Global research landscape and trends of papillary thyroid cancer therapy: a bibliometric analysis

A database was built to retrieve related literature for bibliometric analysis via the Web of Science (WoS) Core Collection database. To prevent errors resulting from database upgrades, data searches and exports were conducted on May 1, 2023, encompassing articles related to PTC therapy published between January 1, 2012, and December 31, 2022. The search strategy was designed as follows: TS= (“papillary thyroid carcinoma” or “thyroid papillary carcinoma” or “thyroid carcinoma, papillary” or “papillary thyroid cancer” or “thyroid papillary cancer” or “thyroid cancer, papillary”) AND TS= (therapy OR therapies OR treatment). We initially retrieved 5503 records using this search strategy. To minimize potential bias in our analysis, the following refining criteria were applied: (1) timespan: January 1, 2012, to December 31, 2022; (2) document types included article and review; and (3) written in English. Consequently, a total of 3954 articles were obtained for in-depth analysis. Two authors independently extracted the data from the included studies. The flowchart of article inclusion is shown in. 1

Microsoft Office Excel 2019 software (Microsoft, Redmond, WA, USA) was used to manage data and analyze annual publications. GraphPad Prism 9 software (Dotmatics, San Diego, CA, USA) was used to generate histograms and bubble diagrams. The “bibliometrix” package of R (v4.1.1) is an open-source tool for performing comprehensive science mapping analysis (13). Additionally, the bibliometric analysis utilized two different software tools: CiteSpace [version 6.1. R2 (64-bit)] and VOSviewer (version 1.6.17). CiteSpace is a JAVA-based citation visualization software developed by Chaomei Chen that performs statistical analysis and converts the raw data into a visualization and analysis of literature networks (14). Furthermore, CiteSpace was employed to analyze keywords and references that exhibited significant citation bursts, thus enabling the creation of visualization maps depicting cocited references and keywords. VOSviewer is a powerful software tool for bibliometric mapping developed by Nees Jan van Eck and Ludo Waltman in 2009 (15). It was used to visualize the citation, collaboration, and co-occurrence relationships among journals, references, countries, and institutions. To achieve a comprehensive overview of the research field and ensure result validation, we employed a combination of both software tools. The bibliometric analysis results were subsequently analyzed with descriptive statistics, presented using tables and graphs. To visualize the number of publications in different countries and regions, we utilized Tableau v10.5.0 to create a world map.

The annual trend publications associated with therapy for PTC from 2012 to 2022 are presented in Figure 1. This study identified a total of 3954 publications meeting the inclusion criteria, including 3468 articles and 486 reviews. The continuous increase in the number of publications over the past decade reflects the increase in attention toward PTC therapy research. The quantity of published documents serves as a crucial indicator of the pace of knowledge updates in this subject and helps to elucidate the trends in the field. When evaluating the number of publications per year, the year with the highest number of publications was 2021 (527, 13.3%). The bibliometric analysis revealed a consistent linear growth trend (R^{2 =} 0.8974) in research related to PTC therapy, indicating a growing interest in this field of study (Supplementary Figure S2).

A total of 87 countries or regions contributed to publications in the field of PTC treatment. Figure 2A visualizes the world map of publications in this field. The figure displays the top 25 countries geographically, and the minimum number of documents from a country was 13. Significant achievements were observed in North America, East Asia, and Western Europe, with Figure 2B demonstrating strong cooperation among these countries. A coauthorship network was constructed by including 43 countries/regions with a threshold of ten documents, resulting in six clustered groups. Each node in the network represented a country or region, with node size corresponding to the number of publications. The lines connecting the nodes represent cooperation between countries. In addition, Table 1 and Figure 2C present the scientific production of countries.

Among the top five countries in terms of productivity, China led with the highest number of papers published (1329, 33.59%), followed by the United States (800, 20.23%), South Korea (364, 9.21%), Italy (344, 8.70%), and Japan (170, 4.30%). Although China ranked first globally in terms of publication volume, its citation/publication rate (11.59) was comparatively low compared to other countries. This suggests that the quality of China’s papers needs further improvement.

The present study involved a total of 3426 institutions, with the top 10 institutions contributing 17.15% of the total papers. Table 2 lists these top 10 prolific institutions, with half of them located in China and the remaining institutions located in the United States, South Korea, and Italy. Shanghai Jiao Tong University made the highest contribution, with 93 papers and 1396 citations, followed by China Medical University (75 papers, 1161 citations) and Fudan University (74 papers, 947 citations). Among these top ten institutions, the University of Pisa exhibited the highest citation/publication rate (56.63). An analysis of institutional cooperation was conducted to reveal the collaborations between institutions (Figure 3). Institutions depicted in the same color demonstrated more active cooperation during the survey. Shanghai Jiao Tong University, University of Pisa, and Memorial Sloan Kettering Cancer Center emerged as central partners. However, there was a notable lack of collaboration among these top research institutions within each region.

The 3954 included papers were published in 860 academic journals over the past decade. Table 3 presents the top 10 productive journals, offering valuable insights into the most prolific and highly cited publications. The bibliographic coupling network of journals related to PTC treatment is shown in Figure 4A. Among the 88 journals that published a minimum of 10 papers, Thyroid published the highest number of papers (186), followed by Frontiers in Endocrinology (87 papers) and Journal of Clinical Endocrinology and Metabolism (J CLIN ENDOCR METAB) (81 papers). The impact factor, quartile, and categories were retrieved from the Journal Citation Reports (JCR). Most of the listed journals were categorized under “endocrinology and metabolism” and “surgery.” Except Oncotarget, which was delisted from WoS in 2017, five of the top ten journals were located in JCR quartile one, indicating their high quality. Furthermore, Figure 4B shows the results of cocitation analysis of 141 journals with a minimum citation count exceeding 200. Among the top 10 cocited journals, half of them were cited more than 1000 times. Thyroid had the highest total number of citations (6034, IF=6.506), followed by Journal of Clinical Endocrinology and Metabolism (3404, IF=6.134) and Surgery (1307, IF=4.348). The significantly higher citation count of Thyroid compared to other journals indicates its substantial productivity and influence.

A total of 18,501 researchers and 48,186 cocited authors made contributions to the 3,954 publications in the field of PTC therapy between 2012 and 2022. Figure 5A illustrates coauthor collaboration and the publication output of each author. Collaboration among authors signifies teamwork, although there is less collaboration among teams, resulting in relatively fragmented research. Cocited authors refer to two or more authors who are simultaneously cited through another paper. As depicted in Figure 5B, the cocitation analysis revealed that Ito Yasuhiro (1,286), Xing MZ (1,093), and Haugen BR (1,145) had the highest number of cocitations. These findings indicate that the aforementioned authors displayed greater interest in PTC-related research.

Figure 6 displays the publication count and H-index of the top 10 productive authors. These ten authors collectively contributed 225 papers (5.69%). Notably, Miyauchi Akira (33), Ito Yasuhiro (29), and Tuttle R Michael (25) emerged as the top three authors in terms of publication volume. The first two authors are affiliated with Kuma Hospital in Japan, an institution specializing in thyroid disorders. Among all authors, Antonelli Alessandro from the University of Pisa had the highest H-index (71), followed by Miyauchi Akira from Kuma Hospital (69) and Fallahi Poupak from the University of Pisa (61).

Among the 1,690 documents retrieved for this study, the top 10 most cited research articles accounted for a total of 2,012 citations, representing 19.75% of the total citations. Table 4 presents the top 10 references with the highest cocitations in descending order. Each of these references received more than 90 cocitations. Eight of the publications were from the United States, while the remaining two were from England. The most cited article, titled “2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer,” was published in 2016 in the journal Thyroid, which is the most cited journal in this field.

CiteSpace software was employed to conduct burst analysis and cluster analysis of cocited references. In bibliometric analysis, reference bursts can indicate research hotspots in academic fields. Figure 7A displays the top 15 cocited references with the strongest citation bursts from 2012 to 2022. The strength of these bursts ranged from 16.32 to 118.56. The blue line segment represents the time interval, while the red line segment indicates the time of frequent citations. The top three references with the most pronounced citation bursts were “Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer” (Strength: 118.56; Publication Year: 2009), “2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer” (Strength: 59.16; Publication Year: 2015) and “Integrated genomic characterization of papillary thyroid carcinoma” (Strength: 40.99; Publication Year: 2016).

As depicted in Figure 7B, the research field’s topic clustering map identified 10 clusters with significant modularity and silhouette scores (Q = 0.7983; S = 0.9214). These two indicators are crucial for evaluating the effectiveness of graph visualization. Q-values greater than 0.3 and S-values greater than 0.5 indicate a stable and highly convincing clustering structure. The largest cluster, labeled “prophylactic central neck dissection,” was followed by clusters focused on “managing papillary thyroid carcinoma” (cluster #1), “radioactive iodine-refractory differentiated thyroid cancer” (cluster #2), and “active surveillance” (cluster #3). Other important clusters included papillary thyroid carcinoma, radioiodine therapy, Hashimoto’s thyroiditis, dynamic risk stratification, and association management guidelines.

Keywords play a crucial role in accurately depicting an article’s topic and reflecting the research frontiers within a specific field of study. In this study, we collected 8,013 keywords from various authors and utilized the CiteSpace cluster function to construct a visual map clustering commonly cited keyword. Through our analysis, we categorized the research documents into nine clusters, as depicted in Figure 8A. The module’s Q value was determined to be 0.9122, and the average contour S value was 0.5. The cluster nomenclature effectively represents the study frontiers in the field. In this study, the clusters were autogenerated and labeled using the log-likelihood ratio (LLR) algorithm. The largest cluster, labeled #0, was “proliferation,” followed by “central neck dissection” (cluster #1), “thermal ablation” (cluster #2), and “BRAF mutation” (cluster #3). Other significant clusters included papillary thyroid carcinoma, radioiodine therapy, Hashimoto’s thyroiditis, dynamic risk stratification, and association management guidelines.

To further visualize the thematic map of PTC therapy, we examined the locations of keywords, which were represented by their density and centrality. This representation indicates the evolution of themes (16). The upper right quadrant (Q1) represents crucial motor themes, while the upper left quadrant (Q2) portrays well-established, albeit somewhat isolated, themes. The lower left quadrant (Q3) indicates emerging or declining themes, and the lower right quadrant (Q4) represents basic or transversal themes. As shown in Figure 8B, the terms “management” and “recurrence” are located in Q1, demonstrating that these topics are well developed and can structure the research field. The presence of “metastasis” and “mutations” in Q3 is expected, as they represent basic topics in this field. Additionally, the presence of “expression,” “invasion,” and “proliferation” in Q2 implies highly developed internal connections but marginal contributions to advancements in PTC therapy. The timeline viewer (Figure 8C) of keywords enables us to observe the evolution of new hotspots and explore the evolutionary trajectory of this field. The keywords were classified into nine clusters, and their evolution can be roughly divided into three periods: the early stage (2012-2015), middle stage (2016-2020), and current stage (2021-2022). Based on the cluster labels and keywords within the clusters, recent research hotspots include “tumor microenvironment,” “fusion oncogene,” “dynamic risk stratification,” and “pediatric thyroid cancer”. Supplementary Figure S3A presents the most frequently occurring keywords in the field of PTC therapy from 2012 to 2022. The size of each keyword indicates its importance and frequency of use within the field. To obtain clustering results, we utilized the FoamTrees function of Carrot2 with keywords as the data source, resulting in 34 clusters (Supplementary Figure S3B). The larger the area of the foam is, the higher the intensity of research activity. Notable keywords in the clusters include “cells in patients,” “Braf in cancer,” and “Expression in papillary thyroid carcinoma cells.”

This study analyzed 3954 publications from 87 countries on PTC therapy over the past decade. The studies were retrieved from the WoSCC database. Bibliometric and visual analysis methods were employed to examine research trends and hotspots in this field from 2012 to 2022. The number of publications on PTC therapy has been steadily increasing each year, with 2021 having the highest volume of literature, thus indicating the growing significance of this topic. China and the United States have emerged as the leading contributors, surpassing other countries/regions in terms of publication count. The United States is the most cited country, and Canada had the highest average number of citations. However, China exhibited a low citation-to-publication ratio, suggesting a need for higher-quality publications. Furthermore, half of the top 10 institutions with the most published papers were located in China, followed by the United States and South Korea.

Regarding journals and cocitations, the most productive journals were Thyroid (186 papers), Frontiers in Endocrinology (87 papers), and Journal of Clinical Endocrinology and Metabolism (81 papers). Thyroid also ranked first in terms of total citations, indicating the inclusion of a substantial number of high-quality articles. Among authors, the most productive individuals were Miyauchi Akira (33), Ito Yasuhiro (29), and Tuttle R Michael (25). The three authors with the highest H-indexes were Antonelli Alessandro (71), Miyauchi Akira (69), and Fallahi Poupak (61), highlighting the superior quality of their papers and establishing them as leading figures in the field of PTC therapy research. Among the top 10 most cited articles, in addition to the guidelines for diagnosing and managing thyroid nodules and cancers, the integrated genomic characterization (17) and the current trend of thyroid cancer (18) also gained researchers’ attention. These articles are all published in JCR Q1 journals, and the quality of the articles is relatively high.

The analysis and discussion of the aforementioned general information reveal that influential authors and references mostly consist of review articles and clinical guidelines from internationally renowned institutions and journals. By combining co-occurrence, clustering, and burst analysis of keywords and references, we identified continuous and emerging hotspots and research trends in the field of PTC therapy. In the past, research primarily focused on surgery and radioactive iodine therapy. However, with the increasing awareness of people’s health and advancements in medical technology, active surveillance, thermal ablation, and genomic analysis are becoming prominent areas of research.

To our knowledge, this study represents the first bibliometric analysis and visualization of research on therapy for PTC. However, our study does possess inherent limitations associated with bibliometric analysis. First, this study only included literature retrieved from the WoS database, and papers not included in this database were omitted. Although the WoS database is widely recognized, extensively used, and comprehensive in bibliometrics (67), the results obtained reflect the overall trends. Second, our study focused on retrieving literature related to PTC therapy published between 2012 and 2022 to capture the latest research trends and landscape in this field, thereby excluding earlier papers. Last, a certain degree of language bias was present, as our analysis only included English-language publications.