A bibliometric analysis of lung cancer and atrial fibrillation [1991–2026]: research hotspots, thematic evolution, and translational landscape in cardio-oncology

Abstract

Background

The intersection of lung cancer (LC) and atrial fibrillation (AF) is common in thoracic surgery, systemic cancer therapies, and cardio-oncology, yet the research landscape remains inadequately characterized. To understand the progression of LC-AF research, a detailed overview is necessary to identify its main research focuses and highlight the most pertinent knowledge gaps for upcoming clinical and translational research. The present study aimed to systematically clarify the global research landscape, translational priorities, and key molecular signatures relevant to contemporary clinical and scientific focus in LC-AF comorbidity via a multi-perspective approach.

Methods

We conducted a bibliometric study on LC-AF literature sourced from the SCI-Expanded database in the Web of Science Core Collection (WoSCC), updated to April 2026, including original articles and reviews after a predefined screening process. Clinical trial records indexed in PubMed were analyzed as an additional module to explore trends in trial-oriented clinical trends. Additionally, a transcriptomic analysis that integrates AF and LC datasets was carried out to give molecular context to repeated bibliometric themes.

Results

The LC-AF field experienced steady publication growth, which became more pronounced post-2015. The primary contributors were the United States and China, with productive institutions clustered in several renowned global academic centers. Research hotspots were mainly centered on perioperative AF, thoracic surgery, cardiovascular complications, and risk-factor assessment, with thematic evolution toward cardio-oncology, treatment-related toxicity, immunotherapy, and precision-oriented research. Supplementary analysis of trials pointed to a need for greater emphasis on precision therapy, age-specific management, and more rigorous study designs. The transcriptomic exploratory study pinpointed shared genes with differential expression, including S100A8, S100A9, and IGFBP2, and underscored inflammatory and metabolic pathways as potential molecular connections to frequent bibliometric topics.

Conclusions

The field of LC-AF research is expanding from a focus on perioperative and descriptive aspects to include a wider cardio-oncology orientation. Future work should prioritize long-term follow-up in comorbid populations, clearer distinction between perioperative and chronic AF scenarios, prospective multicenter validation of risk models, and translational studies linking molecular signals to perioperative surveillance and treatment decision-making.

Highlight box.

Key findings

• Lung cancer and atrial fibrillation (LC-AF) research has grown steadily and has shifted from a predominantly perioperative and descriptive literature toward broader themes involving cardio-oncology, treatment-related toxicity, and precision-oriented research.

• Supplementary trial-oriented mapping suggests increasing clinical attention to precision therapy, age-stratified management, and more rigorous study designs.

• Exploratory transcriptomic analysis identified shared molecular signals, including S100A8, S100A9, and IGFBP2, and highlighted inflammatory and metabolic pathways as potential mechanistic correlates.

What is known and what is new?

• AF is a common cardiovascular complication in patients with LC, particularly in the perioperative setting, and treatment-related cardiovascular events are increasingly recognized in cardio-oncology.

• This study provides an integrated overview of the LC-AF field by combining bibliometric mapping with supplementary trial-oriented clinical analysis and exploratory transcriptomic context.

What is the implication, and what should change now?

• Future research should place greater emphasis on long-term follow-up in comorbid LC-AF populations, clearer differentiation between perioperative and chronic AF scenarios, prospective multicenter validation of clinical and molecular risk models, and translational studies linking biomarker signals to perioperative surveillance and treatment decision-making.

Introduction

Lung cancer (LC) continues to be the leading cause of cancer deaths around the world, significantly burdening both patients and healthcare systems (1). Simultaneously, atrial fibrillation (AF), the most frequent ongoing cardiac arrhythmia, is connected with stroke, heart failure, longer hospitalizations, and increased death rates (2). In the past few years, there has been heightened interest in the clinical connection between LC and AF, notably in the context of thoracic surgery, systemic anticancer therapy, and cardio-oncology (3-6). Present evidence implies a two-way relationship between LC and AF:LC drives AF risk through inflammation ,treatment-related cardiotoxicity, perioperative stress (5-14), while AF serves as a potential warning sign for LC (15-17). Nonetheless, most existing studies have concentrated on the role of LC and its treatment in contributing to AF, while the overall framework and development of this research area have not been thoroughly analyzed.

The clinical importance of this topic is particularly evident in the perioperative setting. Perioperative atrial fibrillation (POAF) is often seen as a cardiovascular complication post-LC surgery, leading to higher postoperative morbidity, extended hospital stays, and adverse long-term outcomes (18,19). Beyond surgery, the expansion of targeted therapy, immunotherapy, and multimodal treatment has further complicated the cardiovascular profile of patients with LC, bringing treatment-related arrhythmia, thromboembolic risk, and cardio-oncology management into sharper focus (20,21). In contrast to well-studied cardio-oncology areas such as chemotherapy-related heart damage, heart failure, or myocarditis, the intersection of LC and AF has been less thoroughly investigated. Existing studies have mainly addressed specific clinical issues, such as postoperative complications, treatment-related cardiovascular events, prognostic evaluation, and risk-factor assessment, rather than providing a structured overview of the developmental trajectory, collaboration patterns, thematic evolution, and knowledge structure of the field.

In this context, bibliometric analysis proves particularly advantageous as it quantitatively delineates aspects of a research domain that are not easily captured through traditional narrative reviews. These aspects include publication growth, collaboration patterns, co-citation relationships, thematic evolution, and the emergence of research fronts over time. To bridge this gap, we conducted an integrated analysis that centered on bibliometric mapping, supplemented by clinical trial-oriented mapping and exploratory transcriptomic analysis. Initially, we employed bibliometric techniques to delineate the global research landscape of the LC-AF field, encompassing publication trends, key contributors, collaborative networks, and thematic evolution. To augment this overview, we further scrutinized PubMed-indexed clinical trial literature to determine whether the themes identified in the bibliometric analysis were also reflected in trial-oriented research. Additionally, we investigated shared transcriptomic signals between AF and LC to elucidate the molecular foundations of recurrent themes identified in the literature.

Taken together, this multi-perspective approach was designed to provide a structured overview of the LC-AF research landscape, its translational priorities, and selected molecular signals relevant to current clinical and academic interest in this field. Consequently, the objective of this study was to elucidate the evolution of LC-AF research, identify significant knowledge gaps within the field, and underscore potential directions that hold substantial relevance for future investigations oriented towards cardio-oncology. We present this article in accordance with the BIBLIO reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0237/rc).

Methods

Study design

In this study, bibliometric analysis served as the principal framework, augmented by targeted clinical-trial mapping and exploratory transcriptomic analysis to furnish further translational and mechanistic insights. The primary objective was to delineate the global research landscape of LC-AF, while the supplementary analyses aimed to contextualize the extent to which bibliometric themes were mirrored in trial-oriented clinical research and specific molecular signals.

Data sources and search strategy

Two complementary databases were utilized in this study. The SCI-Expanded within Web of Science Core Collection (WoSCC) was employed as the primary source for bibliometric analysis due to its standardized citation indexing system, which is well-suited for co-citation analysis, collaboration studies, and knowledge-structure mapping. Additionally, PubMed was incorporated because of its extensive coverage of biomedical and clinically indexed literature, with a particular emphasis on trial-related records and MeSH-based thematic categorization, thereby offering a complementary clinical perspective. To ensure our findings are reproducible, we’ve included the raw bibliometric data as follows: data from the WoSCC database (available online: https://cdn.amegroups.cn/static/public/jtd-2026-1-0237-1.txt), and data from the PubMed database (available online: https://cdn.amegroups.cn/static/public/jtd-2026-1-0237-2.txt).

The WoSCC search was updated until April 17, 2026, and the PubMed search until April 28, 2026. WoSCC was used to explore literature on LC and AF, while PubMed focused on clinical-trial records in this area. The study emphasized depth and consistency by selecting WoSCC for comprehensive bibliometric mapping and PubMed for clinical validation. The methodological design of this study strategically prioritized depth and consistency over breadth. In this study, we specifically selected WoSCC as the sole representative citation-indexed source for network-based bibliometric mapping. This choice was based on its authoritative and comprehensive coverage, which is considered sufficiently representative to support a robust analysis of the knowledge structure within the field. Additionally, we utilized PubMed, a clinically oriented database, for thematic and translational validation. Consequently, Scopus was excluded from the study as the research design prioritized the use of one primary citation-indexed source for network-based bibliometric mapping and one clinically indexed source for thematic and translational validation, rather than duplicating efforts across multiple citation databases. Detailed search strategies are provided in Tables S1,S2.

Eligibility criteria

For the primary bibliometric analysis, eligible records were English-language articles indexed in SCI-Expanded in WoSCC, focusing on LC and AF, including original articles and reviews. Excluded were conference abstracts, editorials, letters, non-English records, and irrelevant studies. The supplementary PubMed-based analysis considered only clinical-trial-related literature directly relevant to LC-AF, serving as a mapping of clinical trends.

Data extraction

For each eligible record, document types were classified as either Article or Review for records from the WoSCC database, and additional Clinical Trial indexing information was gathered for analyses based on PubMed. Core data extracted from all records encompassed the publication year, title, journal, complete list of authors, institutional affiliations, country or region of origin, and keywords. Data specific to WoSCC included the DOI and citation counts, whereas data specific to PubMed included PMID and MeSH terms.

Bibliometric analysis

A bibliometric analysis was conducted, encompassing publication growth, collaboration patterns, co-occurrence, co-citation relationships, thematic evolution, and the emergence of research fronts over time within WoSCC. To enhance interpretability for a general clinical audience, definitions and concise explanations of the primary bibliometric indicators employed in this study are presented in Table S3.

Supplementary trial-oriented clinical mapping

A supplementary analysis utilizing PubMed-based clinical trial records was conducted to characterize trends in clinical trials within the LC-AF field, aiming to determine whether the themes identified in the bibliometric analysis were also evident in the clinically indexed trial literature.

Exploratory transcriptomic analysis

To elucidate the molecular underpinnings of recurrent themes identified through bibliometric analyses, we conducted a comprehensive integrated transcriptomic analysis of AF and LC datasets. Gene expression data for AF were sourced from the Gene Expression Omnibus (GEO) (22), while transcriptomic and clinical data for lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) were obtained from The Cancer Genome Atlas (TCGA) (23). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Software and statistical analysis

Bibliometric analyses were conducted utilizing VOSviewer (24), CiteSpace (24), and the R programming environment with the Bibliometrix package. Default parameter settings were employed unless specified otherwise. The network analyses encompassed co-occurrence, co-citation, and collaboration mapping. To delineate major research domains and temporal trends, keyword clustering and thematic evolution analyses were performed. The quality of clusters was assessed using Modularity Q and Weighted Mean Silhouette S. Comprehensive details regarding software versions, parameter settings, and definitions of bibliometric indicators are available in Table S4.

Exploratory bioinformatics analysis was executed using GEO (AF datasets GSE41177, GSE115574) and TCGA (LUAD/LUSC RNA-seq data). Differentially expressed genes (DEGs) were identified employing limma (25) and DESeq2 (25), with criteria set at an adjusted P value of less than 0.05 and |log₂FC| greater than 1. Protein-protein interaction (PPI) networks were constructed using STRING (26) and visualized in Cytoscape (27), with hub genes identified via cytoHubba (28) using the MCC (29) method. GeneMANIA (30) was employed to predict interacting genes. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using the clusterProfiler (31) package.

Results

Analysis of growth and structure of the LC-AF literature

Figure 1 provides a comprehensive overview of LC-AF research from 1991 to 2026. Following a systematic screening process, 496 publications were included (Figure 1A). The bibliometric analysis indicates that LC-AF research is a relatively focused yet expanding field, characterized by contributions from a diverse array of authors, institutions, and journals (Figure 1B). An analysis of temporal trends reveals a consistent upward trajectory in annual publication output, with a fitted quadratic regression curve (R²=0.9493) demonstrating accelerated growth over the past three decades. This trend exhibits a more pronounced acceleration post-2015, suggesting a sustained increase in scholarly interest in this domain (Figure 1C). Overall, the publication trajectory reflects a field that has experienced steady expansion in recent years, while remaining modest in scale relative to its clinical significance.

Figure 1.

Global bibliometric overview of LC-AF research. (A) PRISMA flow diagram of literature identification and selection. (B) Core bibliometric metrics of included publications. (C) Annual publication output from 1991 to 2025 with fitted quadratic regression curve. LC-AF, lung cancer and atrial fibrillation; NSCLC, non-small cell lung cancer; SCLC, small cell lung cancer.

Analysis of geographic, institutional, and author landscape

At the national level, the United States and China emerge as the two most prolific countries, contributing approximately 47% of the total research output, with Japan following (Figure 2A). It is noteworthy that the majority of countries predominantly depend on SCP, with relatively low proportions of MCP, indicating limited global integration in cross-border research. The collaboration map further illustrates this trend, showing that the most robust collaborative ties are concentrated between the United States and China, while partnerships among other nations remain sparse and fragmented (Figure 2B). These findings underscore the U.S.-China-centric nature of research in the field of LC-AF, highlighting both their leading roles and the necessity for enhanced international collaboration in this interdisciplinary domain.

Figure 2.

International collaboration landscape. (A) Top 20 most productive countries by publication output, with SCP and MCP indicated. (B) Country/region collaboration map illustrating cross-national research collaboration patterns. MCP, multiple country publications; SCP, single country publications.

At the institutional level, Harvard University and its medical affiliates are at the forefront of the field, followed by Shanghai Jiao Tong University, illustrating the predominance of leading U.S. and Chinese academic centers in advancing this interdisciplinary research (Figure 3A). The collaboration network further reveals a highly centralized structure, with the Harvard-centered cluster forming the core of the network, demonstrating the most extensive and interconnected relationships with other major institutions. In contrast, numerous other institutions remain in relatively isolated, sparsely connected clusters, indicating a fragmented global landscape of institutional collaboration (Figure 3B). Collectively, these findings emphasize the concentration of LC-AF research activity within a limited number of high-impact centers, particularly in the United States, and underscore the necessity for expanded cross-institutional partnerships to promote broader collaboration in this emerging field.

Figure 3.

Institutional publication output and collaboration. (A) Top 10 most productive institutions ranked by publication count. (B) Institutional collaboration network generated in R. Node size indicates publication volume, and connecting lines represent collaborative links.

At the author level, among the top ten most prolific authors, as shown in Table 1 and Figure 4A, seven are associated with Japanese institutions, indicating a significant concentration of research activity in this area. The author collaboration network uncovers a clustered arrangement consisting of three main groups: a large, tightly-knit Japanese cluster, a cluster led by the U.S., and various smaller, scattered clusters from Europe and Asia (Figure 4B). This pattern indicates that collaborations remain largely confined within regional and institutional boundaries, with limited cross-regional integration.

Table 1. Top 10 most influential authors publishing on LC and AF (ranked by local impact as measured by H-index).

Rank	Author	Country	H-index	G-index	TC	NP	PY_start
1	Nojiri Takashi	Japan	8	8	213	8	2010
2	Okumura Meinoshin	Japan	8	9	241	9	2010
3	Maeda Hajime	Japan	7	7	205	7	2010
4	Takeuchi Yukiyasu	Japan	7	7	205	7	2010
5	Amar David	USA	5	5	465	5	2007
6	Inoue Masayoshi	Japan	5	5	115	5	2011
7	Yamamoto Kazuhiro	Japan	5	5	161	5	2010
8	Cardinale Daniela	Italy	4	4	231	4	2007
9	D’Amico Thomas A.	USA	4	4	727	4	2007
10	Funakoshi Yasunobu	Japan	4	4	149	4	2010

AF, atrial fibrillation; LC, lung cancer; NP, number of publications; PY_start, publishing year start; TC, total citations.

Figure 4.

Author productivity and collaboration patterns. (A) Top 10 most productive authors, ranked by publication count. (B) Author collaboration network generated using R. Node size represents publication volume, and colors denote collaborative clusters.

Analysis of journals and co-cited journals

Figure 5 presents the interdisciplinary landscape and the flow of disciplinary knowledge within LC-AF research. The temporal distribution of publications across the top 10 journals indicates that core cardiothoracic surgery journals have consistently served as primary venues for publication (Figure 5A). The dual-map overlay visualizes disciplinary citation pathways, revealing that citing journals are predominantly concentrated within the “Medicine, Medical, Clinical” domain. Two principal trajectories are evident: clinical citing journals primarily connect to the “Health, Nursing, Medicine” domain (z=6.30, f=3,033), reflecting a strong translation of LC-AF research into clinical practice and care guidelines. Secondarily, they link to the “Molecular, Biology, Genetics” domain (z=2.48, f=1,263), illustrating how clinical observations inform mechanistic and translational investigations into cancer-related arrhythmias. These bidirectional flows underscore the field’s interdisciplinary nature: clinical research both depends on foundational biological knowledge and stimulates new molecular inquiries, highlighting LC-AF research as a vital bridge between clinical practice and bench science. The co-citation network delineates three distinct disciplinary clusters: one centered on cardiothoracic surgery, a core cluster focused on cardiovascular medicine, and another led by oncology. This configuration highlights the influence of the intersection among these three pivotal disciplinary communities on the development of LC-AF research (Figure 5C).

Figure 5.

Journal distribution, co-citation structure, and knowledge flow pathways in LC-AF research. (A) Annual publication output of the top 10 most productive journals (1991–2026; 2026 data incomplete). (B) Journal dual-map overlay generated by CiteSpace. The left panel represents citing journals, the right panel represents cited journals, and curved lines indicate citation pathways between disciplines. (C) Journal co-citation network. Nodes represent journals, links indicate co-citation relationships, and colors denote disciplinary clusters. LC-AF, lung cancer and atrial fibrillation.

Analysis of research hotspots and thematic evolution

Table 2 delineates the top 10 locally cited documents within the realm of LC-AF research, all of which were published between 1999 and 2014, focusing on POAF following thoracic surgery for LC. These influential studies illustrate a distinct progression over this period: initial foundational research identified risk factors and established epidemiological profiles; subsequent prospective analyses corroborated the adverse impact of POAF on long-term survival; and later studies advanced the field through biomarker-based prediction, assessment of surgical techniques, and the development of evidence-based clinical guidelines and onco-cardiology frameworks. These works were predominantly published in leading journals of cardiothoracic surgery and cardiovascular medicine.

Table 2. Top 10 publications based on local citations in the field of LC and AF.

Rank	Title	Source	Year	Local citations	Global citations
1	Risk factors associated with atrial fibrillation after noncardiac thoracic surgery: analysis of 2588 patients	The Journal of Thoracic and Cardiovascular Surgery	2004	65	269
2	Risk Factors for Atrial Fibrillation After Lung Cancer Surgery: Analysis of The Society of Thoracic Surgeons General Thoracic Surgery Database	The Annals of Thoracic Surgery	2010	62	184
3	Atrial fibrillation complicating lung cancer resection	The Journal of Thoracic and Cardiovascular Surgery	2005	55	155
4	Atrial fibrillation after operation for lung cancer: clinical and prognostic significance	The Annals of Thoracic Surgery	1999	36	76
5	Increased Perioperative N-Terminal Pro-B-Type Natriuretic Peptide Levels Predict Atrial Fibrillation After Thoracic Surgery for Lung Cancer	Circulation	2007	29	86
6	Atrial fibrillation after pulmonary lobectomy for lung cancer affects long-term survival in a prospective single-center study	Journal of Cardiothoracic Surgery	2012	29	102
7	Video-assisted thoracic surgery does not reduce the incidence of postoperative atrial fibrillation after pulmonary lobectomy	The Journal of Thoracic and Cardiovascular Surgery	2007	25	105
8	Predictive value of B-type natriuretic peptide for postoperative atrial fibrillation following pulmonary resection for lung cancer	European Journal of Cardio-Thoracic Surgery	2010	23	49
9	Insights Into Onco-Cardiology: Atrial Fibrillation in Cancer	Journal of the American College of Cardiology	2014	21	319
10	2014 AATS guidelines for the prevention and management of perioperative atrial fibrillation and flutter for thoracic surgical procedures	The Journal of Thoracic and Cardiovascular Surgery	2014	21	236

AATS, American Association for Thoracic Surgery; AF, atrial fibrillation; LC, lung cancer.

Keyword co-occurrence clustering analysis (Figure 6A) identified four main research themes in the LC-AF field, with a high silhouette value of 0.8537 confirming the reliability of the results. Themes include: #0 thoracic surgery and #1 atrial fibrillation as the core clinical framework, #3 direct oral anticoagulants for thromboembolism prevention, #11 cerebral infarction and #7 cardiac tamponade as key perioperative complications, and #6 liposomal doxorubicin related to chemotherapy-induced cardiotoxicity. Figure 6B’s thematic map further highlights the positioning of research topics. The upper-left niche themes, such as video-assisted thoracoscopic surgery (VATS) and crizotinib-targeted therapy, are mature but have low centrality, making them peripheral to mainstream research. In contrast, the lower-right motor themes, focusing on POAF, cancer-related cardiac injury, stroke prevention and thoracic surgery management, are central and have significantly influenced research progress.

Figure 6.

Author keyword clustering analysis and conceptual structure. (A) Keyword clustering network based on co-occurrence analysis. Network parameters (N=205, E=850) are shown with modularity Q=0.604>0.3 and weighted mean silhouette S=0.8537>0.5 indicating cluster quality. (B) Thematic map derived from author keywords. The horizontal axis (centrality) reflects the importance of themes within the overall research field, and the vertical axis (density) suggests the internal development of themes. (C) Factorial map based on author keywords, illustrating major thematic divisions within the field. MCA, multiple correspondence analysis.

The conceptual structure map derived from Multiple Correspondence Analysis (MCA) effectively illustrates the evolutionary trajectory of this research domain. The cluster on the left, depicted in red, establishes a traditional foundation for research, with a focus on thromboembolism, anticoagulation strategies, bleeding risk, and stroke management. In contrast, the cluster on the right, shown in blue, signifies the current predominant research focus, which has shifted towards cardio-oncology, particularly emphasizing immunotherapy-related cardiotoxicity and the application of amiodarone. The green cluster located in the upper-right corner signifies an emerging frontier in the field. Research on cardiac injury induced by targeted therapies, specifically those related to tyrosine kinase inhibitors and osimertinib, is increasingly becoming a focal point, indicating a progression towards precision cardio-oncology (Figure 6C).

The evolutionary trajectory of LC-AF research is thoroughly illustrated in Figure 7, which delineates a distinct three-stage progression. The initial phase of research, prior to 2015, predominantly concentrated on conventional thoracic surgery and fundamental perioperative outcomes, with a focus on terms such as thoracotomy, pneumonectomy, and pulmonary resection. Transitioning into the mid-stage (post-2015), the research emphasis shifted towards minimally invasive techniques and the management of postoperative complications, marked by an increased focus on VATS, lobectomy, and POAF. In the most recent stage, the field has advanced into a new era characterized by cardio-oncology-oriented precision research. As depicted across all three panels, emerging themes such as cardio-oncology, cardiotoxicity, immunotherapy, and non-small cell lung cancer have become predominant. Collectively, the three maps consistently illustrate the evolution of LC AF research from an exclusively surgical focus to a multidimensional cardio-oncology discipline, with a growing emphasis on treatment-related cardiac toxicity and personalized patient management.

Figure 7.

Temporal evolution of research themes. (A) Trend Topics timeline. The horizontal axis represents time, and topic frequency is reflected by the size and continuity of the nodes and connecting lines, illustrating the temporal dynamics of research topics. (B) Keyword co-occurrence network generated by VOSviewer. Node size reflects keyword frequency; the color gradient (blue to yellow) indicates the chronological trend from earlier to more recent periods; lines represent co-occurrence relationships between keywords. (C) Keyword burst analysis.

Supplementary translational and molecular context

Supplementary mapping of trial-oriented clinical trends

Research Hotspots and Thematic Evolution analysis showed that research in this field has shifted from perioperative and descriptive studies to focus on cardio-oncology, drug toxicity, and translational research. Accordingly, we examined if this trend is also evident in clinical trial literature by conducting a targeted analysis of trials from PubMed.

As illustrated in Figure 8, the supplementary analysis centered on recent trials revealed a growing emphasis in trial-based research on precision therapy, particularly in the areas of immune checkpoint inhibitors and biomarkers, as well as age-stratified management for populations aged 80 years and over, including both male and female subjects. This shift is accompanied by the adoption of more rigorous study designs, such as prospective studies, follow-up studies, and double-blind methodologies. The recurrent appearance of high-frequency terms associated with molecular targeted therapy, research on elderly populations, and survival outcomes suggests that clinical research is increasingly characterized by individualized and translational approaches. These findings align closely with the broader bibliometric trends observed in this field.

Figure 8.

Clinical research progress of LC-AF comorbidity based on PubMed clinical trials. (A) Keyword co-occurrence network: node size corresponds to keyword frequency. (B) Research hotspot evolution heatmap: color depth indicates the relative density of keywords across different time periods. LC-AF, lung cancer and atrial fibrillation.

Exploratory shared transcriptomic signals

In light of the bibliometric analysis highlighting themes associated with inflammation and risk, we conducted a further investigation to determine whether AF and LC might share transcriptomic signals pertinent to these recurring clinical patterns.

To provide an exploratory molecular context for these themes, we performed an integrated transcriptomic analysis of datasets pertaining to AF and LC. This analysis revealed shared DEGs, among which several hub genes, including S100A8, S100A9, and IGFBP2, emerged as potentially significant candidates (Figure 9A-9C). Functional enrichment analyses indicated the involvement of inflammatory and metabolic pathways, encompassing signals related to chronic inflammatory responses and carbon metabolism (Figure 9D,9E). While these findings do not establish causation, they offer an exploratory molecular context for the recurrent inflammatory and risk-related patterns identified in the bibliometric analyses.

Figure 9.

Supplementary bioinformatics analysis of shared molecular pathways between AF and LC. (A,B) Venn diagrams of shared (A) upregulated and (B) downregulated DEGs. (C) PPI network of the 65 shared DEGs. The top 20 hub genes are highlighted. (D) GeneMANIA functional interaction network of the hub genes, showing predicted interacting partners. (E) Functional enrichment analysis (GO and KEGG) of the top 20 hub genes, highlighting key inflammatory and metabolic pathways. AF, atrial fibrillation; BP, biological process; CC, cellular component; DEGs, differentially expressed genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; LC, lung cancer; MF, molecular function; PPI, protein-protein interaction.

Discussion

This study offers a comprehensive synthesis of the LC-AF domain by integrating bibliometric mapping with supplementary clinical trial analysis and exploratory transcriptomic evaluation. Collectively, the findings indicate that LC-AF research has transitioned from a predominantly narrow focus on postoperative atrial fibrillation and surgical complications to a more expansive cardio-oncology-oriented framework. This new framework encompasses treatment-related cardiovascular toxicity, risk stratification, and specific mechanistic themes (14).

The analysis of annual publications identified 2015 as a pivotal turning point, marked by a consistent and significant increase in publications related to LC-AF. This surge was propelled by the convergence of several advancements: the widespread adoption of minimally invasive thoracic surgery, increased clinical focus on perioperative complications, rapid advancements in targeted therapy and immunotherapy for LC, and the gradual establishment of cardio-oncology as a recognized interdisciplinary field (14,16,21,29,32-36). Notably, the highly cited 2014 review, “Insights Into Onco-Cardiology: Atrial Fibrillation in Cancer“, laid a robust theoretical foundation for this domain and facilitated subsequent academic interest and research on the LC-AF intersection (37). Collectively, these developments enhanced the clinical relevance and academic impact of LC-AF research. This transition was not merely quantitative but also represented a fundamental shift in the research paradigm. Research priorities have evolved to include the exploration of standardized identification strategies, individualized management, and prognostic risk stratification within diverse anti-tumor treatment contexts.

The analysis of geographic distribution indicated that the United States and China are the foremost contributors in this domain, with institutions of high productivity primarily located in globally recognized academic centers. Notably, an independent and significant regional research community emerged in Japan, characterized by a closely interconnected cluster of authors. Although the collaboration network exhibited a well-established global framework, research activity and academic influence remained relatively concentrated. In particular, China’s substantial publication output, coupled with relatively limited international collaboration, suggests a research landscape predominantly driven by large-scale domestic case series, single-center surgical cohorts, and institution-based retrospective studies. While this structure facilitates considerable research output, it may also impede participation in broader multinational collaborative networks, especially in a field that continues to lack standardized prospective frameworks across regions. This pattern suggests that future advances would benefit from more extensive multicenter, cross-regional, and interdisciplinary collaboration, especially among thoracic surgery, cardiology, oncology, and translational research teams.

The analysis of research hotspots and thematic evolution indicates that current research on AF in LC predominantly focuses on clinical outcomes, perioperative management, risk factors, and cardiovascular complications. This suggests that the field is primarily driven by the imperative to identify high-risk populations, evaluate adverse outcomes, and optimize management strategies in vulnerable clinical settings. Notably, temporal analyses reveal a gradual transition from descriptive perioperative and epidemiological studies to investigations of treatment-related complications, toxicity, cardio-oncology, targeted therapies, and specific mechanistic topics. This shift has significant clinical implications, as AF in LC patients is increasingly recognized not only as a perioperative complication or pre-existing comorbidity but also as an indicator of disease- or treatment-related cardiovascular vulnerability. Consequently, future research should prioritize cardiovascular surveillance, risk stratification, and multidisciplinary management within the cardio-oncology framework.

An analysis of journals and co-cited journals elucidated the publication structure, citation trends, and disciplinary knowledge transfer within LC-AF research. The dual-map overlay analysis revealed that citing journals were predominantly concentrated in the domains of Medicine, Medical, and Clinical, with two distinct knowledge pathways leading to cited disciplinary fields. The primary trajectory was directed towards Health, Nursing, and Medicine, indicating the integration of clinical evidence into routine practice and risk management. The secondary trajectory was linked to Molecular, Biology, and Genetics, highlighting the continuous incorporation of fundamental biological findings into clinical research to support mechanistic interpretation and translational exploration of cancer-related arrhythmias. The journal co-citation network identified three disciplinary clusters encompassing cardiothoracic surgery, cardiovascular medicine, and oncology, further affirming that LC-AF constitutes an interdisciplinary research domain shaped by the convergence of these core disciplines.

The supplementary trial-oriented mapping facilitates the contextualization of these findings within a translational framework. The heightened emphasis on precision therapy, age-related terminology, and robust study designs within the trial literature indicates a shift in the field from retrospective descriptions to more clinically structured research inquiries. This shift is significant, as bibliometric growth alone does not ensure clinical translation. By demonstrating that certain emerging bibliometric themes are also evident in trial-oriented research, the supplementary PubMed analysis reinforces the perspective that LC-AF research is progressing towards more individualized and methodologically advanced clinical investigations.

While inflammation is frequently implicated in LC-AF research, its precise role remains a gap in knowledge rather than a fully elucidated mechanism. Current studies often identify inflammatory associations at a broad level but do not consistently differentiate between perioperative inflammatory stress and chronic cancer-related inflammation. Furthermore, they fail to clearly delineate which pathways are shared, treatment-specific, or clinically actionable. This issue is particularly pertinent in the context of immunotherapy and precision treatment, where cardiotoxicity may vary across treatment classes and mechanistic pathways remain incompletely characterized. Therefore, inflammatory processes should be regarded as a recurrent and plausible mechanistic thread rather than a fully established unifying explanation for the entire field. Correspondingly, the transcriptomic analysis, while exploratory in nature, provides valuable context for this transition. The identification of S100A8/S100A9 and IGFBP2 is significant not merely because these genes have emerged as central nodes, but because they may serve as a translational link between inflammatory risk profiles and clinically observable cardiovascular events. Should these molecular signatures be validated in future research, they could facilitate the identification of patients with an elevated inflammatory and arrhythmogenic profile, enhance perioperative risk stratification, and potentially augment existing clinical variables such as age, surgical approach, treatment exposure, and baseline cardiac status. Consequently, these molecular findings could ultimately contribute to more personalized cardiovascular monitoring in patients with LC.

Several significant knowledge gaps persist in the existing literature on LC-AF. Firstly, there is a paucity of long-term follow-up data concerning comorbid LC-AF populations, particularly in relation to survival rates, recurrent cardiovascular events, and treatment continuity. Secondly, the current body of evidence disproportionately emphasizes the contribution of LC and its treatment to AF, while the potential impact of AF-related physiological disturbances on LC outcomes remains inadequately explored. Thirdly, perioperative AF and chronic or pre-existing AF are frequently discussed within overlapping frameworks, despite probable differences in their pathophysiology, prognosis, and management requirements. Fourthly, biomarker-guided risk models, especially those that integrate molecular candidates with clinical variables, remain largely unvalidated. Lastly, the scarcity of prospective multicenter studies and standardized randomized trial designs continues to constrain both external validity and clinical applicability.

In summary, the three analytical layers of this study collectively offer a more comprehensive understanding of the LC-AF field. The bibliometric analyses indicate a shift in the literature from perioperative and descriptive issues to broader themes such as cardiotoxicity, cardio-oncology, and precision-oriented research. The additional trial-focused analysis suggests that this shift is also evident in clinically indexed research, where precision therapy, age-adapted management, and rigorous trial design have gained prominence. Furthermore, the transcriptomic findings provide exploratory molecular context for these recurring themes by highlighting shared inflammatory and metabolic signals, thereby linking macro-level research evolution with potential mechanistic substrates. While this integrated framework does not establish causation, it elucidates why perioperative risk, treatment-related cardiovascular events, and biomarker-informed stratification have become converging priorities within LC-AF research.

This study is subject to several limitations. Firstly, the exclusive reliance on the WoSCC may result in an underrepresentation of non-English and regional research, potentially biasing analyses of geographic contributions. Secondly, the keyword co-occurrence analysis identifies only thematic associations rather than causal relationships, thereby limiting mechanistic interpretations. Thirdly, citation metrics tend to favor older publications, which may lead to an underemphasis on recent advancements, such as immunotherapy-related AF. Lastly, the transcriptomic analysis is exploratory and lacks experimental validation, which precludes drawing causal inferences regarding the molecular interactions between LC and AF.

Conclusions

In summary, research on LC-AF has shown consistent growth and is transitioning from primarily perioperative and descriptive studies to a more interdisciplinary focus within the field of cardio-oncology. Presently, key areas of interest include postoperative atrial fibrillation, cardiovascular complications related to treatment, identification of risk factors, and emerging themes centered on precision medicine. Supplementary trial-oriented mapping indicates an increasing clinical focus on age-specific management, precision therapies, and the implementation of more robust study designs. Concurrently, exploratory transcriptomic analyses reveal common inflammatory and metabolic pathways, which may provide valuable context for these advancements.

Future research should prioritize long-term follow-up studies in populations with comorbid LC and AF, as well as a clearer differentiation between perioperative and chronic AF scenarios. Additionally, there is a need for prospective multicenter validation of clinical and molecular risk models, alongside more structured translational studies that connect biomarker signals to perioperative surveillance and treatment decision-making. By fostering stronger interdisciplinary collaboration, the field may advance towards more integrated strategies aimed at optimizing both oncologic and cardiovascular outcomes in patients affected by LC and AF.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.