Rheumatoid arthritis is a systemic, chronic inflammatory pathology and an autoimmune disorder1. It is characterized by progressive and destructive arthropathy stemming from synovial inflammation and hyperplasia, along with an elevation in autoantibodies such as rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP)2. In addition to musculoskeletal involvement, rheumatoid arthritis is associated with extra-articular manifestations, particularly cardiovascular disease, which contributes to diminished life expectancy among affected individuals3. Rheumatoid arthritis is a complex polygenic disease influenced by the interplay of genetic predisposition and environmental triggers, both of which play instrumental roles in disease susceptibility and progression4.

Significant strides in genetic research have markedly enhanced our understanding of rheumatoid arthritis pathogenesis, identifying numerous loci and molecular pathways involved in disease susceptibility, severity, and therapeutic response5. These insights have highlighted critical immune-related cell types and mechanisms, providing the groundwork for personalized medicine approaches. Current and future challenges include leveraging genetic variants as instrumental variables to establish causal relationships with intermediate phenotypes (e.g., transcriptomic markers), developing polygenic risk scores for diagnosis and treatment stratification, and identifying novel therapeutic targets4.

While interest in rheumatoid arthritis genetics continues to intensify, bibliometric analyses of this expanding research domain remain limited. Extant studies have primarily focused on specific subtopics, such as deoxyribonucleic acid (DNA) methylation6, non-coding ribonucleic acids (RNA)7, genome-wide association studies (GWAS)8, RNA-binding proteins9, and microRNAs10. However, a comprehensive evaluation of the global research landscape concerning rheumatoid arthritis genetics is still lacking. This study aims to fill that gap by mapping publication trends, research impact, collaborative networks, leading journals, influential authors, the most frequently cited articles, and the thematic evolution within the field. By offering an in-depth bibliometric overview, this analysis provides valuable insights that can guide future research directions and inform clinical applications11,12.

A total of 3,731 documents were included in the final analysis to map publication trends, research impact, collaborative networks, leading journals, influential authors, the most cited articles, and the thematic evolution within the field.

Our results indicate a clear trend of growth in the number of publications on rheumatoid arthritis and genetics over time. The annual trends from 1953 to 2024 reveal a steady evolution, beginning with negligible publication rates until the 1980s. A pivotal turning point occurred in 1987, when a high-impact study on the shared epitope hypothesis by Gregersen et al. generated a significant citation spike, signaling foundational contributions to the field. Research activity accelerated throughout the 1990s, culminating in the most productive epoch between 2000 and 2015. This rise may have been facilitated by the concurrent launch of the Human Genome Project (HGP) in October 1990. The HGP was a multinational project involving the United States, the United Kingdom, France, Japan, Canada, Germany, and China, undertaking to sequence the entire human genome and classify human genes 29. Although citations have slightly declined since 2016, likely due to the natural lag in citation accrual, publication rates have remained consistently high, indicating sustained global interest and active investigation in this domain.

The analysis of country contributions and international collaboration highlights both established and emerging global players. While China leads in publication volume, the United States excels in citation impact and research influence, as reflected in its top average citations per publication and h-index. European countries like Sweden and the Netherlands also show high citation efficiency, producing fewer but highly impactful papers. VOSviewer’s co-authorship network visualization indicates that these influential nations are central to global collaboration clusters, with the United States, United Kingdom, Netherlands, Sweden, and Spain forming key hubs. In contrast, countries with lower publication output, such as Thailand and Lithuania, display limited international connectivity. However, emerging contributors, particularly China, Iran, Pakistan, and several Middle Eastern nations, are becoming more active, indicating expanding global participation and growing research potential despite currently modest citation impact. Notably, China’s high publication output is largely driven by large-scale initiatives like the China Kadoorie Biobank (CKB), GWAS consortia, and other cohort studies.

The leading journals, authors, and most-cited articles reveal a tightly interlinked landscape of scholarly influence, scientific advancement, and interdisciplinary growth. High-impact journals like Arthritis and Rheumatism and Annals of the Rheumatic Diseases serve as key publication venues for seminal works, such as Gregersen et al. (1987)19, MacGregor et al. (2000)24, and Padyukov et al. (2004)27, highlighting their role in disseminating groundbreaking genetic discoveries. The prominence of generalist outlets like PLOS One and immunology-focused platforms like Frontiers in Immunology signals the increasing cross-pollination between rheumatology, genomics, and broader biomedical research. This aligns with the evolving scope of top-cited papers, which span from early heritability studies to cutting-edge GWAS, epigenetics, and gene–environment interactions. Importantly, this pattern is not only of academic interest but also highlights the translational trajectory of rheumatoid arthritis genetics research. For instance, GWAS that identified variants in the Janus kinase (JAK)- signal transducer and activator of transcription (STAT) or JAK–STAT signaling pathway provided the scientific foundation for the development of JAK inhibitors such as tofacitinib and baricitinib, which are now approved for clinical use in rheumatoid arthritis5,20,30.

A strong correlation also emerges between the productivity of influential authors and the top-cited articles. Researchers such as Martín (Spain), Gregersen (United States), and Klareskog (Sweden), all of whom appear in the top 10 author rankings, are also behind some of the most cited articles, reinforcing their leadership in shaping the field. The keyword co-occurrence analysis using VOSviewer reveals a rich and evolving thematic structure in rheumatoid arthritis and genetics research, segmented into four distinct yet interconnected clusters. Cluster 1 (red) emerges as the largest and most recent, showcasing the field’s shift toward molecular precision and personalized therapy31,32,33. Core concepts such as rheumatoid arthritis, methotrexate, gene expression, and cytokines anchor this cluster in the disease's biological basis and treatment landscape. However, the appearance of newer keywords like machine learning, RNA-seq, and bioinformatics signals a growing reliance on high-throughput and computational methodologies to dissect complex genetic patterns34,35,36. The inclusion of Mendelian randomization, COVID-19, and interstitial lung disease further reflects an expanding research scope that now incorporates global health concerns and comorbidity profiling37,38,39.

The second largest cluster (green) emphasizes polymorphisms, SNPs, and meta-analysis, as well as comparisons with related autoimmune diseases like SLE40,41. The persistent use of older keywords such as juvenile rheumatoid arthritis underscores how foundational studies in genetic susceptibility continue to inform current hypotheses. These insights highlight opportunities for future studies to explore shared genetic mechanisms across autoimmune conditions and to refine genotype-based risk stratification models42,43. Clinically, such advancements could support the development of more targeted diagnostic tools and personalized therapeutic strategies in rheumatoid arthritis and related disorders44. In contrast, Cluster 3 (blue) focuses on well-established immunogenetic hallmarks that continue to shape clinical practice. Keywords like HLA, shared epitope, and PTPN22 are deeply embedded in rheumatoid arthritis's pathogenic narrative45,46, while the presence of anti-CCP, cardiovascular disease, and disease activity highlights a maturing focus on translational outcomes47,48. Lastly, Cluster 4 (yellow) bridges basic and translational science through exploration of functional gene variants and genotype–phenotype relationships. Keywords like VDR, etanercept, and TNF-α underscore the genetic underpinnings of therapeutic response49,50,51, with vitamin D and gene polymorphisms suggesting newer angles on immune modulation52,53. The coexistence of older methods such as RFLP and immunogenetics with newer molecular targets illustrates the ongoing synthesis of traditional genetic tools with modern approaches54,55.

Beyond bibliometric trends, a closer look at the four keyword clusters reveals both consensus and ongoing controversies in rheumatoid arthritis genetics. In Cluster 1, molecular and transcriptomic studies consistently implicate cytokine dysregulation, particularly IL6, TNF-α, and JAK–STAT pathway components, as core mediators of inflammation and drug response5,31,33. These findings underpin the rationale for targeted therapies such as JAK inhibitors, yet cross-cohort transcriptomic analyses show inconsistent predictive signatures for methotrexate or TNF inhibitor efficacy. Cluster 2 highlights strong consensus on shared autoimmune susceptibility genes such as HLA-DRB1 and PADI4, which exhibit reproducible risk effects across ancestries19,23,46. However, variants such as PTPN22 display marked population-specific heterogeneity, strongly associated with rheumatoid arthritis in European and North American cohorts but weak or absent in East Asian and African populations21,45. These inconsistencies suggest allelic heterogeneity or gene–environment interactions shaping regional disease risk. In Cluster 3, HLA-DRB1 remains the most robust universal locus, whereas other loci such as STAT4 and CTLA4 show divergent significance across populations, as demonstrated by studies in Iraqi, Mexican, and Algerian cohorts46,47,53. Such discrepancies limit the transferability of polygenic risk scores and highlight the need for ancestry-inclusive GWAS. Finally, Cluster 4 highlights the translational challenges in linking functional polymorphisms to treatment response. Variants in VDR, TNF, and TP53 have been correlated with biologic response, remission, or disease activity in some studies, yet replication remains inconsistent and clinical utility uncertain49,50,52,55.

Collectively, these patterns illustrate a maturing yet heterogeneous field, where established immunogenetic mechanisms coexist with ancestry-specific associations and persistent barriers to clinical translation. This study has several limitations that warrant consideration. First, the analysis relied exclusively on Scopus as the data source. While Scopus provides broad coverage, reliable citation metadata, and strong compatibility with bibliometric tools13,14, the use of a single database may have excluded relevant publications indexed only in Web of Science or PubMed. Variations in indexing standards, author affiliations, and institutional naming conventions across databases also pose challenges for data harmonization. Second, our search strategy required the presence of inclusion search terms in article titles, which may have led to the omission of some relevant studies where these terms were reported only in abstracts or keywords. Third, the analysis was based on raw citation counts rather than normalized measures. As a result, some influential recent publications may be underrepresented in citation-based rankings. Fourth, the overlay visualization generated using VOSviewer is based on the average publication year. While this approach is useful for indicating the relative recency of research themes, it does not capture the frequency dynamics of keyword usage over time. Accordingly, our findings should be interpreted in light of these constraints.

Bibliometric analysis of rheumatoid arthritis and genetics research reveals a field marked by steady growth and evolving scientific priorities. From foundational breakthroughs in the late 20th century to the recent surge in interest in pharmacogenetics, epigenetics, and data-driven methodologies, the field has matured into a highly interdisciplinary domain. Thematic keyword clusters further reflect the field’s transition from traditional immunogenetics toward personalized medicine, complex comorbidity research, and computational genomics, signaling a future direction centered on precision healthcare and integrative approaches to disease understanding and management.

ACPA: anti-citrullinated protein antibody, anti-CCP: anti-cyclic citrullinated peptide, CKB: China Kadoorie Biobank, COVID-19: coronavirus disease 2019, CTLA4: cytotoxic T-lymphocyte-associated protein 4, DNA: deoxyribonucleic acid, GWAS: genome-wide association studies, HGP: Human Genome Project, HLA: Human Leukocyte Antigen, JAK-STAT: Janus kinase - signal transducer and activator of transcription, PTPN22: protein tyrosine phosphatase non-receptor type 22, QTLs: epigenomic quantitative trait loci, RFLP: restriction fragment length polymorphism, RNA: ribonucleic acid, SLE: systemic lupus erythematosus, SNP: single-nucleotide polymorphism, TNF: tumor necrosis factor, VDR: vitamin D receptor.

None.

(I) Conceptualization, methodology development, data curation, formal analysis, and drafting of the original manuscript: WNAZ (II) Administrative support, critical review, and manuscript editing: AW, WSWG, NDMS (III) Data acquisition, validation, and analysis: WNAZ, AW (IV) Final approval of the manuscript and accountability for the integrity of the work: All authors.

None.

Supplementary files, including the bibliometric dataset (.csv) and reference file (.ris), are available in Zenodo at https://doi.org/10.5281/zenodo.17096085

Not applicable.

Not applicable.

The authors declare that they have not used generative AI (a type of artificial intelligence technology that can produce various types of content including text, imagery, audio and synthetic data).

The authors declare that they have no competing interests.