Global research trends in diagnostic biomarkers for colorectal cancer: a bibliometric and visualization analysis.

Introduction
According to statistics from the International Agency for Research on Cancer, in the global cancer incidence statistics for 2022, colorectal cancer (CRC) accounted for 9.6% of all cancers, ranking it as the third most common cancer worldwide[1]. A survey on the incidence and mortality of gastrointestinal cancers from 2021 also revealed that the incidence of gastrointestinal cancers was 5.26 million, with 3.7 million deaths, the largest proportion of which was attributed to CRC[2]. Current clinical screening methods for early CRC detection primarily involve fecal occult blood tests and colonoscopy[3]. However, since fecal occult blood tests are not entirely reliable for diagnosing CRC, and colonoscopy screening imposes a significant burden on patients, scientists are still committed to finding a simple, convenient, cost-effective, and effective diagnostic biomarker for screening early CRC and assessing the prognosis of CRC.
HIGHLIGHTS
First bibliometric mapping of global research (2014–2024) on diagnostic biomarkers for colorectal cancer (CRC), analyzing 255 publications via CiteSpace/VOSviewer.

Reveals China’s dominance (62.75% publications) and the unexpected leadership of the Egyptian Knowledge Bank in institutional output.

Identifies genes, microRNAs, lncRNAs, and proteins as core biomarker categories, with multi-omics integration as a key emerging trend.

Tracks dynamic hotspot evolution: from gene expression (2014–2017) to lncRNAs (2017–2019), proteins (2019–2020), and DNA methylation (2020–2024).

Provides evidence-based guidance for future research priorities, emphasizing non-invasive sampling and biomarker panels to improve early CRC detection.

At the beginning of the 20th century, bibliometrics emerged as an independent discipline and has been widely used in the study and analysis of literature[4]. Bibliometrics can systematically analyze relevant literature in a particular field, providing a visual reflection of the research countries, institutions, journals, authors, and keywords, which helps to understand the frontier dynamics and development trends of the field. This article conducts a bibliometric analysis of CRC and diagnostic biomarkers, utilizing CiteSpace, VOSviewer, and Excel software to map the scientific knowledge graph of this research, in order to explore the hotspots and global development trends in this field. The research is compliant with the TITAN Guidelines 2025 – governing declaration and use of AI[5]. AI-assisted linguistic refinement and English translation were applied solely to enhance manuscript readability without impacting scientific content. No generative AI tools were utilized in the conceptualization, analysis, or interpretation of this research. AI-assisted translation tools were employed during manuscript preparation, with all authors assuming full responsibility for the accuracy and integrity of the final content.

Method

Databases and search strategy
The work has been reported in line with the STROCSS criteria[5]. The literature for this study was retrieved from the Science Citation Index Expanded database. To ensure the comprehensiveness and accuracy of the retrieved documents, the search scope was limited to the Science Citation Index Expanded, with the search period set from 1 January 2014 to 10 September 2024, the document type was restricted to “Articles,” and the language type was selected as “English.” We searched for topics related to CRC and diagnostic biomarkers, with the search expression shown in the flowchart (Fig. 1). A total of 564 relevant documents were identified, two researchers (Jiawei Gao and Weiwei Cao) read the titles and abstracts of the aforementioned documents. Following a manual screening process,309 irrelevant documents were excluded, leaving a final selection of 255 documents (exclusion criteria are presented in Table 1). These studies were saved as a plain text file on 12 September 2024.

Software for bibliometric and visualization analysis
We employed three software tools – CiteSpace (v.6.3.R1), VOSviewer (1.6.18.0), and Excel (2021) – to construct a knowledge map of the 255 documents. CiteSpace (v.6.3.R1) and VOSviewer (1.6.18.0) were utilized for generating collaboration and clustering networks, while Excel (2021) was used to create pie charts, bar charts, and line graphs.
CiteSpace is a scientific literature analysis tool designed for bibliometric analysis on publications within a specific research field. It primarily utilizes set theory-based data standardization methods to measure the similarity of knowledge units. Similarity algorithms are applied to obtain Timezone and Timeline views within time slices, thereby providing a clear depiction of the knowledge evolution process and the historical span of literature within a particular cluster, and this approach presents the research structure and developmental progress of the field. In this study, CiteSpace (6.3.1) was used to analyze the publication patterns of relevant countries, institutions, authors, publication counts, and keywords related to literature publication, and to visualize these data as images.
VOSviewer is based on probabilistic data normalization method and offers various visualization views for keywords, co-authorship, co-citation, and other aspects, including Network Visualization, Overlay Visualization, and Density Visualization. This study used VOSviewer (1.6.18.0) to investigate keywords and co-cited reference.

Results

Trend in literature publication over the past decade
A total of 255 articles related to CRC and diagnostic biomarkers were identified in this study. As depicted in Fig. 2, the number of publications on CRC and diagnostic biomarkers began to increase in 2015, reaching a peak in 2021 with 37 publications, the lowest number of publications was recorded in 2015, with only 16 articles. The annual publication volume has fluctuated over the past decade, showing a general upward trend followed by a slight decline. Nevertheless, the cumulative number has steadily increased, indicating a significant growth and progress in research on CRC and diagnostic biomarkers.

Publications and collaboration networks of countries/regions, institutions, and authors
We analyzed the publication output and collaboration networks of relevant countries, regions, institutions, and authors involved in CRC and diagnostic biomarker research. In the collaboration network diagram, larger nodes indicate a higher number of publications. A higher centrality value indicates a stronger collaborative relationship between the node and other nodes.

Analysis of publications and collaborative relationships by country/region
We conducted an analysis of the publication output (Fig. 3A, Fig. 3B) and collaborative networks (Fig. 3C) of CRC and diagnostic biomarker research across different countries and regions. The results presented in Fig. 3A and 3B indicate that China, Egypt, Iran, and the United States are the top four countries in terms of publication volume, and they also occupy central positions in the collaboration network. China ranks first with 160 publications, followed by Egypt (20), Iran (16), and the United States (13). The research indicates that these four countries are the main forces in CRC and diagnostic biomarker-related studies, while other countries/regions have published fewer than 10 papers. Furthermore, the network and density depicted in Fig. 3C demonstrate the research impact of countries such as China, Egypt, Iran, and the United States, the largest connected component in the country/region co-occurrence network consists of 37 nodes and 45 connections (density = 0.0676), as shown in Fig. 3C, China has the greatest research influence.

Analysis of institutional publications and collaborative relationships
Figure 4A and B illustrates the publication counts from different institutions, while Fig. 4C presents the centrality and collaborative network of these institutions. According to the results shown in Fig. 4A and Fig. 4B, the majority of publications were contributed by three research institutions: the Egyptian Knowledge Bank, Cairo University, and Nanjing Medical University. The Egyptian Knowledge Bank ranks first with 20 publications, followed by Cairo University (15) and Nanjing Medical University (13), with other institutions publishing fewer than 10 articles. Additionally, the network and density shown in Fig. 4C highlight the research impact of institutions such as the Egyptian Knowledge Bank, Cairo University, Nanjing Medical University, and Sun Yat-sen University, the largest connected component in the institutional co-occurrence network consists of 220 nodes and 356 connections (density = 0.0148). As indicated in Fig. 4C, some institutions independently conducted research related to CRC and diagnostic biomarkers, while others completed relevant studies through inter-institutional collaboration.

Journals
Research related to CRC and diagnostic biomarkers has been published in 143 different journals. Table 2 presents the top 10 journals with the highest publication volume along with their 2023 Impact Factors (IFs). According to Table 2, Oncology Letters has the highest publication volume, ranking first, with an IF of 2.5 for the year 2023, and is classified in JCR as Q3, and this journal primarily publishes research related to clinical oncology. The second and third most prolific journals are Frontiers in Oncology (IF: 3.5, JCR: Q2) and PLoS One (IF: 2.9,JCR: Q1). Among the top 10 journals, three are classified in JCR as Q1,five as Q2, and two as Q3,with six journals having an IF greater than 3.

Author
Figure 5 illustrates the collaborative relationships among authors in CRC and diagnostic biomarker-related publications. It can be observed that the network and density depicted in Figure 5 demonstrate the extent of publication contributions by different authors. The largest connected component within the co-authorship network comprises 294 nodes and 709 edges (density = 0.0165). As indicated in Figure 5, most authors have engaged in research on CRC and diagnostic biomarkers through mutual collaboration. However, the cohesiveness of collaboration among different groups of authors is not high, and the majority of authors have published a relatively small number of papers.

Keyword co-occurrence analysis
Keywords reflect the main content of articles, with higher frequency indicating greater attention. This can be used to explore the research hotspots in CRC and diagnostic biomarkers. We utilized VOSviewer software to construct a keyword co-occurrence knowledge map, yielding 1209 keywords, with a threshold set at six occurrences, displaying a total of 78 keywords. The resulting keyword co-occurrence map is depicted in Fig. 6. In this map, keywords with the same color belong to the same cluster, the size of the nodes indicates the frequency of keyword occurrence, and the distance between nodes represents the degree of closeness between keywords.

Keyword clustering analysis
Building upon the keyword co-occurrence analysis, we performed a clustering analysis of the keywords using CiteSpace software. Each cluster was required to contain at least 25 nodes, resulting in a total of 13 clusters (Fig. 7). The clustering map consists of 299 nodes and 620 connections. The modularity (Q-value) and silhouette value (S value) of the network are crucial indicators reflecting the effectiveness and rationality of the overall network structure. The Q-value for clustering analysis in this field is 0.7514 (>0.3), indicating that the clustering is effective; the S value is 0.8905 (>0.5), suggesting that the homogeneity within the clusters is high, and thus the clustering results are reasonable. Based on the clustering results, the main research content of diagnostic biomarkers for CRC includes #0 tumor suppressor, #1 microRNA, #2 colorectal carcinoma, #3 human colon, #4 colorectal neoplasms, #5 risk, #6 bioinformatics analysis, #7 early detection, #8 diagnostic biomarker, #9 DNA, #10 biomarker screening, #11 rectal cancer, and #12 early diagnosis. Additionally, based on the clustering results, we have created a timeline(Fig. 8). This map uses the cluster labels as the Y-axis and the years of keyword occurrence as the X-axis, providing a clear visual representation of the time span for each cluster and the interrelationships between clusters. According to the display of the clustering timeline map, it can be observed that cluster labels such as microRNA, colorectal carcinoma, bioinformatics analysis, early detection, diagnostic biomarker, DNA, and diagnostic biomarker screening persist throughout the period from 2014 to 2024.

Keyword emergence analysis
Keyword emergence analysis highlights keywords whose frequency increases abruptly within a short period. The larger the value, the more active the keyword is in the field, providing a better focus on the cutting-edge research (Fig. 9). Based on the diagram, we can roughly deduce the evolving trends in research hotspots for CRC diagnostic biomarkers over the past decade. Genes, long noncoding RNA, proteins, and DNA methylation are among the hot topics that scientists have been studying in the context of CRC diagnostic biomarkers, but the emergence of these keywords occurs in different time periods. The chart can be roughly divided into four segments. From 2014 to 2017, gene expression was the primary research hotspot; from 2017 to 2019, long noncoding RNA emerged as a key focus in the literature; from 2019 to 2020, protein research gained prominence; and from 2020 to 2022, research shifted back toward genetic-related topics.

Reference citation analysis
We employed (used) VOSviewer to generate a knowledge map of co-cited references. Among the 255 articles related to CRC and diagnostic biomarkers, a total of 9038 references were cited with the minimum citation frequency being six times and a total of 53 references were depicted in the graph. Figure 10 illustrates the final relationships of co-cited references, where the network diagram shows that highly co-cited references are divided into five clusters, each represented by a different color: red, yellow, green, blue, and purple. The top five most-cited articles are as follows:
“Global cancer statistics,” published in CA: A Cancer Journal for Clinicians in 2011[6].

“Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method,” published in Methods in 2001[7].

“Global patterns and trends in CRC incidence and mortality,” published in Gut in 2017[8].

“Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012,” published in International Journal of Cancer in 2015[9].

“Circulating exosomal microRNAs as biomarkers of colon cancer,” published in PLoS One in 2014[10].

Discussion

Trends and necessity of literature research
CRC, as a gastrointestinal malignancy with a high mortality rate, emphasizes the importance of early screening. Although colonoscopy remains the most reliable method for diagnosing CRC, its repetitive and invasive nature limits its widespread application as a screening tool[11]. Furthermore, carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) are the two most commonly used biomarkers for diagnosing CRC; however, their sensitivity during treatment is still insufficient, making them unsuitable for dynamic monitoring of CRC patients[12,13]. Therefore, further identification of CRC diagnostic biomarkers would greatly contribute to improving CRC screening rates and reducing the mortality of gastrointestinal malignancies. Currently, the exploration of CRC diagnostic biomarkers is flourishing, with scientists increasingly favoring easily obtainable specimens such as plasma and feces, to facilitate early CRC screening in clinical settings[14,15]. This study is the first bibliometric analysis of CRC and diagnostic biomarkers, examining related research from 2014 to 2024. By analyzing relevant research from 2014 to 2024 using bibliometric methods, it reflects the publication trends, current status, and hotspot topics in this field over the past decade, providing researchers with a systematic and intuitive overview of the research landscape.
This study includes 255 papers from 1894 authors affiliated with 460 institutions across 37 countries, which were published in 143 journals and cited 9038 references from 1562 journals. In the past decade of research on CRC and diagnostic biomarkers, publications in 2015 were relatively few. However, the number of published papers began to gradually increase thereafter, with a significant surge in publications between 2018 and 2019. The number peaked in 2021 and then gradually declined, yet the overall publication rate remains high. This indicates that the field of CRC and diagnostic biomarkers has garnered increasing interest from scientists, with progressive deepening in this field and the achievement of certain research outcomes.

Geographical distribution of publications
In the realm of research related to CRC and diagnostic biomarkers, China has published the highest number of articles, with a total of 160 papers accounting for 62.75% of the overall output, thereby occupying a leading position in the field. According to the latest data from the Journal of the National Cancer Center (JNCC), in 2022, China reported 517 100 new cases of CRC and 240 000 deaths[16], making it (CRC) the second most common malignant tumor in the country. Additionally, the average 5-year survival rate for CRC in China is 56.9%[17], with one of the main reasons being the relatively low rate of early diagnosis among patients. The steadily increasing number of new cases and the suboptimal 5-year survival rate may be one of the main reasons why China is focusing on research into CRC and diagnostic biomarkers. Countries such as Egypt, Iran, and the United States have also made contributions to related research. However, despite certain achievements in their respective studies, these countries appear not to have engaged in international collaboration. Broader international cooperation could enhance the quality of research in this field.

Affiliated institutions of publications
Among the top 10 institutions with the highest number of publications on CRC and diagnostic biomarkers research, seven are from China, two from Egypt, and one from Iran, which is consistent with the distribution of publications across countries. The top three institutions by publication count are the Egyptian Knowledge Bank, Cairo University, and Nanjing Medical University. This indicates that although China leads in publication output, the research institutions are relatively dispersed and independent, with less collaboration. This could be related to the shared interests of researchers within these institutions and the funding they receive. Two of the top three institutions with the highest publication output are from Egypt, despite Egypt being the second-largest contributor in terms of publication volume, significantly lower than China. However, the Egyptian Knowledge Bank is the leading institution in this study, indicating its specialization in CRC and diagnostic biomarkers research. It plays a crucial role in driving research in this field.

Journal affiliation of publications
Peer review of journals is crucial to the publication of scholarly articles, and analyzing the journals in which publications can help researchers submit their work to appropriate fields. Among the top 10 journals publishing research related to CRC and diagnostic biomarkers, Oncology Letters has the highest number of publications, with nine publications and an IF of 2.9. Bibliometric analysis indicates that journals publishing research on CRC and diagnostic biomarkers are predominantly classified as Q1 or Q2, which suggests that the quality of research in this field is relatively high, and the results of bibliometric analysis are therefore more reliable.

Research foundation of CRC diagnostic biomarkers
Co-citation refers to the references that are commonly cited by researchers in their articles. Conducting a co-citation analysis of the literature helps identify the foundational research between CRC and diagnostic biomarkers, aiding in the understanding of the research background. We retrieved the top five most highly cited articles, three of which addressed the global epidemiology of cancer or CRC at the time, one focused on the 2(-Delta Delta C(T)) and PCR experimental techniques, and the other investigated the sensitivity of exosomal microRNAs as diagnostic biomarkers for colon cancer. According to the highly co-cited literature, epidemiological studies from 2008 and 2012 indicated that CRC is the third most common malignant tumor in men and the second in women globally[6,9]. Furthermore, the global incidence and mortality rates of CRC exhibit a 10-fold difference, with rapidly rising incidence and mortality rates in low-and middle-income countries, while trends in developed countries show either stability or a decline[8].
Against the backdrop of CRC’s high incidence and mortality rates, scientists have been dedicated to mitigating the impact of this disease. Polymerase chain reaction (PCR) technology, a method for rapidly amplifying specific DNA fragments in vitro, fundamentally simulating the semi-conservative DNA replication process within cells, invented by American scientist Kary Mullis[18]. With the continuous improvement and refinement of PCR technology[7], it has been widely applied in medical fields including oncological diseases[19], infectious diseases[20,21], and cardiovascular diseases[22]. The high incidence and mortality rates of CRC have led researchers to focus on the disease, and PCR technology has provided technical support for the study of CRC and diagnostic biomarkers to some extent. Researchers have utilized this technology to detect differential substances between CRC patients and healthy individuals, aiming to identify non-invasive diagnostic biomarkers for CRC. Exosomal microRNA represents a fundamental and prominent area of focus, and bibliometric results also indicate that microRNA as a diagnostic biomarker for CRC is one of the main directions of interest for researchers[10].

Identifying hotspots and frontier knowledge through keyword co-occurrence, clustering, and emergence analysis
Keywords summarize the core content of literature, and by conducting co-occurrence, clustering, and emergence analysis of keywords, we can gain insights into the distribution and development of research hotspots in a specific field[23]. The keyword clustering analysis ultimately identified 12 potential research directions, including “tumor suppressor,” “microrna,” “colorectal carcinoma,” “human colon,” “colorectal neoplasms,” “risk,” “bioinformatics analysis,” “early detection,” “diagnostic biomarker,” “dna,” “biomarker screening,” “rectal cancer,” and “early diagnosis.” Additionally, the most frequently mentioned keywords include “colorectal cancer,” “expression,” “biomarker,” “metastasis,” “diagnosis,” “proliferation,” “prognosis,” “invasion,” “cells,” and “microRNAs.” The aforementioned text indicates that research related to CRC focuses on the study of diagnosis and prognosis, with biomarker studies delving into cellular, molecular, and genetic levels.
Following these keywords, we reviewed the progress and current status of research related to CRC and diagnostic biomarkers over the past decade. Combined with keyword emergence analysis, we found that research related to CRC and diagnostic biomarkers involves genomics[24], transcriptomics[25], and proteomics[26] methodologies. By utilizing multi-omics research methods, genes such as APC[27], TP53[28], SPATA18[29], and PIK3CA[30] have been identified at the genetic level. These genes are highly correlated with the inhibition or promotion of CRC development and are considered potential directions for CRC diagnosis research. With the support of transcriptomics technologies, various potential biomarkers for diagnosing CRC have been continuously discovered, among which MicroRNAs stand out as a significant finding. MicroRNAs are a class of endogenous small non-coding RNAs, approximately 22 nucleotides in length[31,32], and since the discovery of miRNAs in Caenorhabditis elegans by Victor Ambros[33] and Gary Ruvkun[34] in the 1990s, over 450 microRNAs have been identified. Furthermore, various experiments have confirmed that CircRNAs[35,36] and lncRNAs[37,38] also have potential as early diagnostic biomarkers for CRC. Additionally, plasma proteins, due to their presence in the circulatory system through secretion or cell leakage, can reflect both health and disease states, and are regarded by researchers as potential biomarkers related to CRC[39,40].

Identify and validate hotspots and cutting-edge knowledge from the latest literature
Recent studies indicate that circRNAs[41] and lncRNAs[42] remain one of the focal points in the research of CRC diagnostic biomarkers, confirming that bibliometric analysis of CRC and diagnostic biomarkers has captured the current research hotspots and frontier topics. Moreover, recent literature combines both plasma and fecal samples and employs metabolomics to identify metabolites that change during the development of CRC and colorectal adenomas. These metabolites have been validated as having potential as metabolic biomarkers[14]. This suggests that building upon past research related to CRC and diagnostic biomarkers, future research utilizing multi-sample collection methods and multi-omics study approaches to investigate combined biomarkers may represent a promising trend for advancing and improving CRC diagnostic biomarkers.

Comparisons with previous studies
This study highlights an emerging focus on the research trends of long non-coding RNAs (lncRNAs) and other novel biomarkers. In contrast to preceding investigations that predominantly centered on conventional markers such as CEA and CA19-9, our work demonstrates greater methodological novelty and intellectual foresight. By delineating evolving research priorities in this field, our findings hold substantial implications for refining diagnostic strategies and therapeutic interventions in oncology.

Strengths and limitations
A primary advantage of this investigation resides in its methodological rigor derived from a large dataset. By comprehensively analyzing literature within the Web of Science Core Collection database, the study achieves robust sample sizes, thereby ensuring the reliability and statistical validity of its conclusions. A multidimensional analytical framework was employed to examine geographic distribution patterns, institutional contributions, journal IFs, authorship networks, and thematic keyword clusters. This approach not only maps the global research landscape but also identifies pivotal institutions, high-impact publication venues, and emerging scientific frontiers, offering a holistic perspective for scholars navigating this field. Furthermore, the study demonstrates considerable innovation by foregrounding lncRNAs and other novel biomarkers as critical research trajectories, thereby establishing clear priorities for future investigative efforts and fostering conceptual advancements within CRC diagnostics.
Several constraints warrant acknowledgment. The analysis was restricted to English-language publications, potentially excluding significant contributions from non-English speaking regions where unique research priorities or methodological approaches may exist. While numerous promising biomarkers and research foci were identified, the study does not systematically evaluate their clinical translational potential or validation status – a critical gap given the ultimate goal of improving diagnostic accuracy and therapeutic outcomes. Additionally, although policy implications are acknowledged in principle, the manuscript lacks concrete recommendations for translating these findings into actionable clinical guidelines or public health strategies. Future iterations of this work could fruitfully explore mechanisms to bridge the research-to-practice gap, particularly through policy engagement initiatives that accelerate the clinical implementation of validated diagnostic biomarkers.

Clinical implications for clinicians and policy-makers
For clinicians, this study elucidates emerging research frontiers in diagnostic biomarkers for CRC, with a particular focus on genetic markers, proteins, and lncRNAs. These findings offer critical guidance for future research priorities and hold significant potential to advance precision medicine approaches in CRC diagnostics. Notably, the identified biomarkers could serve as valuable adjunctive diagnostic tools, enabling clinicians to achieve more accurate detection of CRC, particularly in early-stage disease, thereby enhancing patient survival outcomes and quality of life.
For policymakers, the documented research trends and identified knowledge gaps provide a strategic framework for resource allocation. Specifically, targeted investment in emerging areas such as lncRNA research could accelerate innovation and foster breakthroughs in this critical field. Furthermore, by mapping the global research landscape in CRC biomarkers, this study offers policymakers a platform to stimulate international scientific collaboration, facilitate knowledge exchange, and optimize resource utilization across institutional and national boundaries. Leveraging these insights, policymakers could develop evidence-based initiatives to promote the clinical translation of validated biomarkers, ultimately improving early detection rates, therapeutic efficacy, and disease burden reduction in CRC management.

Conclusion
This study analyzed 255 publications on CRC diagnostic biomarkers, revealing key trends and research gaps. China emerged as the leading contributor, with studies focusing on genetic markers, lncRNAs, and epigenetic factors. While highlighting promising biomarkers and international collaboration opportunities, the analysis is limited by English-language bias and lack of clinical validation data. Future research should prioritize multi-center validation studies and policy frameworks to accelerate translational applications.