Abstract

Objectives:

Bibliometrics was adopted to analyze the research trends of colorectal cancer organoids, and to identify the global research hotspots and prospective directions.

Methods:

Publications related to colorectal cancer organoids published from January 1, 2010 to December 31, 2025 were obtained from the Web of Science Core Collection (WOSCC) and Scopus databases. Bibliometric and visualization analyses were conducted utilizing Bibliometrix R package, Python, CiteSpace and VOSviewer to explore the publication trends, countries, institutions, journals, authors and thematic evolution.

Results:

A total of 2239 relevant publications were included. The analysis revealed an explosive growth in the number of publications in this field over the past 16 years. China and the United States were the main contributors, and Utrecht University was the most productive institution. Hans Clevers had the highest total citations and number of publications, reflecting his leading influence in the field. Cancers was the most productive journal, while Nature was the most frequently cited. Keyword analysis and thematic evolution revealed that the current research hotspots primarily include drug screening, the tumor microenvironment, tumor immunity, and precision oncology.

Conclusion:

Research on colorectal cancer organoids is shifting from the construction and optimization of conventional organoid models to the development of high-fidelity disease models. Current studies focus on the simulation of the tumor microenvironment and personalized cancer therapy, and colorectal cancer organoids will achieve broader clinical applications in precision oncology in the future.

1 Introduction

Colorectal cancer (CRC) is the third most prevalent malignancy and the second most common cause of cancer mortality worldwide (1). While colorectal cancer incidence is declining among older adults in high-income countries, rates continue to rise in emerging economies and among young adults (age <50 years) worldwide (2, 3). Distant organ metastasis is the leading cause of death from CRC, with a 5-year relative survival rate of only approximately 14% (4). Surgery is the main treatment for CRC, supplemented by adjuvant chemotherapy and radiotherapy (5). However, treatment is often hindered by drug resistance (6). Additionally, many patients are diagnosed at advanced stages, losing the opportunity for curative surgery, or suffer from postoperative recurrence and metastasis. To improve both basic research and clinical management of CRC, traditional preclinical models are no longer adequate to meet current demands, necessitating the exploration of preclinical models with high fidelity (7).

Organoids are three-dimensional in vitro cultured tissues that recapitulate key structural and functional aspects of their in vivo counterparts. They can be derived from pluripotent stem cells, tissue-resident stem cells, progenitor cells or differentiated cells from normal or diseased tissues (e.g., epithelial and non-epithelial tumors) (8, 9). Patient-derived tumor organoids (PDTOs) can maintain interpatient heterogeneity, accurately recapitulate the histological and molecular characteristics of original tumors, and enable long-term culture with structural and genetic integrity (10–12). Currently, an increasing number of prospective clinical trials employ patient-derived tumor organoids (PDTOs) to evaluate drug responses and guide cancer treatment, particularly for colorectal tumors (13). Colorectal cancer organoids are also being employed to investigate cancer differentiation, invasion and metastasis, and hold promising potential for applications in personalized medicine (14–18).

Since 2010, the number of publications on colorectal cancer organoids has grown explosively. However, systematic quantitative analyses of the academic structure, research trends and emerging frontiers in this field are still lacking. In contrast to traditional narrative reviews, bibliometrics employs statistical approaches for the quantitative analysis of scientific literature, serving as a robust tool to reveal the knowledge structure within databases (19, 20). Such analysis not only reveals the fundamental status and research hotspots of the field, but also provides valuable insights and guidance for future studies focusing on model optimization, precision medicine, and interdisciplinary translational research. Therefore, this study utilized bibliometric methods to quantitatively and visually clarify its academic evolution and provide valuable insights into future developmental trends.

In recent years, driven by the rapid advancement of organoid research, related bibliometric and knowledge mapping analyses have increasingly emerged (21–25). Existing studies have largely provided macroscopic overviews of the organoid field or encompassed various disease types, offering valuable references for understanding its overall development. However, systematic analyses focusing on specific tumor types remain relatively insufficient within this body of literature, with studies specifically dedicated to CRC organoids being particularly limited.

Against this backdrop, the present study focuses specifically on CRC organoids. Utilizing a larger, more up-to-date dataset spanning an extended timeframe, we combine multiple bibliometric and visual analysis methods to systematically delineate research hotspots and developmental trends. Furthermore, we place particular emphasis on translational directions such as drug screening, the tumor microenvironment, and personalized medicine, thereby providing a more disease-specific and clinically-oriented analytical perspective.

2 Materials and methods2.1 Data sources and processing

The literature search was implemented on February 09, 2026, utilizing the Web of Science Core Collection (WOSCC) database and Scopus database. The Scopus CSV files were downloaded and converted into plain text format by Python (version 3.14) to unify the data.

To realize data uniformity, this study adopted Python (version 3.14) to translate Scopus CSV files into a plain text format compatible with WOSCC. We also utilized Bibliometrix R package (v5.2.1) to perform data cleaning steps, including DOI-based deduplication, removal of records with “[Anonymous]” in the author field, and merging and standardization of duplicate institution names. Furthermore, the disambiguation function of the Bibliometrix R package was utilized.

2.2 Search strategy

We developed a comprehensive search strategy focusing on organoids and colorectal cancer, as shown in Figure 1. Early 2026 publications were still under continuous updates and incomplete database coverage at the time of retrieval. To ensure dataset integrity and consistency, and avoid bias caused by incompletely indexed records, we only included literature fully indexed and incorporated by December 31, 2025. Literature types were limited to articles and reviews. A final total of 2,239 English publications dated from January 1, 2010, to December 31, 2025 were included. We performed data collection, screening, and data analysis in adherence to the PRISMA systematic review protocol (26). The detailed search strategy is laid out in Supplementary Table 1.

Flowchart of the study.

2.3 Bibliometric analysis and visualization

The Bibliometrix R package (v5.2.1) served as the primary tool for conducting descriptive and quantitative bibliometric analyses to characterize publication trends, productivity of authors, institutions and journals, collaboration networks, citation metrics of top-cited articles, and theme evolution trends (27). CiteSpace (v6.4.R1) was used for the dual map overlay of journals analysis, high-frequency keyword clustering, and burst detection of cited references, while VOSviewer (v1.6.20) was employed to visualize journal co-citation networks (28–30). Additionally, Python’s pandas and matplotlib packages were used for plotting.

3 Results3.1 Descriptive bibliometric characteristics and publication trends

The final dataset was derived by merging the Web of Science Core Collection (WOSCC) and Scopus databases, with the merged data exhibiting trends consistent with those of the individual databases. After removing duplicates and excluding irrelevant articles in accordance with the predefined criteria, a total of 2,239 publications (including 1,794 articles and 445 reviews) were included. The corpus encompasses a substantial body of references, Keywords Plus, and author keywords from 2010 to 2025 (Table 1; Supplementary Figure 1). International collaboration accounted for 31.8% of publications, with an average of nearly eleven co-authors per document. An annual growth rate of 42.03% indicates that this field is in a phase of extremely rapid expansion. It is noteworthy that the proportion and average annual growth rate of internationally collaborative papers in the two databases are nearly identical. As shown in Figure 2A, only 2 papers were published in 2010, whereas this number reached 386 in 2025, representing more than a one-hundred-fold increase.

DescriptionWOSCCScopusWOSCC+ScopusTimespan2010:20252010:20252010:2025Sources (Journals, Books, etc)471507601Documents174319012239Annual Growth Rate %4040.5842.03Document Average Age4.44.284.35Average citations per doc44.114645.73References706231248374527Keywords Plus (ID)3439134159511Author’s Keywords (DE)310135063894Authors147531560818498Authors of single-authored docs112227Single-authored docs112328Co-Authors per Doc11.511.711.2International co-authorships %32.0132.6131.8article145015991794review293302445

Descriptive bibliometric statistics of datasets.

Publication trends and geographic distribution. (A) Number of publications in different databases. (B) Annual publication trends with fitted regression curves. (C) Annual trends in mean total citations per article. (D) Top 10 countries contributing to publications. (E) Top 10 institutions contributing to publications.

As illustrated in Figure 2B, curve fitting was performed on the annual publication counts. The fitted curve closely followed the observed annual publication counts, demonstrating a high goodness of fit (R2 = 0.98). The average number of citations per article for publications from 2010 to 2025 exhibits notable peaks in 2011 and 2015, suggesting that those years witnessed the publication of high-impact articles associated with potential technological breakthroughs (31, 32) (Figure 2C). Further details are available in Supplementary Table 2.

Analysis of national publication output showed that China ranked first with 533 publications, succeeded by the United States (469) and Germany (162) (Table 2; Figure 2D). The three leading contributors represented over 50% of global output. Multiple country publications (MCPs) reflected the number of collaborating authors from different nations and regions.

CountryArticlesArticles %SCPMCPMCP %CHINA53323.84359818.4USA46920.933113829.4GERMANY1627.2897345.1JAPAN1587.11283019NETHERLANDS1456.5846142.1UNITED KINGDOM1255.6566955.2ITALY954.2544143.2KOREA843.875910.7AUSTRALIA492.2252449SPAIN431.9261739.5

Publication output and international collaboration patterns of top 10 countries.

Utrecht University in the Netherlands led institutional output with the highest number of publications (n=231). Other top contributors included the Helmholtz Association (173), the German Cancer Research Center (DKFZ) (141), Sun Yat-sen University (135), Harvard University (122), and Fudan University (111). Additional high-output institutions included the University of California System (109), the University of London (94), the University of Texas System (94), and Shanghai Jiao Tong University (90) (Supplementary Table 3; Figure 2E). In the realm of conventional scholarly institutions, Utrecht University stands out for its prominent research output. The results highlight a worldwide research landscape in the field of colorectal cancer organoids, with scholarly input originating from a diverse array of global institutions.

3.2 Visualized analysis of the dual-map overlay of journals

The dual-map overlay of journals can reveal the research evolution across multiple disciplinary fields and capture the information flow at the journal level. As shown in Figure 3, the left portion corresponds to citing journals, and the right portion stands for cited journals. The colored curves in the Figure denote citation paths. The yellow path indicated that articles published in journals of Molecular/Biology/Genetics were frequently cited by journals in Molecular/Biology/Immunology. This analysis revealed the cross-citation relationships among multidisciplinary journals, reflecting the interdisciplinary collaboration and knowledge dissemination within the research field.

Dual-map overlay of journals in colorectal cancer organoid research.

3.3 Analysis of author, country and journal productivity

Author productivity and impact analyses revealed that Hans Clevers was the most productive author (NP = 58, TC = 18291) and held the highest h-index, g-index, and m-index (Table 3). Authors’ production over time showed that Hans Clevers has maintained consistent output and high mean citations per year since 2010, with particularly notable increases in 2018 and 2020 (Figure 4A). Additionally, Onno Kranenburg ranked second in publication output (NP = 29) and held the second-highest g-index, showing a significant surge in productivity in recent years. By utilizing Lotka’s Law, the analysis found that 81.7% of authors contributed only one publication (Figure 4C).

Authorh_indexg_indexm_indexTCNPPY_startCLEVERS HANS41582.41218291582010SATO TOSHIRO18241.1258296242011SANSOM OWEN J.17201.4171095202015YU JUN17232.1252131232019GOEL AJAY15181.667698182018KRANENBURG ONNO13291.625887292019VOEST EMILE E.131413124142014VAN DE WETERING MARC12130.85282132012BATLLE EDUARD11150.9173134152015FARIN HENNER F.11140.917928142015

Top 10 authors by quantitative academic metrics.

Analysis of author, country, and journal productivity. (A) Annual publication trends of the top 10 authors. (B) Cumulative publication trends of the top 10 countries. (C) Author productivity through Lotka's law. (D) Annual publication trends of the top 10 journals.

Analysis of the cumulative number of publications by country (Figure 4B) revealed that the United States (USA) has consistently ranked first in terms of cumulative publications in this field from 2010 to 2025. Since 2010, the number of publications from all countries has shown a steady upward trend, with a marked acceleration in growth after 2015. Particularly notable increases were observed for China and the United States. Quantitative academic metrics at the journal level (Table 4) and annual publication trends (Figure 4D) collectively revealed the influence characteristics and output dynamics of core journals in this field. In terms of academic influence, Nature Communications ranked at the forefront with a comprehensive performance of h-index = 29, g-index = 51, and m-index = 2.636, boasting a total citation count of 3502 and 51 cumulative publications since its debut in 2016. The annual publication output of journals showed significant fluctuations over time. Cancers recorded the highest total number of publications among all journals, while Nature had the highest total citation count.

Sourceh_indexg_indexm_indexTCNPPY_startNATURE COMMUNICATIONS29512.6363502512016GASTROENTEROLOGY28411.6476452412010CANCERS25352.7781664942018NATURE23251.5339868252012ONCOGENE23401.6431645492013CANCER RESEARCH21371.751426382015SCIENTIFIC REPORTS20341.8181220472016CELL REPORTS19331.4621874332014CELLULAR AND MOLECULAR GASTROENTEROLOGY AND HEPATOLOGY19242.111791242018JOURNAL OF EXPERIMENTAL & CLINICAL CANCER RESEARCH18291.8890322017

Top 10 journals by quantitative academic metrics.

3.4 Collaboration analysis

In order to assess the consistency and reliability of bibliographic data for visualizing academic partnerships, collaboration networks were developed at the institutional, national, and individual author levels utilizing both the Scopus and WOSCC databases. Collaboration networks of authors, countries, and institutions were visualized using the Bibliometrix R package (v5.2.1), confirming the stability and complementarity of the two databases in revealing collaborative network structures. While both databases consistently identified major collaborative groups, slight discrepancies in local collaborations suggest the value of combining data from multiple sources.

Author collaboration networks revealed distinct co-authorship clusters, with the size of the nodes determined by author centrality and the line thickness indicating the strength of the connection. The network topology indicated that in both the WOSCC network and the Scopus network (Figures 5A, B), Hans Clevers exhibited extremely high centrality, demonstrating his leadership role in thematic research development. Onno Kranenburg followed closely as a prominent node. However, the overall structure remained characterized by several smaller, closed sub-clusters with strong geographical localization. This structural feature reflects limited extensive collaboration across different author communities.

Collaboration networks at the author, country, and institutional levels. (A) Author collaboration network of WOSCC. (B) Author collaboration network of Scopus. (C) Country collaboration network of WOSCC. (D) Country collaboration network of Scopus. (E) Institutional collaboration network of WOSCC. (F) Institutional collaboration network of Scopus.

The national collaboration network revealed how countries collaborate in colorectal cancer organoid research (Figures 5C, D). Several nations stood out as major contributors. Analyses from both WOSCC and Scopus showed that the USA appeared as the largest central node, while China and Japan both maintained close collaboration with the USA. European nations like the UK, Germany, and the Netherlands exhibited strong internal collaboration, forming distinctive European industrial clusters and establishing cooperative relationships with the USA.

The institutional collaboration map revealed unique patterns in global research center connectivity (Figures 5E, F). In both the WOSCC and Scopus networks, institutions centered around Utrecht, Netherlands, formed the largest collaborative networks. WOSCC data showed KNAW (Hubrecht Institute/Royal Netherlands Academy of Arts and Sciences), Utrecht University, and UMC Utrecht forming the densest core. This centrality was further validated in Scopus data, with Oncode being another central node. Additionally, the German Cancer Research Center (DKFZ) was another key node in Europe.

3.5 Citation analysis

The top high-impact publications on colorectal cancer organoids, ranked by total citations and journal impact factor, are shown in Figure 6A. These highly cited papers exhibit total citations ranging from 600 to over 3,000, with journal IFs primarily concentrated between 20 and 60. The top 20 most-cited papers in colorectal cancer organoid research, covering basic publication information, DOI, global total citations, normalized citation counts, and journal impact factors, were summarized in Table 5. Despite not having the highest total citations, some papers in high-ranking journals exhibit excellent normalized citation counts and IF, underscoring their significant academic impact in the field.

Citation and co-citation analysis. (A) Top papers by total global citations and impact factor. (B) Top 30 references with the strongest citation bursts. (C) Journal citation network