This study utilized the Web of Science Core Collection (WOSCC) as the literature search database. Visual mapping was conducted using software such as CiteSpace, VOSviewer, Bibliometric tools, and Microsoft Excel. The search terms were “TS =(neoplasm of the liver OR liver neoplasm OR carcinoma liver OR carcinoma of the liver OR liver cancer OR cancer of the liver) AND TS = (bone metastas* OR skeletal metastas* OR metastas* carcinoma of bone OR metastas* tumor of bone OR metastas* cancer of bone) OR TS = (liver cancer bone metastas* OR bone metastas* from liver cancer)”. The following adjustments were made: Document type: “Article/Review Article”; Language: “English”; Time: “from 2004.01.01 to 2024.10.31”. The field label “TS = Topic” was used with the publication date as the selection criterion and duplicate documents removed, resulting in the final retrieved material ( Figure 1 ).

Bibliometric analysis is a method used to describe and analyze the dynamics and progress of research in a particular field. With the assistance of modern computing technology, it visualizes the analysis results in the form of knowledge maps, which not only facilitates data analysis however also helps to uncover the intrinsic connections within this information.

Bibliographic coupling, co-authorship, co-occurrence, citation, and theme patterns of each publication were examined using VOSviewer (version 1.6.20) (22), including network visualization, overlay visualization, and density visualization, it is characterized by simple mapping and visually appealing images.

Professor Chaomei Chen developed CiteSpace (version 6.3 R1) (23), as a tool to analyze the evolutionary process of knowledge. A set theory-based method standardizes data to measure the similarity of knowledge units. This method is used within time slices to create regional views and timeline graphs. This approach clearly outlines the evolutionary process of knowledge and the historical span of literature within a cluster. As a result, it helps to understand the development and trends in the field.

We utilize R-bibliometrics software (Version R 4.0) to analyze research hotspots and trends, focusing on keywords, trending topics, and thematic maps (24).

We retrieved a total of 4,068 papers related to liver cancer with bone metastasis. After screening, we selected 3,845 papers, which included 3,347 “articles” and 498 “reviews,” as illustrated in Figure 2 . we noted that the number of publications increased each year from 2004, and reached a peak of 286 papers in 2021, with an annual growth rate of 8.3%, since then, the number of publications has remained relatively stable.

A total of 92 countries contributed to these publications. The top 10 publications were dominated by Asia (n=4), the Americas (n=2), Europe (n=3), and Eurasia (n=1), with Asia standing out as particularly prominent. China had the highest number of publications (n=970, 29%), followed by the United States (n=648, 19.4%), Japan (n=317, 9.5%), and Germany (n=228, 6.8%) ( Figures 3a, b ; Table 1 ). The annual publication output of the United States had seen rapid and steady growth, while China had significantly increased its number of publications annually since 2019 ( Figures 3c, d ). Canada’s international collaboration rate was significantly higher at 44.8% compared to China (9.3%), the United States (23.1%), and Japan (2.8%), suggesting that Canada had a more extensive international collaboration network. The publication with the highest number of contributions from China requires further improvement in its international collaboration efforts ( Figures 3e, f ).

We established a minimum requirement of 5 publications and utilized VOSviewer to identify a total of 13,188 organizations. Using the publication counts and the relationships among organizations, we created a visual collaboration network divided into 5 clusters, as illustrated in Figure 4a . The top 10 publishing organizations were primarily from 3 countries: China (n = 6), the United States (n = 3), and Germany (n = 1), with publication rates of 10.6%, 5.14%, and 1.39%, respectively ( Table 2 ).

A total of 830 journals have published research on liver cancer with bone metastasis. The journal with the highest number of publications was “Frontiers in Oncology” (n=88, 3.86%), followed by “BMC Cancer” (n=65, 2.86%), “Medicine” (n=59, 2.59%), and “PLOS ONE” (n=57, 2.5%). Among the top 10 journals ranked by impact factor, the highest was the Journal of Nuclear Medicine (IF = 9.1), followed by the European Journal of Nuclear Medicine and Molecular Imaging (IF = 8.6) and Frontiers in Oncology (IF = 6.24). Following this, we filtered 187 journals that had at least five related publications and constructed a journal network, additionally, we observed strong cooperation among countries and various institutions ( Table 3 ; Figure 5a ). Figure 5b showed the publication curve of the top 5 journals from 2004 to 2024. Table 4 revealed that all of the top 10 co-cited journals received over 1000 citations. “J Clin Oncol” led with 4275 citations, followed by “Cancer Res” with 2421 citations.

The next highest citations were for “Cancer Res” (2421), “New Engl J Med” (1983), “Cancer-Am Cancer Soc” (1936), and “Clin Cancer Res” (1639). “New Engl J Med” had the highest impact factor of 158.5, followed by “J Clin Oncol” with an impact factor of 44.54. After this, we filtered 188 journals that had at least 100 related publications and constructed a journal network, as illustrated in Figure 5c . The dual map overlay illustrates the citation relationships between journals and their co-cited counterparts. The left side represents the citing journals, while the right side shows the co-cited journals. As showed in Figure 5d shows that these two paths are mainly green and yellow, reflecting a diverse array of cited papers. Such as yellow cluster4 (MOLECULAR, BIOLOGY, IMMUNOLOGY), green cluster9(DENTISTRY, BERMATOLOGY, SURGERY), and cluster2 (MEDICINE, MEDICICAL, CLINICAL) journal articles mainly cited to cluster8 (MOLECULAR BIOLOGY, GENETICS) and cluster5(HEALTH, NURSING, MEDICINE), etc.

A total of 23,710 authors contributed to research on liver cancer with bone metastasis. Among the top 10 authors by publication count, 6 authors published more than 10 papers. Wang Xin and Zhang Chao had the highest number of publications, each with 18 papers ( Table 5 ). We constructed a collaboration network based on authors who had published 5 or more papers ( Figure 6a ). Additionally, close collaboration was noted among several authors. For instance, Wang, Xin and Zhang, Chao had a close collaboration, as did Saad, Fred; Briganti, Alberto; Abdel-Rahman, Omar and Tong, Yuexin. From the 54,118 co-cited authors, we included 153 articles in the visual analysis by setting a minimum citation threshold of 30. There were 9 authors with more than 100 co-citations, with Jemal, A. leading at 313 citations. Active collaboration was also observed among various co-cited authors, including Jemal A., Siegel RL., Motzer RJ., Eisenhauer EA., and Coleman RE. ( Table 6 ; Figure 6b ). Figure 6c showed that Wang, Z. was the first to publish on this topic in 2007, followed with Zhang Y.; Wang Y. and Zhang J. who researched the topic in 2008, and they continue to work in this direction to date.

Through the co-occurrence analysis of keywords, we could quickly capture research hotspots in a certain field. Figure 7a showed the top 20 high-frequency keywords in research of liver cancer with bone metastasis. Among them, metastasis became the most prominent keyword.

We filtered keywords with the number of occurrences more than or equal to 10 and performed cluster analysis through VOSviewer. The larger the node, the more keyword references; The thicker the line between nodes, the stronger the connection between keywords. As shown in Figure 7b , we obtained 8 clusters in total and represented 8 research directions. For example, the keywords in purple clusters consisted of prostate cancer, pet, pet/ct, staging etc. green clusters consisted of prognosis, bone metastasis, nomogram, colorectal cancer, etc. red clusters consisted of metastasis, breast cancer, hepatocellular carcinoma, metastases, etc. blue clusters consisted of bone metastases, immunotherapy, breast cancer, brain metastases, lung metastases, etc.

The trend topic analysis of the keywords ( Table 7 ; Figure 7c ) showed that from 2004 to 2009, the research in this period mainly focused on fdg pet, messenger-rna, bone sialoprotein, polymerase-chain-reaction, liver-tumors, hormone-related protein, prognostic factors, etc. Since 2010, there had been an explosion of research interest in this direction, mainly involving survival, cancer, therapy, management, expression, metastases, bone-marrow, chemotherapy, etc; In addition, from 2022 to 2024, keywords such as “immunity,” “gastrectomy,” and “guidelines” had frequently appeared. And the emerging drug denosumab might represent the current research hotspot of liver cancer with bone metastasis.

A timeline image could display the time periods and trends for different research keywords. The term co-occurrence analysis revealed a total of 35 major clusters, including “#0breast cancer”, “#1nomogram”, “#2breast cancer subtype”, “#3renal cell carcinoma”, “#4docetaxel”, “#5molecular subtype”, and “#6microenvironment”, among others. Moreover, the research interest in #0, #1, #2, and #3 had surged since 2010 ( Figure 7d ).

The timeline view of clustered maps of 79935 co-cited references from 2004 to 2024. In top 10 co-cited references, there were 3 documents that had been cited more than 100 times, with the highest citation count reaching 167 times (19, 25–33) ( Table 8 ; Figure 8a ). We selected references with co-citation more than or equal to 15 for the construction of the co-citation network map. According to Figure 8b , the citation count and total link strength of the paper Jemal A, 2011, CA-Cancer J Clin significantly exceed those of other publications. Reference with citation bursts refers to those references that were frequently cited by scholars in a certain field over a period of time. In our study, 116 references with strong citation bursts were identified by CiteSpace. As shown in Figure 8c , every bar represents a year, and the red bar represents strong citation burstiness. Citation bursts for references appeared as early as 2008 and as late as 2022. These 20 references had an overall range of 6.45 to 42.46 for burst strength and 3 to 5 years for endurance strength. The most citation surging paper (intensity=42.46) was “Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries (25)” by Sung H et al., published in CA Cancer J Clin in 2021, with an impact factor of 503.1. The citation surge occurred from 2022 to 2024. It described the magnitude and distribution of the overall disease burden and major types of cancer in 2020, briefly comments on the relevant risk factors and the prospects for preventing the major cancers observed globally, and finally predicted the severity of the disease in 2040 based on global population projections.

Based on the statistical analysis of this study, China was the country with the highest number of published papers, accounting for 29% of the total literature, which might be related to the highest incidence of liver cancer in China. In terms of international collaboration rate, Canada ranked the highest, far surpassing China, the United States, Japan, and others, indicating that these major contributing countries needed to strengthen international exchange and cooperation. In the top 10 institutional contribution analysis, Chinese universities had made the greatest contribution (n=6), including Fudan University, Sun Yat-sen University, Shanghai Jiao Tong University, Zhejiang University, Tongji University, and Huazhong University of Science and Technology; followed by American institutions (n=3), such as The University of Texas MD Anderson Cancer Center, Memorial Sloan Kettering Cancer Center, and Harvard University. In terms of journal publication volume and citation frequency, the top 10 journals had a total co-citation count of over 1000, with “Frontiers In Oncology” having the highest output; “European Journal of Nuclear Medicine and Molecular Imaging” had the highest citation rate, and the journal with the highest impact factor was “New Engl J Med” (IF=158.5), demonstrating the importance and reference value of the papers. The dual-map coverage indicated that these journals cover a wide range of fields, including molecular, biological, immunological, cosmetic, surgical, pharmaceutical, medical, nursing, and clinical disciplines. A total of 23,710 authors had participated in research related to liver cancer with bone metastasis, with Wang, Xin and Zhang, Chao being the individual authors who had published the most papers. There had been 54,118 co-cited authors, with Jemal, A (n=313) having the highest number of citations, indicating active collaborative relationships among different authors. Wang, Z was the first to publish research on this topic in 2007, while Zhang Y, Wang Y, and Zhang J conducted research on this topic simultaneously in 2008 and had been actively involved in this research direction to date.

Current diagnostic research in liver cancer bone metastasis primarily centers on molecular markers and imaging technologies. Keywords such as “PET/CT,” “nomogram,” and “serum lnc34a” ( Table 7 ) highlight the emphasis on non-invasive detection and prognostic modeling. Studies have explored functional imaging for metastatic site identification and serum biomarkers (e.g., lncRNAs) for early metastasis prediction. However, diagnostic gaps persist: (1) Limited sensitivity of current imaging in detecting micro-metastases; (2) Lack of standardized molecular markers for routine clinical use; (3) Inconsistent diagnostic criteria across studies, hindering cross-validity. The emerging trend toward multi-omics integration (e.g., genomics and radiomics) aims to address these gaps but remains in the preclinical stage.

We had identified and summarized the research hotspots and frontiers of bone metastasis from liver cancer through the analysis of cited literature, citation bursts, keyword clustering, and keyword topic trends in liver cancer with bone metastasis. Our study highlighted the necessity of liver cancer treatment methods as well as the early monitoring and identification of bone metastasis, involving the following aspects:

Firstly, through keyword co-occurrence analysis, the research had been clustered into eight directions, with “metastasis” being the most prominent keyword. This included cancer detection and treatment methods such as PET, PET/CT, radiotherapy, immunotherapy, targeted therapy, and drugs; types of cancer metastasis such as breast cancer, hepatocellular carcinoma, and colorectal cancer; sites of cancer metastasis involving bone, brain, and lung metastases. Meanwhile, the timeline visualization could display the time periods and trends of different research keywords, revealing a total of 35 main clusters. Since 2010, there had been a significant increase in research interest related to #0 breast cancer, #1 nomogram, #2 breast cancer subtypes, and #3 renal cell carcinoma. The emerging drug denosumab may be a current research hot spot for the treatment of liver cancer with bone metastasis (21, 34–39).

Secondly, the literature construction network graph showed that the paper titled “Global Cancer Statistics (26) “ by Jemal A, et al., published in the CA-Cancer J Clin journal in 2011, had the highest number of citations (n=167) and the highest impact factor (IF 503.1), with a Web of Science (WOS) quartile of Q1 and a total link strength of 364. This paper analyzed cancer cases for the year based on GLOBOCAN 2008 and compared them with common causes of cancer death over the past decade, suggested that the low survival rate of cancer was likely related to late diagnosis and limited opportunities for timely or standard treatment. It highlighted the positive role of promoting cancer control knowledge, such as implementing tobacco control, vaccination (for liver cancer and cervical cancer), early detection and treatment plans, as well as public health campaigns promoting physical activity and healthier dietary intake in preventing the global cancer burden.

Thirdly, CiteSpace identified 116 papers with strong citation bursts, with the earliest citation burst appeared in 2008 and the latest in 2022. The overall range of burst strength for the top 20 references was from 6.45 to 42.46, and the overall range of endurance strength was from 3 to 5 years. The paper with the highest citation strength (strength=42.46) was “Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality of 36 Cancers in 185 Countries Worldwide (25)” by Sung H et al., published in CA Cancer J Clin in 2021, with an impact factor of 503.1 and the citation burst occurred from 2022 to 2024. It conducted a statistical analysis of the size and distribution of the overall global cancer burden and types of cancer, and predicted the severity of the disease in 2040, guiding research in the field.

This study was the first to employ bibliometric methods to analyze the current research status of liver cancer with bone metastasis and use three bibliometric tools for data visualization, which helped researchers better understand the field and explored new research directions. However, there were some limitations. First, we only selected the Web of Science (WOS) database and English languages as the data source, which might result in the omission of other research. Second, there were significant differences in the incidence rate of liver cancer, diagnostic methods and treatment drugs among countries, and accordingly, research focuses and advancements were also varied. Finally, recent high-quality articles might not receive sufficient attention due to their short publication time and low citation frequency. For example, the International Agency for Research on Cancer (IARC) published the latest global cancer statistics for 2022 in an article in May 2024 (“Global Cancer Statistics 2022: GLOBOCAN Estimates of Incidence and Mortality for 36 Cancers in 185 Countries Worldwide”) and this piece was overlooked precisely because of this reason, despite its high quality (40). however, due to the varying cancer incidence rates in each country, indicated the necessity to combine with the specific circumstances of each country. In summary, despite these limitations, our study offered a comprehensive overview of the clinical research, monitoring/treatment methods and prognosis of liver cancer with bone metastasis, which were also of significant value for understanding the current research trends in this field.