reduce trial complexity and improve outcome comparability.

In translational research, efforts to elucidate NPSLE pathogenesis must be expanded. Key objectives include deciphering the mechanisms of neuroimmune barrier disruption, identifying how microglial “innate immune memory” drives chronic injury, and clarifying the regulatory dynamics of the gut-brain axis. Concurrently, optimizing animal models and applying multi-omics strategies are crucial to translating novel interventions—such as blood-brain barrier neuroprotection and microglial modulation—into clinical practice. Furthermore, tailored investigations into distinct subsets, notably pediatric NPSLE, require increased attention.

Synergistic advancement across diagnostics, clinical trials, and translational research is critical to resolving the disparity between the high disease burden of NPSLE and the low quality of therapeutic evidence. Ultimately, these efforts should facilitate the transition from empiric management to mechanism-driven, precision interventions.

Limitations

Several methodological limitations should be considered when interpreting these findings. First, to ensure standardized citation metrics and cross-study comparability, this analysis was restricted to English-language publications. While this criterion maintains methodological consistency, it inevitably excludes relevant research published in other languages, such as Chinese, Japanese, or Korean. Although these studies hold contextual value, their integration into the global citation network is generally more limited.

Furthermore, inherent discrepancies exist between the Web of Science Core Collection (WoSCC) and Scopus regarding indexing scope, update frequency, and document types. Specifically, WoSCC offers a more extensive historical archive and a stable citation structure, whereas Scopus provides a broader indexing scope and faster update cycles. To reconcile these differences, cross-database deduplication and metadata standardization protocols were implemented. Nevertheless, subtle variations in indexing logic may still marginally affect the precise mapping of institutional affiliations and collaborative networks. Additionally, the complete standardization of author and institutional names is limited by the capabilities of current bibliometric software. Moreover, because the present study was based on bibliometric metadata (e.g., authors, affiliations, countries, and citations) rather than structured extraction from full texts, we were unable to systematically quantify institution-specific responsibilities in multicenter studies (e.g., patient recruitment, biomarker testing, and imaging assessment). Consequently, the publication output at the country and institutional levels may not be directly comparable because different counting conventions were applied.

However, given the large sample size and the extensive timeframe evaluated, these confounding variables are unlikely to meaningfully alter the overarching trends observed. Ultimately, bibliometric analyses provide a retrospective overview of academic progress and cannot replace prospective clinical trials or foundational mechanistic studies. Within this context, the present study provides an objective, quantitative framework for understanding the evolutionary trajectory of NPSLE research.

Conclusion

Based on a comprehensive bibliometric analysis of 1,569 publications retrieved from the Web of Science Core Collection and Scopus databases, integrated with evidence from core clinical trials, this study systematically delineates the evolutionary trajectory of neuropsychiatric systemic lupus erythematosus (NPSLE) research from 2006 to 2025, including diagnostic and therapeutic advancements and central bottlenecks. Our findings reveal a robust average annual growth rate of 6.04% in publication volume over the past two decades, with China and the United States emerging as the preeminent global contributors. Importantly, the epicenter of research interest has undergone a profound paradigm shift from rudimentary disease definition and foundational diagnosis toward increasingly precision-oriented therapeutics and deeper mechanistic exploration. Concurrently, diagnostic frameworks are rapidly evolving into multidimensional, highly integrated paradigms, while therapeutic strategies are experiencing a definitive shift from broad-spectrum immunosuppression to precision-targeted interventions.

Despite these advances, the field remains critically encumbered by major systemic challenges, including the lack of consensus diagnostic cutoff values and a dearth of robust evidence-based support for targeted therapeutics. Consequently, NPSLE continues to face a persistent developmental paradox characterized by high clinical prevalence alongside comparatively low-quality evidence. Moving forward, it is essential to prioritize the construction of standardized diagnostic and evaluative ecosystems, initiate large-scale multicenter clinical trials exclusively dedicated to patients with NPSLE, and further deepen translational research into underlying pathogenesis. Addressing these priorities will catalyze the transition of NPSLE clinical management from empiric symptomatic treatment to mechanism-driven, precision interventions, thereby improving long-term patient outcomes and overall functional status.

Edited by: Neeraj Kumar , National Institute of Pathology (ICMR), India

Reviewed by: Jose Luis Aguilar , Universidad Peruana Cayetano Heredia, Peru

Lijuan Jiang , Huazhong University of Science and Technology, China

Data availability statement

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

Author contributions

G-LD: Writing – review & editing, Supervision, Conceptualization, Funding acquisition, Project administration. MB: Methodology, Investigation, Formal analysis, Visualization, Software, Data curation, Writing – original draft. XZ: Data curation, Writing – original draft, Methodology. J-SW: Investigation, Writing – review & editing, Data curation. XJ: Writing – review & editing, Supervision, Data curation.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

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.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fimmu.2026.1846304/full#supplementary-material

References

1 Kiriakidou M Ching CL . Systemic lupus erythematosus . Ann Intern Med . ( 2020 ) 172 : Itc81 – itc96 . doi: 10.7326/aitc202006020 . PMID: 32479157

2 Legge AC Hanly JG . Recent advances in the diagnosis and management of neuropsychiatric lupus . Nat Rev Rheumatol . ( 2024 ) 20 : 712–28 . doi: 10.1038/s41584-024-01163-z . PMID: 39358609

3 Rice-Canetto TE Joshi SJ Kyan KA Siddiqi J . Neuropsychiatric systemic lupus erythematosus: a systematic review . Cureus . ( 2024 ) 16 : e61678 . doi: 10.7759/cureus.61678 . PMID: 38975440 PMC11227614

4 Sarwar S Mohamed AS Rogers S Sarmast ST Kataria S Mohamed KH . Neuropsychiatric systemic lupus erythematosus: a 2021 update on diagnosis, management, and current challenges . Cureus . ( 2021 ) 13 : e17969 . doi: 10.7759/cureus.17969 . PMID: 34667659 PMC8516357

5 Xie X Yu H He Y Li M Yin F Zhang X . Bibliometric analysis of global literature productivity in systemic lupus erythematosus from 2013 to 2022 . Clin Rheumatol . ( 2024 ) 43 : 175–87 . doi: 10.1007/s10067-023-06728-z . PMID: 37668951

6 Koo M . Systemic lupus erythematosus research: a bibliometric analysis over a 50-year period . Int J Environ Res Public Health . ( 2021 ) 18 : 7095 . doi: 10.3390/ijerph18137095 . PMID: 34281030 PMC8295925

7 Yan YL Su YS Cao JJ Lin J Sun JL Zhang CF . Research trends of the aetiology of systemic lupus erythematosus in the past 10 years: a bibliometric analysis, (2014-2023) . Clin Exp Rheumatol . ( 2025 ) 43 : 1650–60 . doi: 10.55563/clinexprheumatol/yvqg5k . PMID: 40314992

8 Zhang D Liang C Yin Q Li Y . Trends and collaborations in discoid lupus erythematosus research: a bibliometric analysis from 2010 to 2024 . Front Med (Lausanne) . ( 2025 ) 12 : 1556976 . doi: 10.3389/fmed.2025.1556976 . PMID: 40620431 PMC12226284

9 Yu H Xie Y Zuo M Xu J Jiang L Liu T . Mapping theme evolution and identifying hotspots in biomarkers of systemic lupus erythematosus based on global research . biomark Med . ( 2024 ) 18 : 321–32 . doi: 10.2217/bmm-2023-0774 . PMID: 38648095 PMC11218803

10 Synnestvedt MB Chen C Holmes JH . Citespace ii: visualization and knowledge discovery in bibliographic databases . AMIA Annu Symp Proc . ( 2005 ) 2005 : 724–8 . PMC1560567 16779135

11 Arruda H Silva ER Lessa M Proença D Bartholo R . Vosviewer and bibliometrix . J Med Libr Assoc . ( 2022 ) 110 : 392–5 . doi: 10.5195/jmla.2022.1434 . PMID: 36589296 PMC9782747

12 Bamahel AS Liu J Zhang Z Deng Y Xue C Sun X . Global research trends on macrophage polarization in osteoarthritis: a bibliometric analysis . Front Immunol . ( 2025 ) 16 : 1659938 . doi: 10.3389/fimmu.2025.1659938 . PMID: 41181141 PMC12575332

13 Govoni M Bortoluzzi A Padovan M Silvagni E Borrelli M Donelli F . The diagnosis and clinical management of the neuropsychiatric manifestations of lupus . J Autoimmun . ( 2016 ) 74 : 41 – 72 . doi: 10.1016/j.jaut.2016.06.013 . PMID: 27427403

14 Hanly JG . Diagnosis and management of neuropsychiatric sle . Nat Rev Rheumatol . ( 2014 ) 10 : 338–47 . doi: 10.1038/nrrheum.2014.15 . PMID: 24514913

15 Fanouriakis A Kostopoulou M Alunno A Aringer M Bajema I Boletis JN . 2019 update of the eular recommendations for the management of systemic lupus erythematosus . Ann Rheum Dis . ( 2019 ) 78 : 736–45 . doi: 10.1136/annrheumdis-2019-215089 . PMID: 30926722

16 Wójcicka-Frankiewicz A Pawlak MA Kalinowska A . Clinical, immunological and neuroimaging spectrum of cns lupus: can we reliably differentiate it from ms ? Neurol Neurochir Pol . ( 2024 ) 58 : 557–68 . doi: 10.5603/pjnns.103538 . PMID: 39737583

17 Postal M Lapa AT Reis F Rittner L Appenzeller S . Magnetic resonance imaging in neuropsychiatric systemic lupus erythematosus: current state of the art and novel approaches . Lupus . ( 2017 ) 26 : 517–21 . doi: 10.1177/0961203317691373 . PMID: 28394232

18 Magro-Checa C Zirkzee EJ Huizinga TW Steup-Beekman GM . Management of neuropsychiatric systemic lupus erythematosus: current approaches and future perspectives . Drugs . ( 2016 ) 76 : 459–83 . doi: 10.1007/s40265-015-0534-3 . PMID: 26809245 PMC4791452

19 Moore E Huang MW Putterman C . Advances in the diagnosis, pathogenesis and treatment of neuropsychiatric systemic lupus erythematosus . Curr Opin Rheumatol . ( 2020 ) 32 : 152–8 . doi: 10.1097/bor.0000000000000682 . PMID: 31895125 PMC7548105

20 Schwartz N Stock AD Putterman C . Neuropsychiatric lupus: new mechanistic insights and future treatment directions . Nat Rev Rheumatol . ( 2019 ) 15 : 137–52 . doi: 10.1038/s41584-018-0156-8 . PMID: 30659245 PMC8023338

21 Bertsias GK Ioannidis JP Aringer M Bollen E Bombardieri S Bruce IN . Eular recommendations for the management of systemic lupus erythematosus with neuropsychiatric manifestations: report of a task force of the eular standing committee for clinical affairs . Ann Rheum Dis . ( 2010 ) 69 : 2074–82 . doi: 10.1136/ard.2010.130476 . PMID: 20724309

22 Carrión-Barberà I Salman-Monte TC Vílchez-Oya F Monfort J . Neuropsychiatric involvement in systemic lupus erythematosus: a review . Autoimmun Rev . ( 2021 ) 20 : 102780 . doi: 10.1016/j.autrev.2021.102780 . PMID: 33609799

23 Barile-Fabris L Ariza-Andraca R Olguín-Ortega L Jara LJ Fraga-Mouret A Miranda-Limón JM . Controlled clinical trial of iv cyclophosphamide versus iv methylprednisolone in severe neurological manifestations in systemic lupus erythematosus . Ann Rheum Dis . ( 2005 ) 64 : 620–5 . doi: 10.1136/ard.2004.025528 . PMID: 15769918 PMC1755456

24 Vacca S Porcu M Piga M Mannelli L Chessa E Suri JS . Structural brain mr imaging alterations in patients with systemic lupus erythematosus with and without neuropsychiatric events . AJNR Am J Neuroradiol . ( 2024 ) 45 : 802–8 . doi: 10.3174/ajnr.A8200 . PMID: 38637023 PMC11288607

25 Tomalla V Schmeisser MJ Weinmann-Menke J . Mouse models, antibodies, and neuroimaging: current knowledge and future perspectives in neuropsychiatric systemic lupus erythematosus (npsle) . Front Psychiatry . ( 2023 ) 14 : 1078607 . doi: 10.3389/fpsyt.2023.1078607 . PMID: 36970286 PMC10031066

26 Bortoluzzi A Scirè CA Govoni M . Attribution of neuropsychiatric manifestations to systemic lupus erythematosus . Front Med (Lausanne) . ( 2018 ) 5 : 68 . doi: 10.3389/fmed.2018.00068 . PMID: 29594125 PMC5861139

27 Chalhoub NE Luggen ME . Screening for cognitive dysfunction in systemic lupus erythematosus: the montreal cognitive assessment questionnaire and the informant questionnaire on cognitive decline in the elderly . Lupus . ( 2019 ) 28 : 51–8 . doi: 10.1177/0961203318815299 . PMID: 30482092

28 Hirohata S Sakuma Y Matsueda Y Arinuma Y Yanagida T . Role of serum autoantibodies in blood brain barrier damages in neuropsychiatric systemic lupus erythematosus . Clin Exp Rheumatol . ( 2018 ) 36 : 1003–7 . doi: 10.26226/morressier.56e174d7d462b8028d88aba5 29846157

29 Hanly JG Su L Urowitz MB Romero-Diaz J Gordon C Bae SC . A longitudinal analysis of outcomes of lupus nephritis in an international inception cohort using a multistate model approach . Arthritis Rheumatol . ( 2016 ) 68 : 1932–44 . doi: 10.1002/art.39674 . PMID: 26991067 PMC5858760

30 Magro-Checa C Zirkzee EJ Beaart-van de Voorde LJJ Middelkoop HA van der Wee NJ Huisman MV . Value of multidisciplinary reassessment in attribution of neuropsychiatric events to systemic lupus erythematosus: prospective data from the leiden npsle cohort . Rheumatol (Oxford) . ( 2017 ) 56 : 1676–83 . doi: 10.1093/rheumatology/kex019 . PMID: 28339952

31 Shimizu F Nakamori M . Blood-brain barrier disruption in neuroimmunological disease . Int J Mol Sci . ( 2024 ) 25 : 10625 . doi: 10.3390/ijms251910625 . PMID: 39408955 PMC11476930

32 Kamintsky L Beyea SD Fisk JD Hashmi JA Omisade A Calkin C . Blood-brain barrier leakage in systemic lupus erythematosus is associated with gray matter loss and cognitive impairment . Ann Rheum Dis . ( 2020 ) 79 : 1580–7 . doi: 10.1136/annrheumdis-2020-218004 . PMID: 33004325

33 Mader S Brimberg L . Aquaporin-4 water channel in the brain and its implication for health and disease . Cells . ( 2019 ) 8 : 90 . doi: 10.3390/cells8020090 . PMID: 30691235 PMC6406241

34 Linge P Fortin PR Lood C Bengtsson AA Boilard E . The non-haemostatic role of platelets in systemic lupus erythematosus . Nat Rev Rheumatol . ( 2018 ) 14 : 195 – 213 . doi: 10.1038/nrrheum.2018.38 . PMID: 29559714

35 Aw E Zhang Y Carroll M . Microglial responses to peripheral type 1 interferon . J Neuroinflamm . ( 2020 ) 17 : 340 . doi: 10.1186/s12974-020-02003-z . PMID: 33183319 PMC7659169

36 Lood C Blanco LP Purmalek MM Carmona-Rivera C De Ravin SS Smith CK . Neutrophil extracellular traps enriched in oxidized mitochondrial dna are interferogenic and contribute to lupus-like disease . Nat Med . ( 2016 ) 22 : 146–53 . doi: 10.1038/nm.4027 . PMID: 26779811 PMC4742415

37 Suurmond J Diamond B . Autoantibodies in systemic autoimmune diseases: specificity and pathogenicity . J Clin Invest . ( 2015 ) 125 : 2194–202 . doi: 10.1172/jci78084 . PMID: 25938780 PMC4497746

38 Ramos PC Mendez MJ Ames PR Khamashta MA Hughes GR . Pulse cyclophosphamide in the treatment of neuropsychiatric systemic lupus erythematosus . Clin Exp Rheumatol . ( 1996 ) 14 : 295–9 . 8809444

39 Mo S Li Y He J Lin L . Progress of rituximab in the treatment of systemic lupus erythematosus and lupus nephritis . Front Med (Lausanne) . ( 2024 ) 11 : 1472019 . doi: 10.3389/fmed.2024.1472019 . PMID: 39430591 PMC11486751

40 Furie R Rovin BH Houssiau F Malvar A Teng YKO Contreras G . Two-year, randomized, controlled trial of belimumab in lupus nephritis . N Engl J Med . ( 2020 ) 383 : 1117–28 . doi: 10.1056/NEJMoa2001180 . PMID: 32937045

41 Atisha-Fregoso Y Malkiel S Harris KM Byron M Ding L Kanaparthi S . Phase ii randomized trial of rituximab plus cyclophosphamide followed by belimumab for the treatment of lupus nephritis . Arthritis Rheumatol . ( 2021 ) 73 : 121–31 . doi: 10.1002/art.41466 . PMID: 32755035 PMC7839443

42 Morand EF Furie R Tanaka Y Bruce IN Askanase AD Richez C . Trial of anifrolumab in active systemic lupus erythematosus . N Engl J Med . ( 2020 ) 382 : 211–21 . doi: 10.1056/NEJMoa1912196 . PMID: 31851795

43 Gupta R Agapoff M Diamond B . Neuropsychiatric systemic lupus erythematosus - current and novel treatments . Curr Opin Rheumatol . ( 2026 ) 38 : 115–20 . doi: 10.1097/bor.0000000000001141 . PMID: 41251208 PMC12863625

44 Hanly JG Urowitz MB Su L Bae SC Gordon C Wallace DJ . Prospective analysis of neuropsychiatric events in an international disease inception cohort of patients with systemic lupus erythematosus . Ann Rheum Dis . ( 2010 ) 69 : 529–35 . doi: 10.1136/ard.2008.106351 . PMID: 19359262 PMC2929162

45 Ohl K Tenbrock K Kipp M . Oxidative stress in multiple sclerosis: central and peripheral mode of action . Exp Neurol . ( 2016 ) 277 : 58 – 67 . doi: 10.1016/j.expneurol.2015.11.010 . PMID: 26626971 PMC7094520