A bibliometric and visualization analysis of radiotherapy for hepatocellular carcinoma with portal vein tumor thrombosis from 2013 to 2025

Abstract

Background:

The management of hepatocellular carcinoma (HCC) combined with portal vein tumor thrombosis (PVTT) poses considerable clinical challenges, and radiotherapy (RT) has emerged as a vital therapeutic option. Despite its importance, comprehensive analyses in this domain are lacking. Therefore, this study aims to conduct a bibliometric analysis to elucidate the knowledge structure and evolution of RT for PVTT in patients with HCC from 2013 to 2025.

Methods:

Publications were retrieved from PubMed, yielding 34 initial studies, of which only 25 met the inclusion criteria for final analysis. A bibliometric analysis was performed using the R programming language and the bibliometrix package.

Results:

The annual publication output exhibited notable growth, rising from 1 to 3 publications per year during 2013–2018 to a peak of five publications in 2023. Institutional analysis revealed that The Catholic University of Korea was the most prolific contributor, with its output increasing from 2 to 7 publications annually. Co-citation network analysis identified Llovet JM (2008) as the seminal reference, with a betweenness centrality of 558.24. The thematic evolution of the field has progressed from the establishment of fundamental radiotherapeutic techniques to the incorporation of radiomics and targeted therapies.

Conclusion:

This bibliometric analysis provides a systematic, quantitative synthesis of the research landscape regarding RT for PVTT in patients with HCC. It emphasizes the importance of establishing staging systems and indicates a notable transition toward image-guided RT and combination-based precision medicine. These findings offer a strategic framework for clinicians to effectively navigate the future of this critical therapeutic area.

Introduction

Globally, hepatocellular carcinoma (HCC) poses a considerable oncological challenge, with recent epidemiological studies indicating an escalating incidence, particularly in Western countries^[1,2^]. This upsurge is markedly associated with cases related to the hepatitis B virus (HBV) and non-alcoholic steatohepatitis^[3,4^]. The clinical management of HCC is confounded by portal vein tumor thrombosis (PVTT), a condition currently recognized not only as a marker of disease progression but also as an active driver of tumor dissemination^[5,6^]. This dissemination occurs via intricate molecular interactions involving vascular endothelial growth factor-mediated angiogenesis, angiopoietin-2 signaling, and epithelial–mesenchymal transition pathways^[7^].

HIGHLIGHTS

• The leadership of East Asian researchers in this field is evident.

• The vital role of established staging systems.

• A is a notable transition toward image-guided radiotherapy and combination-based precision medicine.

In the preceding 5 years, the therapeutic approach to PVTT in patients with HCC has undergone significant redefinition^[8^]. The era of tyrosine kinase inhibitors, initiated by the landmark Sorafenib Hepatocellular Carcinoma Assessment Randomized Protocol trial, established a foundation for systemic treatment strategies^[9^]. However, the emergence of immunotherapy has profoundly transformed treatment protocols. The IMbrave150 trial demonstrated that the atezolizumab–bevacizumab combination outperforms sorafenib, establishing a novel standard of care and prompting a critical re-evaluation of the role of locoregional therapies in modern treatment modalities^[10^]. Concurrently, radiotherapy (RT) has undergone substantial methodological refinement, encompassing magnetic resonance-guided RT, stereotactic body radiotherapy (SBRT) with real-time tumor tracking, and proton beam therapy. These innovations have enabled significant dose escalation within portal veins affected by thrombus while maintaining strict organ-at-risk parameters^[11,12^]. Recent evidence from prospective studies, including multi-institutional Asian consortium research conducted during the 2020–2024 period, consistently indicates that the mechanisms of RT extend beyond local cytoreduction^[13^]. These mechanisms may encompass immunomodulatory effects that potentially synergize with contemporary systemic agents.

Recent iterations of international guidelines, including those from the American Association for the Study of Liver Diseases and the European Society for Medical Oncology, have increasingly incorporated RT into hierarchical treatment algorithms for selectively chosen cases of PVTT, reflecting the evolving evidence base^[14^]. Nevertheless, considerable knowledge gaps remain within the scientific literature, particularly in the systematic characterization of the research landscape itself. The rapid proliferation of research outputs has culminated in a fragmented evidentiary landscape that lacks systematic structural analysis. The topology of scientific collaboration networks, identification of knowledge brokers, and mapping of intellectual exchange pathways remain insufficiently characterized. Furthermore, while bibliometric methodologies have proven effective in delineating scientific landscapes across oncology, to date, no comprehensive bibliometric analysis has specifically addressed the rapidly evolving research ecosystem regarding RT for PVTT in patients with HCC. Such an analysis is essential not only to map current research frontiers but also to uncover thematic transitions, identify gaps in evidence, and guide future clinical trial design and funding strategies.

A preliminary scoping search across PubMed, Web of Science, and Scopus, prior to this study, confirmed the absence of any dedicated bibliometric investigation specifically targeting the use of RT for PVTT in HCC. Therefore, the present work constitutes an initial systematic effort to map this research domain. Our analytical framework aims to (1) quantitatively map publication trends and citation network structures; (2) construct multi-level collaboration networks at the author, institutional, and national levels; (3) identify emerging research fronts and thematic transitions via temporal slicing analysis; (4) deconstruct the intellectual framework through co-citation network analysis and bibliographic coupling; and (5) project emerging research trajectories using burst detection algorithms. The synthesized findings are expected to provide a data-driven roadmap for clinicians, researchers, and funding bodies, highlighting both established knowledge and promising future directions in this rapidly evolving domain.

Materials and methods

Data source and systematic retrieval protocol

The bibliometric analysis employed a systematic data collection approach utilizing PubMed (https://pubmed.ncbi.nlm.nih.gov/), leveraging its expansive coverage of high-impact scientific literature and robust citation indexing features. The retrieval strategy was developed in alignment with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Search (PRISMA-S) guidelines to ensure methodological rigor and reproducibility^[15^].

The search algorithm utilized an advanced Boolean framework encompassing three conceptual domains, which were iteratively refined based on consultations with hepatology and radiation oncology experts. The defined search strategy was as follows: (“Carcinoma, Hepatocellular” [Mesh] OR “Liver Neoplasms” [Mesh] OR hepatocellular[tiab] OR HCC[tiab] OR liver cancer[tiab] OR liver tumor[tiab] OR liver tumour[tiab] OR hepatic cancer[tiab] OR hepatoma[tiab]) AND (“Portal Vein“[Mesh] OR portal vein[tiab] OR PVTT[tiab] OR PVT[tiab]) AND (thrombosis[tiab] OR thrombus[tiab] OR tumor thrombus[tiab] OR tumour thrombus[tiab] OR tumor thrombosis[tiab] OR tumour thrombosis[tiab] OR neoplastic thrombosis[tiab]) AND (“Radiotherapy” [Mesh] OR “External Beam Radiotherapy” [Mesh] OR radiotherapy[tiab] OR radiation therapy[tiab] OR radiation[tiab] OR irradiation[tiab] OR “Stereotactic Body Radiotherapy” [Mesh] OR SBRT[tiab] OR radiosurgery[tiab] OR “Three-Dimensional Conformal Radiotherapy” [tiab] OR 3D-CRT[tiab] OR IMRT[tiab] OR “Intensity-Modulated Radiotherapy” [tiab] OR “Image-Guided Radiotherapy” [tiab] OR IGRT[tiab] OR “Particle Therapy” [tiab] OR “Proton Therapy” [tiab] OR “Carbon Ion Therapy” [tiab] OR “Neoadjuvant Radiotherapy” [tiab] OR “Adjuvant Radiotherapy” [tiab] OR “External Beam” [tiab] OR EBRT[tiab]) AND (2013/01/01[dp]: 2025/10/26[dp]).

The temporal scope of this study was limited to publications from 1 January 2013 to 26 October 2025, covering a 13-year period to facilitate an analysis of both historical trends and contemporary developments in scientific research. Selection criteria were confined to peer-reviewed original research articles and systematic reviews published in English. Exclusion criteria were rigorously applied to eliminate conference proceedings, editorials, letters, case reports, book chapters, and retracted publications, thus securing the integrity of the analysis (Fig. 1).

Figure 1.

Publication screening flowchart.

The identification and screening process followed a multi-phase approach: the initial search results underwent deduplication through automated algorithms, followed by manual verification; title and abstract screening was performed based on predetermined inclusion criteria; and full-text assessments were conducted against quality thresholds for final selection. This comprehensive process yielded 34 initial studies, with one duplicate discarded, resulting in the screening of 33 publications, of which 25 satisfied all inclusion criteria for final analysis.

Data processing and advanced analytical framework

The 25 selected records, exported from PubMed in MEDLINE format, were subjected to a multi-stage preprocessing pipeline combining algorithmic procedures with manual curation to ensure data integrity and reproducibility.

Duplicate removal. Records were screened for duplicates using a two-step strategy: automated matching by DOI, followed by pairwise comparison of titles and author lists using R’s duplicated() function; all flagged candidates were manually verified, and one duplicate was excluded.

Author name disambiguation: author names were standardized through a hierarchical protocol: ORCID-based matching, where available; otherwise, surname, first initial, and institutional affiliation were used to aggregate publications attributable to the same individual. Homographs were resolved by cross-referencing co-authorship networks and subject expertise, with ambiguous cases adjudicated manually.

Institutional affiliation standardization: Institutional names were normalized to a unified authority list via automated string-matching algorithms (Levenshtein distance ≤ 2) coupled with a custom lookup dictionary that mapped variant aliases to canonical forms; subunits were consolidated under their parent institutions.

Keyword normalization: Author-supplied keywords and MeSH terms underwent consistent preprocessing: lowercasing, removal of extraneous whitespace and punctuation, consolidation of synonyms and abbreviations (e.g., PVTT → portal vein tumor thrombosis; SBRT → stereotactic body radiotherapy), and stemming of plurals. The standardized lists were cross-checked against MeSH vocabulary for semantic coherence. Emerging research topics were identified through temporal keyword analysis, defined as keywords with a median publication year of 2021 or later that exhibited a significant upward trajectory in frequency during the study period, supplemented by burst-detection analysis using the bibliometrix package.

Country assignment: Corresponding author affiliations were used to extract country information; when multiple affiliations were present, the primary affiliation was selected. Country names were standardized using the countrycode package in R. For articles with corresponding authors from different countries, each contributing country was credited separately to reflect multinational collaboration. A quantitative bibliometric analysis and network visualization were performed using the R statistical computing environment (version 4.3.1; R Foundation for Statistical Computing, Vienna, Austria), utilizing the bibliometrix package (version 4.1.2; Aria & Cuccurullo, Naples, Italy) for essential bibliometric functions. Network visualization was executed using force-directed algorithms, specifically the Fruchterman–Reingold and Kamada–Kawai methods, implemented via the igraph package (version 1.5.1; Csárdi & Nepusz, Vienna, Austria), prioritizing clarity and interpretability. This comprehensive analytical workflow guaranteed a rigorous examination of the research landscape while maintaining methodological transparency and computational reproducibility.

Co-citation networks were constructed using biblioNetwork (type = “co-citation“) with edges weighted by co-citation frequency; references with < 5 citations were excluded. Bibliographic coupling used type = ”coupling”, requiring ≥ 2 shared references per link. Thematic clustering employed thematicMap() with a co-occurrence threshold of three and Walktrap community detection (100 iterations). Collaboration networks were generated via collaboration() at the author, institutional, and country levels, with edge weights reflecting co-authorship frequencies. Centrality metrics (betweenness, closeness, PageRank) were computed using igraph’s betweenness(), closeness(), and page_rank(). All network visualizations used the Fruchterman–Reingold layout (layout_with_fr).

Goodness-of-fit for Lotka’s Law was assessed via the bibliometrix package’s lotka() function, which estimates the empirical β coefficient by regression and performs a Kolmogorov-Smirnov (K-S) one-sample test comparing observed author productivity with the theoretical inverse-square distribution (β = 2).

Results

Analysis of publication output and citation impact

Figure 2 presents the annual distribution of publications and their corresponding average citations (AvgCitations) related to RT for PVTT in patients with HCC from 2013 to 2025. The data indicate a generally increasing trend in publication output during this period. Initially, the field demonstrated low productivity, with only one to three publications per year from 2013 to 2018. A significant rise in annual output occurred in 2023, with five publications, suggesting a considerable increase in research activity in recent years. An analysis of AvgCitations revealed a clear temporal pattern within the dataset. Earlier publications, particularly from 2013, exhibited the highest AvgCitations (AvgCitations = 5.31), reflecting the longer citation window available to them. In contrast, the AvgCitations for publications from 2020 onward showed a noticeable decline. The most recent publications, from 2023 to 2025, have recorded minimal AvgCitations to date, an occurrence attributable to their relatively brief time since publication. This inverse correlation between publication recency and AvgCitations is a common feature of bibliometric datasets. It suggests that while foundational literature from earlier years remains frequently cited, there is a concomitant rapid expansion of contemporary research output in the field.

Figure 2.

Trends in publications and citations on radiotherapy for PVTT in patients with HCC from 2013 to 2025.

Three-factor analysis

Figure 3 depicts the collaborative network among authors, author affiliations, and publication venues within the research domain of RT for PVTT in patients with HCC. The analysis identifies Park Hee Chul and Yu Jeong Il as two of the most prolific contributors to this field. Notable author affiliations that significantly contribute to the literature include The Catholic University of Korea, Zhejiang University School of Medicine, and The Fifth Medical Center of PLA General Hospital. Knowledge dissemination primarily occurs through established oncology journals, particularly Cancers, BMC Cancer, and Hepatology Research. The network reveals consistent publication pathways between specific institutions and journals, highlighting structured channels for knowledge dissemination. This configuration underscores a clinically-oriented research field centered around major academic medical centers, characterized by well-defined patterns of scholarly output.

Figure 3.

Three-factor analysis of radiotherapy for PVTT in patients with HCC.

Journal distribution analysis

An examination of the publications reveals a concentrated body of knowledge within this research domain. As illustrated in Figure 4, the 14 publications are distributed across 10 journals, with three journals forming the core of the knowledge base: Radiation Oncology (n = 3), Cancers (n = 2), and Frontiers in Oncology (n = 2). The journal distribution indicates a strategic balance between specialized journals focused on radiation oncology and more comprehensive oncology publications. This pattern underscores the dual nature of the field, encompassing both the technical aspects of RT and the broader management of HCC. The predominance of specialized radiation oncology journals (Radiation Oncology) alongside high-volume open-access oncology journals (Cancers, Frontiers in Oncology) reflects the field’s dual emphasis on technical RT innovation and broad clinical-oncological applicability.

Figure 4.

Top 10 journals.

Author productivity analysis

An analysis of author productivity was conducted to evaluate research contributions in the field of RT for PVTT in patients with HCC. As illustrated in Figures 5A and 10, authors emerged as the most prolific contributors, each of whom had published 2 works. These key authors include Aikata Hiroshi, Duan Xuezhang, Ji Xiaoquan, Kim Myungsoo, Kwon Jung Hyun, Park Hee Chul, Shui Yongjie, Wang Quan, Yu Jeong Il, and Yue Jinbo. The application of Lotka’s Law, depicted in Figure 5B, revealed a significant deviation from the expected theoretical distribution of scientific productivity. Among the 200 authors analyzed, 190 (95%) contributed only one publication, which notably exceeded the theoretical expectation of 80%. In contrast, the observed proportion of authors contributing two publications (5%, n = 10) was considerably lower than the projected 20%. The K-S goodness-of-fit test yielded an empirical β of 2.82 (R² = 0.94, P < 0.001), confirming a statistically significant deviation from Lotka’s theoretical distribution. This productivity distribution underscores a research environment primarily characterized by transient contributors, with a small core group comprising consistently productive authors. The field exhibits characteristics of an emerging research domain, marked by limited sustained academic engagement and collaborative networks dispersed across multiple institutions, with no predominant individual contributors.

Figure 5.

Author analysis of publications on radiotherapy for PVTT in patients with HCC. (A) Analysis of author productivity. (B) Author productivity according to Lotka’s law.

Figure 10.

Network analysis. (A) Coupled-cluster analysis. (B) Network of co-cited references. (C) Author collaboration network.

Institutional contribution analysis

Institutional productivity analysis, depicted in Figure 6A, reveals a research landscape predominantly shaped by major academic medical centers. Leading this domain is The Catholic University of Korea, with seven publications, followed by Seoul National University Hospital and The Fifth Medical Center of PLA General Hospital, each with six publications, whereas Zhejiang University School of Medicine has five. Figure 6B presents a temporal analysis of productivity trends for the top five contributing institutions. Both Sungkyunkwan University and Zhejiang University have demonstrated consistent research output since 2016 and 2018, respectively. Notably, the Catholic University of Korea has exhibited substantial growth, increasing its annual publications from 2 (2020–2022) to 7 (2023–2025). Likewise, Seoul National University Hospital and The Fifth Medical Center have displayed remarkable increases in productivity during the 2023–2024 period. This evolving pattern highlights a dynamic research environment, wherein established centers maintain their output while emerging institutions rapidly enhance their contributions, indicative of an expanding global engagement in this specialized field.

Figure 6.

Analysis of institutional contributions. (A) Institutional productivity analysis. (B) Productivity trends over time for the highest contributors.

Distribution of countries and collaborative patterns

An analysis of the distribution of research output among countries, as presented in Figure 7A, revealed a significant concentration of productivity in East Asia. Notably, China and South Korea are the leading contributors, with 34 and 31 publications, respectively. In contrast, Japan and India produced moderate output, with 5 and 7 publications, respectively. An examination of the distribution of corresponding authors indicates limited international collaboration in the field of RT for PVTT in patients with HCC, as depicted in Figure 7B. China had the highest number of single-country publications (n = 11), while South Korea exhibited the greatest degree of international collaboration, as evidenced by two multiple-country publications (MCPs). Figure 7C illustrates a notable increase in research output from both China and South Korea, with China demonstrating exceptional growth, rising from one publication in 2013–34 by 2024. South Korea exhibited consistent growth throughout the study period, culminating in 31 publications by 2025. The citation analysis presented in Figure 7D indicates a correlation with publication volume, with China leading in total citations (n = 106), followed by South Korea (n = 50) and Japan (n = 25). This distribution suggests that the academic impact of national research programs aligns with their productivity levels. Overall, the landscape of research in this field is predominantly shaped by Asian nations, characterized by limited international collaboration and an association between productivity and citation impact.

Figure 7.

Analysis of country distribution and collaborative patterns in radiotherapy for PVTT in patients with HCC. (A) Distribution of corresponding authors by country. (B) Distribution of authors by country. (C) Research output by country over time. (D) Top four most-cited countries.

Keyword evolution and research trends

A treemap depicting the frequency of author keywords elucidates the conceptual evolution within the research domain of RT for PVTT in patients with HCC, as illustrated in Figure 8A The predominant keywords consistently focus on “HCC,” “PVTT,” and “RT,” indicating a stable research emphasis throughout the study period, as demonstrated in Figure 8B. Emerging topics were identified based on temporal keyword dynamics, specifically keywords with a median publication year of 2021 or later and a sustained increase in frequency, and were further validated by burst-detection analysis. The thematic evolution is characterized by three distinct phases, as illustrated in Figure 8C.

Figure 8.

Visualization analysis of author keywords. (A) Treemap of author keywords. (B) Changing trends in the top 10 author keywords over time. (C) Trend topic analysis of author keywords.

The initial phase (2014–2018) established foundational concepts, introducing “stereotactic body radiotherapy” alongside basic anatomical terminology. The intermediate phase (2019–2022) incorporated prognostic evaluations, such as “prognostic factors” and “survival,” as well as technological advancements, such as “radiomics,” reflecting an increase in methodological sophistication. The current phase (2023–2025) underscores emerging research frontiers, particularly the integration of targeted therapies, such as “lenvatinib”, and enhanced disease characterization.

Trend analysis revealed a growing emphasis on personalized, data-driven treatment approaches. The recent emergence of “radiomics” (median year: 2021) and “lenvatinib” (median year: 2024) signifies a transition from traditional RT methods to integrated modalities that combine advanced imaging analytics with molecular therapeutics. This evolution in key terminology illustrates a dynamic research paradigm – transitioning from the development of fundamental radiotherapeutic techniques to the formulation of personalized treatment strategies – characterized by the integration of advanced radiation technologies, quantitative imaging, and targeted systemic therapies.

Thematic structure and knowledge trajectory

Co-occurrence network analysis uncovered a well-established knowledge framework centered on “HCC,” marked by significant structural importance, as evidenced by the highest betweenness (43.49) and PageRank (0.3274) centrality values, as illustrated in Figures 9A,B. The network delineates five distinct thematic clusters: foundational disease knowledge, radiotherapeutic interventions, prognostic evaluation, integration of targeted therapies, and technical optimization. Thematic evolution analysis demonstrated notable conceptual stability, with primary themes exhibiting strong continuity across temporal intervals (stability index: 0.20–0.25), as illustrated in Figure 9C. The consistently elevated weighted inclusion index (0.625) for key concepts indicates a sustained research focus while allowing for conceptual refinement. Significantly, emerging research dimensions identified via factor analysis suggest a notable shift in research emphasis, as depicted in Figure 9D. The integration of immune checkpoint inhibitors into conventional RT reflects an emerging research direction that emphasizes multimodal treatment combinations. This evolution reflects a transition from establishing technical foundations to developing comprehensive therapeutic frameworks that synergize localized radiation with systemic immunomodulation. Structural analysis indicates a cohesive yet dynamically evolving research landscape, where stable core concepts underpin the integration of novel therapeutic strategies, reflecting a field that is consolidating its methodological foundations while actively integrating emerging therapeutic concepts.

Figure 9.

Analysis of thematic structure and high-frequency keywords. (A) Co-occurrence network of keywords. (B) Thematic map of keywords. (C) Evolution of themes using the most frequently used keywords. (D) Factor analysis.

Network analysis of co-citation and author collaboration

Coupled cluster analysis revealed two structurally distinct yet complementary research paradigms, as depicted in Figure 10A. Cluster one demonstrates significant thematic cohesion, integrating topics such as HCC, PVTT, and SBRT, with elevated confidence levels ranging from 63.6% to 75%. This cluster signifies a specialized research stream dedicated to precision RT. Conversely, Cluster two encompasses a broader array of therapeutic investigations employing conventional RT methods. The comparable centrality metrics of the two clusters (6.349 versus 6.122) underline their equal importance in shaping the research landscape. Moreover, the co-citation network analysis identified Llovet JM (2008) as a pivotal reference, exhibiting exceptional betweenness centrality (558.24) and serving as the primary conceptual bridge across research domains, as presented in Figure 10B. The network configuration illustrates an advanced knowledge architecture, with Nakazawa T (2008) and Finn RS (2020) as additional key nodes. Remarkably, recent contributions by Shui Y (2018) and Bettinger D (2019) demonstrate high closeness centrality (0.2), indicating their rapid integration into the research discourse and reflecting the dynamic evolution of the field. An examination of the author collaboration network reveals a decentralized structure comprising 12 distinct research groups, as depicted in Figure 10C. Aikata Hiroshi has emerged as the principal connector within the network, possessing the highest betweenness centrality (45.0), thereby facilitating knowledge exchange across clusters. The identification of multiple research teams with high closeness centrality (0.5–1.0) suggests efficient internal communication; however, limited cross-cluster connections indicate opportunities for enhanced interdisciplinary collaboration. Collectively, these network analyses illustrate a well-developed research domain typified by clearly delineated knowledge structures, established collaboration patterns, and identifiable key contributors acting as conceptual intermediaries. The synergistic relationship between specialized and comprehensive therapeutic approaches, along with the dynamic incorporation of recent evidence, underscores the ongoing evolution and methodological refinement of the field.

Analysis of high-impact publications

The analysis of highly cited publications with substantial impact factors (IFs) revealed distinct patterns of scholarly influence within the research domain, as depicted in Figure 11. Seminal works by Xi Mian (2013) and Shui Yongjie (2018) have established foundational evidence, accumulating 69 and 42 citations, respectively, despite their publication in journals with moderate IFs (PLOS ONE, IF: 2.6; Radiation Oncology, IF: 3.2). These studies primarily focused on validating SBRT as an effective treatment for HCC with PVTT. Recent high-impact publications indicate a shift in research focus, with studies appearing in prestigious journals such as the International Journal of Radiation Oncology, Biology, Physics (IF: 6.5) and Hepatology International (IF: 6.1), as outlined in Supplemental Digital Content Table S1, available at: https://links.lww.com/MS9/B458. These investigations explore advanced therapeutic combinations that integrate RT with immunotherapy and targeted therapies. Nevertheless, their recent publication dates correlate with relatively low citation counts. The disparity between citation metrics and journal IFs reflects the evolution of the field: foundational technical studies retain substantial scholarly significance, while contemporary research targets innovative therapeutic paradigms through high-impact publication outlets, as illustrated in Supplemental Digital Content Table S2, available at: https://links.lww.com/MS9/B458. This trend emphasizes both the established clinical importance of radiotherapeutic approaches and the ongoing innovation in treatment strategies.

Figure 11.

Authors of papers with relatively high citation rates and impact factors.

Discussion

To the best of our knowledge, this study provides one of the first systematic mappings of the intellectual structure and evolutionary trajectory of research on RT for PVTT in patients with HCC. Unlike narrative reviews, this bibliometric approach provides a quantitative, data-driven synthesis of publication trends, collaboration patterns, and thematic shifts in this domain. Our analysis reveals a rapidly evolving research domain characterized by distinct geographical concentrations, a marked transition toward multimodal and image-guided therapeutic paradigms, and a collaborative network indicative of the emergence of an interdisciplinary specialty.

The notable geographical concentration of research productivity – predominantly associated with institutions in East Asia, particularly in China and South Korea – represents a principal finding of this study. This distribution starkly deviates from the wider landscape of oncological research, which typically features significant contributions from North America and Europe. This disparity is attributable to several factors, including the higher prevalence of HBV-related HCC and the consequent increase in the clinical burden of PVTT in East Asia^[16–18^]. Such conditions have necessitated a more insistent and innovative approach to locoregional control in these regions^[19^]. Our results quantitatively corroborate the qualitative findings of prior systematic reviews, including a study by Kong et al (2021), which highlighted the substantial contribution of Asian studies to the evidence base for PVTT RT^[20^]. This geographical concentration, however, raises questions regarding the generalizability of these findings to Western populations, where non-alcoholic steatohepatitis constitutes an increasingly prevalent etiology^[1^]. The biological behavior of PVTT and its response to RT may differ across etiological contexts, underscoring the need for multi-regional validation studies. The sustained increase in research output from China and South Korea underscores their indispensable role in generating clinical evidence, often through extensive single- and multi-institutional studies that have prospectively validated advanced techniques such as SBRT^[21,22^].

Our analysis of keywords and trending topics identified a thematic evolution that provides an empirical narrative illustrating the scientific maturation of the field. Initially, the discipline experienced a technical foundation phase (2014–2018) that focused on establishing the safety and efficacy of SBRT and 3D-CRT. This was succeeded by an intermediate phase of prognostic sophistication (2019–2022), characterized by the inclusion of survival analysis and the emerging application of radiomics. Currently, we are entering a precision integration phase (2023–2025), typified by the convergence of RT with molecularly targeted therapies, such as lenvatinib, and enhanced disease characterization^[23,24^]. This progression objectively reflects the adaptation of the field to broader oncological trends. The introduction of “radiomics” (median year: 2021) signifies a pivotal advancement toward data-driven, personalized predictions of treatment response, a paradigm increasingly advocated within radiation oncology across diverse disease sites^[25^]. Nevertheless, the clinical translation of radiomic signatures into routine decision-making for PVTT remains constrained by limited external validation and the absence of standardized feature extraction protocols. Similarly, the recent prominence of “lenvatinib” (median year: 2024) highlights ongoing investigations into synergistic combinations, a research frontier that is consistent with clinical hypotheses proposed in reviews, such as the review by He et al (2022). These hypotheses suggest the potential for RT to modulate the tumor microenvironment and enhance the efficacy of systemic therapies^[26^]. However, the evidence base for such combinations in PVTT remains predominantly retrospective, necessitating prospective randomized trials to define optimal sequencing and patient selection.

Our co-citation network analysis offers detailed structural insights into the foundational intellectual components of the field. The identification of Llovet JM (2008) as a pivotal reference, marked by an impressive betweenness centrality of 558.24, is particularly noteworthy. This seminal work, which introduced the Barcelona Clinic Liver Cancer staging system, serves as the primary conceptual link connecting discussions on patient stratification and therapeutic allocation. Its vital role objectively reinforces the view that the management of PVTT in patients with HCC, despite technological advancements, remains fundamentally grounded in established prognostic frameworks. This persistence suggests that future staging systems may benefit from incorporating RT-specific parameters – such as target volume or radiomic features – alongside conventional tumor-node-metastasis and liver function assessments. Furthermore, the bibliographic coupling analysis reveals two structurally distinct yet complementary research paradigms: one focused on precision SBRT and the other encompassing a broader range of conventional RT approaches. The balanced centrality metrics between these clusters indicate that targeted technical optimization is advancing alongside the development of more comprehensive therapeutic strategies.

The analysis of authorship and collaboration within the field suggests that it is still in a phase of consolidation. The significant divergence from Lotka’s Law, with 95% of authors being transient contributors, characterizes this area as emerging, highly specialized, and collaborative. This trend contrasts with more established oncology subfields and indicates that knowledge production is currently driven by multiple, often institutionally bound teams. While this reflects a dynamic and growing interest, it also underscores potential challenges in maintaining long-term, comprehensive investigative programs. The author collaboration network, despite being decentralized, has identified key knowledge brokers, such as Aikata Hiroshi, who demonstrates high betweenness centrality and facilitates cross-institutional exchanges. Nonetheless, the observed limited international collaboration, as evidenced by a low MCP ratio, presents a significant opportunity. Solidifying global partnerships will be essential for validating findings across diverse patient populations and accelerating the translation of novel strategies into global standards of care^[27^]. Strategic funding and policy initiatives should prioritize the formation of transnational consortia to overcome these collaborative barriers.

The observed shift toward integrating RT with immune checkpoint inhibitors or tyrosine kinase inhibitors aligns with contemporary multimodality oncology paradigms. Yet, as noted above, prospective evidence for these combinations in PVTT remains sparse, and the optimal timing, sequencing, and biomarker-driven patient selection are undefined. Clinicians should therefore interpret existing retrospective data with appropriate caution.

By quantitatively delineating thematic transitions and collaboration networks that were previously discernible only through narrative summaries, this study fills a critical methodological gap and offers a data-driven reference for future research planning. Importantly, these bibliometric findings reflect patterns in the published literature and should not be equated with direct evidence of clinical efficacy. The clinical utility of specific interventions identified in this analysis requires independent validation through rigorous prospective trials. This bibliometric analysis is subject to certain limitations. First, the data were exclusively sourced from PubMed, which may overlook relevant studies indexed in Web of Science, Scopus, or Embase. Second, the inclusion criteria were restricted to English-language articles and reviews, potentially introducing a language bias. Third, publication bias may affect the observed trends, as positive findings are more likely to be published and cited than null or negative results. Fourth, as a bibliometric study, it provides an overview of publication structure and trends without assessing the quality of the clinical evidence presented in these publications. Fifth, citation-based metrics inherently suffer from citation lag, whereby recent publications have had insufficient time to accumulate citations and may thus be underrepresented. The small sample size (n = 25) represents a critical methodological caveat, as it constrains the statistical power and reliability of the advanced bibliometric techniques employed. The sparsity of co-citation and co-occurrence matrices renders centrality measures and cluster solutions highly susceptible to the disproportionate influence of a few key publications. Moreover, network metrics are sensitive to threshold and algorithm choices; although parameters were specified in the methods, the robustness of these findings to alternative configurations remains unexamined. With annual outputs ranging from 2 to 5 articles, temporal trend detection – including thematic phase allocation and burst identification – is prone to stochastic year-to-year variability. Moreover, regression-based and factor-analytic models are at risk of overfitting with limited observations, while collaboration indices (e.g., the multiple-country publication ratio) may merely reflect sample sparsity rather than genuine international engagement. Author name disambiguation also poses a persistent challenge, particularly for East Asian names, and despite ORCID-based matching and manual verification, residual inaccuracies cannot be excluded. Accordingly, the quantitative outputs should be interpreted as exploratory and hypothesis-generating, necessitating future corroboration using expanded datasets from additional bibliographic sources.

In conclusion, future research should prioritize prospective randomized trials to validate the promising combination therapies identified in recent high-impact publications. Moreover, the formation of international consortia will be essential to address geographic evidence biases and enhance the generalizability of the findings. Finally, thorough translational investigations into the radiogenomic and immunomodulatory mechanisms of radiation in the specific context of PVTT are necessary to rationally inform future combination strategies.

Conclusions

This bibliometric analysis provides a systematic, quantitative synthesis of the research landscape regarding RT for PVTT in patients with HCC. It objectively confirms the leadership of East Asian researchers in this field, emphasizes the vital role of established staging systems, and indicates a notable transition in research focus toward image-guided RT and combination-based therapeutic strategies. Furthermore, it reveals previously overlooked patterns, such as dual research paradigms, a decentralized collaboration network, and the precise timing of thematic transitions, which were not apparent in earlier qualitative analyses. These findings offer a strategic framework for researchers, clinicians, and funding bodies to effectively navigate the future of this critical therapeutic area.

Ethical approval

Not applicable.

Consent

Informed consent was obtained from the participants.

Sources of funding

This study was supported by the First Affiliated Hospital of Zhengzhou University Joint construction projects (LHGJ20220393).

Author contributions

X.L.: data curation (equal); formal analysis (equal); investigation (equal); project administration (equal); writing – original draft (lead). H.G.: data curation (equal); formal analysis (equal); project administration (equal); writing – original draft (equal). K.L.: data curation (equal); project administration (equal). X.L.: data curation (equal). M.G.: data curation (equal); formal analysis (equal); investigation (equal); project administration (equal); writing – review and editing (equal).

Conflicts of interest disclosure

The authors have no conflicts of interest.

Research registration unique identifying number (UIN)

Not applicable.

Guarantor

Xiufang Lv.

Provenance and peer review

The article was submitted directly by the authors and had been underwent external review.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.