Web of Science (WoS) is one of the most reliable databases with multidisciplinary coverage, including over 12,000 international academic journals, and was selected as the primary database for this study. In addition, WoS can track more comprehensive citations of publications for better citation analyses, especially older citations, but PubMed lack this part of data, and WoS is considered more user-friendly for bibliometric analysis compared to other databases, such as PubMed and Scopus (22). All literature published in the WoS Core Collection (WoSCC) was searched and exported on April 24, 2024. We used the following search strategy: [ALL = (immune therapy OR immunotherapy OR immune OR immunity OR immunotherapies) AND ALL = (cancer* OR tumor* OR carcinoma* OR neoplasms *) AND ALL = (traditional Chinese medicine OR TCM OR Chinese medicine OR Chinese herbal medicine* OR natural ingredients OR natural compounds OR natural products) AND Document Type = (Article AND Review) AND Language = English]. Full records and cited references were extracted from the retrieved publications and saved in plain text format for further analysis. Subsequently, the two authors(YL and JS) excluded articles that was not relevant to the topic and keywords in the “Address” field. Divergent viewpoints would be resolved through discussion or a third party (LY and WY).

This study used the Bibliometrix package in R software (version 4.4.0), VOSviewer (version 1.6.17), and CiteSpace (version 6.3. R2 Advanced) for bibliometric analysis and the creation of scientific knowledge maps (23–25).

The Bibliometrix R package was mainly used for analyzing annual production, country-wise production, authors’ contributions over time, local impacts of sources based on the H index, and trending topics ( Figure 1 ). VOSviewer software was used to perform co-citation and co-occurrence analyses, with co-citation analysis focusing mainly on cited journals and references. CiteSpace software was used to analyze citation bursts in references and keywords, visualize the cooperation network of countries and institutions, and perform cluster analysis of keywords. The relative parameter settings of VOSviwer main about the minimum number of documents/citations/times. In details,1) for co-authorship analysis, the minimum number of documents of an author was defined as more than 5; 2) for co-citation analysis, the minimum number of citations of a source and cited reference were defined as more than 30 and 15 respectively; 3) for co-occurrence analysis, the minimum number of occurrences of a keyword was defined as more than 15. The parameters of Citespace were set as follows: Timespan:1994-2024, years per slice:1, selection criteria (g-index:K = 25), link retaining factor (LRF = 3), look back years (LBY = 5), e for top N(e = 1), and minimum duration (MD = 2 for keywords; MD = 2 for references).

Based on the search strategy, 4,005 publications were collected from WoSCC without duplicates. Finally, 1,657 publications were identified after excluding those unrelated to the topic. The trend in global publications on TCM for cancer immunotherapy has increased steadily each year ( Figure 2 ). From 1994 to 2018, the number of relevant publications increased gradually. From 2018 to 2023, the number of publications increased rapidly, reaching 341 in 2023.

The national publication output analysis showed that 64 countries/regions published articles in relevant fields. The distribution of publications by country is shown in Figure 3A . China (n = 1,405) was the most productive country, accounting for 84.8% of the total publications, followed by the United States (n = 64, 3.9%), the Republic of Korea (n = 47, 2.8%), India (n = 16, 1.0%), Australia (n = 13, 0.8%), and Japan (n = 13, 0.8%) ( Figure 3B , Table 1 ). The results demonstrated that compared to other countries and regions, China had a much larger number of articles in this field, most of which were SCP, accounting for 91.2% of the total publications from China. Interestingly, Australia had the highest MCP ratio (0.769) compared to China (0.088).

Figure 3C shows a further analysis of the collaboration among countries and regions. The most frequent collaboration was between China and the United States (frequency = 70), followed by China and Australia (frequency = 16), China and the United Kingdom (frequency = 14), and China and Japan (frequency = 13). Among the top 10 collaborations, except for two relationships between the United States and India, and the United States and Italy, all international collaborations involved China.

A global collaboration network analysis was also conducted using CiteSpace. China exhibited the highest volume output and worked closely with other countries, indicating its strongest international collaborations ( Figure 3D ). The size of the circle represents the collaboration strength, whereas the color outside the circle represents the collaboration distribution. The four countries with the highest degrees of centrality were China (centrality = 0.69), the United States (centrality = 0.29), India (centrality = 0.26), and Brazil (centrality = 0.23), indicating that these four countries held leading positions in this field.

A total of 1356 institutions conducted research on TCM in cancer immunotherapy. The top 25 most productive institutions were all from China ( Figure 4A ). Nanjing University of Chinese Medicine had the most publications (n = 107), followed by Shanghai University of Traditional Chinese Medicine (n = 99) and Beijing University of Chinese Medicine (n = 75). To further investigate collaborations between institutions, we conducted a co-authorship analysis using CiteSpace ( Figure 4B ). Interestingly, the Chinese Academy of Sciences published 62 papers, ranking 7th in terms of the number of articles, but had the strongest influence in this research field (centrality = 0.22). The Shanghai University of Traditional Chinese Medicine ranked second (centrality = 0.10). These findings suggest that Chinese medical institutions have published the most articles in this research area. Furthermore, the collaboration between these institutions should be strengthened.

The most productive authors were Wang Wei and Li Yan, each with 15 publications, while Wang Yitao was the most-cited author, with 688 citations ( Table 2 ). Figure 5A shows the authors’ production over time. The size of the circle represented the number of publications, while the color represented the total number of citations per year. Notably, Wang Yitao had the highest number of citations in 2019, despite having only two publications. Figure 5B illustrates the co-authorship clustering network among researchers, revealing their collaborative relationships. Fifty-four authors were classified into nine clusters centered around the most productive authors ( Table 2 ).

This study included 1,657 articles published in 464 journals. Figure 6A and Table 3 list the top 10 journals ranked by publication quantity and their latest 2023 impact factors (IF) (26). These include Frontiers in Pharmacology (n = 100, IF = 4.4), Journal of Ethnopharmacology (n = 91, IF = 4.8), Evidence-Based Complementary and Alternative Medicine (n = 61, IF = 2.65), American Journal of Chinese Medicine (n = 51, IF = 4.8), and Phytomedicine (n = 45, IF = 6.7). We used R-Bibliometrix to measure journal impact based on the H-index. The Journal of Ethnopharmacology had the highest H-index ( Figure 6B ) (27). Notably, the limited presence of publications in top-tier journals on the achievements of TCM in cancer immunotherapy suggests the need to improve the quality of research in this field.

The source titles of the co-citation analysis were analyzed using VOSviewer, and journals with at least 30 citations were included. Five hundred eighty-one journals were identified based on their total link strength ( Figure 6C ). The top five journals with the highest total link strength were the Journal of Ethnopharmacology (total link strength = 123818), International Journal of Molecular Sciences (total link strength = 118496), PLOS One (total link strength = 110081), Biomedicine & Pharmacotherapy (total link strength = 108686), and Oncotarget (total link strength = 101613).

The top 10 most-cited publications were listed in Table 4 . The publication titled “Resveratrol as an Anti-Inflammatory and Anti-Aging Agent: Mechanisms and Clinical Implications” received 558 citations, followed by “The Scientific Rediscovery of an Ancient Chinese Herbal Medicine: Cordyceps sinensis Part I”, with 383 citations, and “Advances in Saponin-Based Adjuvants,” with 324 citations. Highly cited papers were indicators of the Essential Science Indicators (ESI) database used to explore the research frontier (27). Among the 10 publications, four were highly cited papers that provide significant guidance in the field.

Co-cited references were analyzed using VOSviewer to identify the most influential publications ( Figure 7A ). Additionally, CiteSpace was used to identify the most significant citation bursts of TCM in cancer immunotherapy. This analysis yielded 42 references with the strongest citation bursts, 25 of which were presented in Figure 7B . Among them, “Global Cancer Statistics 2018: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries” (strength: 40.74), “Cancer Statistics in China, 2015” (strength: 17.83), and “The Advantages of Using Traditional Chinese Medicine as an Adjunctive Therapy in the Whole Course of Cancer Treatment Instead of Only Terminal Stage of Cancer” (strength: 8.65) were the top three references with the most influential citation bursts.

The keywords in the publications accurately reflected active research topics. To reveal current research trends in TCM in cancer immunotherapy, 258 keywords were gathered from publications, and a network map was created to visualize keyword clusters. Accordingly, “tumor-associated macrophages” (Cluster 0), “inflammation” (Cluster 1), “pharmacological effects” (Cluster 2), “network pharmacology” (Cluster 3), “innate immunity” (Cluster 4), “meta-analysis” (Cluster 5), “colorectal cancer” (Cluster 6), “pyroptosis” (Cluster 7), “research progress” (Cluster 8), “Chinese medicine” (Cluster 9), “breast cancer”(Cluster 10), “cancer treatment” (Cluster 11), and “pluronic F127” (Cluster 12) have been research hotspots since 1994 ( Figure 8A ).

Identifying ‘burst keywords’ could help predict future study trends and emerging areas of interest. The CiteSpace algorithm was also used to detect keyword bursts based on burst detection analysis. As presented in Figure 8B , the keywords ‘mice’ (burst duration from 1998 to 2018), ‘cancer patients’ (burst duration from 1999 to 2018), and ‘alternative medicine’ (burst duration from 2020 to 2018) have received the most consistent attention over time. TCM was an important part of complementary alternative medicine. In addition to being an alternative medicine in the form of herbs, herbal extracts, compound preparations, etc., TCM could also be made into nanomedicines, which were involving in the modulation of tumor immunosurveillance to enhance antitumor activity. These results also showed that the development of the novel dosage form of TCM for anti-tumor immunity was a very important research direction. Interestingly, the result was consistent with the keyword clustering results shown in Figure 8A . For instance, “pluronic F127” (Cluster 12) was known as a surfactant used in drug delivery systems, after coating immune adjuvant, nanomedicines were obtained, which maintained high stability during their passage through the gastrointestinal tract, achieving efficient colonic mucus infiltration and specific tumor penetration. In addition, keywords such as ‘molecular docking’ (burst duration from 2022 to 2024), ‘tumor necrosis factor’ (2016-2020), ‘nf-kappa b’ (2016-2018), and ‘immunomodulatory activity’ (2012-2020), have gained traction more recently. As seen from the burst keyword chart ( Figure 8B ), we can get a general idea of how themes are evolving. For instance, the research trend of TCM tumor immunotherapy has changed from observing the changes of immune indexes in patients and model animals at the very beginning to exploring the mechanism of TCM tumor immunotherapy in more and more studies. In recent years, researchers have begun to dig deeper into the pharmacodynamic molecules of anti-tumor immunity. All these lay the foundation for the development of precision targeted drugs. The increasing interest in research on molecular mechanisms and potentially effective components of TCM in cancer immunotherapy has probably opened new research frontiers in this field.

Keyword co-occurrence analysis was a prevalent method for investigating popular research topics and areas, and also played an important role in surveying the developments of scientific research. Keywords, defined as terms used ≥ 15 times in the titles or abstracts of all papers, were included in the study. The 107 keywords identified were divided into four clusters according to their relevance. As shown in Figure 8C , the red cluster was related to the “tumor immune microenvironment,” with keywords such as “tumor-associated macrophage,” “tumor microenvironment,” “epithelial-mesenchymal transition,” and “astragalus polysaccharides,” indicating the pharmacological action of TCM in anti-tumor immunity. Cluster 2, shown in green, focused on “pharmacological response,” with keywords such as “flavonoids,” “pharmacology,” “anti-inflammatory,” and “immunomodulatory,” indicating modes of interaction between the TME and tumor. The blue cluster was related to “clinical research,” focusing on the clinical efficacy of conventional treatments, with keywords such as “biomarkers,” “chemotherapy,” “immune checkpoint,” “cancer immunotherapy,” “PD-L1/PD-1,” and “prognosis.” The yellow cluster was mainly related to the “anti-tumor mechanism of TCM regulating immunity,” with keywords such as “ferroptosis,” “immunogenic cell-death,” “autophagy,” “cell-cycle arrest,” and “mitochondria”. These results showed that the most prominent applications of TCM in cancer immunotherapy include the aforementioned four directions. TCM in cancer immunotherapy can be based on the type of tumor and the patient’s physique to develop a personalized treatment plan to improve the therapeutic effect. However, in general, the study of TCM in tumor immunotherapy is not deep enough, there is a lack of large-scale clinical research evidence, and the evaluation criteria of efficacy are not clear. Although we have made some achievements in the study of pharmacological action and molecular mechanism, the study of side effects, toxicity and modern preparation of TCM is insufficient.

Additionally, these co-occurrence keywords were coded with different colors by VOSviewer based on their average frequency of appearance in all published papers ( Figure 8D ). Blue indicated keywords that appeared earlier, whereas yellow indicated those that appeared later. Early research primarily focused on “anti-tumor”, “immunity”, “ginseng”, “polysaccharides”, “cytokines”, and “mouse-model”. On the contrary, topics related to “PD-1/PD-L1”, “prognosis”, “quercetin”, “molecular docking”, and “macrophage polarization” gained more attention in recent years. These results indicated that the use of TCM in cancer immunotherapy has gradually shifted from in vivo and in vitro models to clinical efficacy. Simultaneously, the research focus has changed from compound TCM preparations or classes of ingredients to specific pharmacodynamic ingredients, with corresponding targets transitioning from cytokines to immune checkpoints.

Our analysis revealed a notable upward trajectory in the publication of research related to TCM in cancer immunotherapy over the past 3 decades, with a marked increase in the last 5 years. This trend suggested a growing interest and focus among researchers in this field. However, the low publication numbers observed in 2024 can be attributed to the timing of the data export, which was conducted in April of that year. Overall, these insights underscored the increasing relevance of TCM in cancer immunotherapy and suggest that it is an area of study warranting further exploration and investment. According to the global publication trend forecast, the number of publications in this field might continue to rise in the future.

Figure 5B showed the clustering network diagram with authors such as Wang Yitao as the center node, indicating that these authors were in a leading position in this field. In addition, combined with the author’s production over time ( Figure 5A ), we found that Wang Yitao’s articles have outstanding advantages in citation. After digging deeply into the research content of Professor Wang Yitao’s team, we found that the team has achieved fruitful results in the research of TCM anti-tumor immunity. It includes the mechanism research (29), the development of new immune active substances of TCM (30), the research of new targeted drug delivery system (31) and the application of multi-omics research methods (32). The team provided us with theoretical and technical guidance from multiple research directions of TCM anti-tumor immunity, and the research methods and research ideas are worthy of our reference.

About the geographic distribution of research outputs, the research indicated that the highest pooled prevalence of the usage of herbal medicine in adults with cancer was in Africa (47%), followed by Asia (30%), America (21%) and European (18%). From an economic viewpoint, low- and middle-income countries (33%) were more likely to have a high prevalence of herbal medicine use by adults with cancer than high-income countries (19%) (33). It may due to socio-economic reasons, many African people have resorted to traditional medicine using natural products to treat diseases, including cancer (34). While this study showed that, except for China and India, the America and Germany have a higher centrality on research collaborations. This is contrary to the prevalence of herbal medicine usage, largely due to the insufficient economic support for scientific research in African countries.

Some studies about cancer with complementary and alternative medicine showed that, Japan (35) and Thailand (34) maintain a positive attitude towards the integrative cancer therapy, this may be related to cultural, national policies and other factors. For example, in 2013, President Xi Jinping launched the Belt and Road Initiative, and the Chinese government has identified TCM as an important health commodity for the global market. TCM was also promoted as cultural power along the Southeast Asian belt and road corridor (36).

In additional to TCM, as a complementary and alternative medicine (CAM) for cancer immunotherapy, there are Ayurvedic medicine, European medieval medicine, Kampo medicine worldwide. The researches about European medieval medicine are much less than other CAM practices (37). Ayurveda was originated and evolved in India and had been practiced for thousands of years, and Curcumin, Ashwagandha, and Triphala were three of the most promising and commonly researched Ayurveda herbs or formulas related to cancer. Ayurvedic medicine foucused on ‘one target-one herb’, and the researchers has realized this approach’s limitation (38). Kampo medicine (Japanese traditional medicine) was originated from ancient China, and had been improved since the Edo period. Most Kampo preparations were available as extract formulations, which was greatly different from China, and only 148 formulations were covered by health insurance system (39). Compared with China, there were 2282 kinds of TCM collected by the 2023 edition of the “China Medical Insurance Directory” (40). The current researches indicated that TCM was used as adjuvant therapy to treat dozens of cancers efficiently, and they exhibited prospective antitumor potential with multiple targets, multiple signaling pathways and less side effects (41, 42). Moreover, the Chinese government also strongly supports the development of TCM through national policies such as the Belt and Road Initiative and the Forum on China-Africa Cooperation (43, 44). Under these policies, China aims to construct a TCM international medical service system, so it will also make the distinctive contributions to the field of cancer immunotherapy.

The first hotspot and trend in TCM for cancer immunotherapy were cancer types. Compound TCM preparations used either alone or as adjunct therapies, have been widely used in cancer treatment (45). Keyword co-occurrence analysis revealed that “lung cancer”, “liver cancer”, “colorectal cancer”, “breast cancer”, “gastric cancer”, “cervical carcinoma” and “melanoma” were the most researched cancer types in this field. For example, Tang demonstrated that Jin-Fu-An decoction targets lung cancer by modulating macrophage polarization via β-catenin synergy (46). Chen et al. found that Dahuang Zhechong pills enhanced anti-tumor immunity in liver cancer by secreting IFN-γ to activate CD8⁺ T cells and promote Treg differentiation (47). Jiang et al. showed that Tong-Xie-Yao-Fang inhibits the growth of colorectal tumors by promoting the maturation of dendritic cells and stimulating T cell-mediated immune responses (48). XIAOPI was confirmed to inhibit breast cancer lung metastasis by inhibiting CXCL1 secretion from tumor-associated macrophages (49). Bu-zhong-yi-qi decoction modulates peripheral immunity and suppresses the immune escape of tumors via PD-1/PD- L1-dependent T cell immunization, showing promise for gastric cancer therapy (50). Additionally, Zhang et al. revealed that flavonoid extracts from P. heterophylla have anti-tumor and immunomodulatory effects in cervical carcinoma-bearing mice by activating more CD4⁺ and CD8⁺ T lymphocytes (51). Furthermore, triptolide significantly inhibited tumor growth by stimulating antigen presentation of dendritic cells (DCs) and the proliferation of cytotoxic T lymphocytes (52). In general, there were fewer studies on tumors of the upper respiratory tract, cervical cancer, and melanoma. It may be related to the relatively low incidence of cancers mentioned above. According to the Global Cancer Statistics 2022, lung cancer is the most commonly diagnosed cancer worldwide (12.4% of the total cases), followed by cancers of the female breast (11.6%), colorectum (9.6%), prostate (7.3%), and stomach (4.9%) (53). In addition, cancers such as melanoma, which have a high degree of malignancy, rapid disease progression, and a short course of disease, may not be very suitable for the use of TCM as an alternative therapy.

The second hotspot and trend in TCM related to cancer immunotherapy was the evolution of research methodologies. Despite limited research and support for the effectiveness of TCM, it has gained significant attention in recent years, prompting more scientists to employ scientific methods to elucidate its fundamental mechanisms, safety, and efficacy. Existing research methods included network pharmacology, bioinformatics, multi-omics analysis, meta-analysis, in vitro and in vivo experiments, and clinical trials (54–58). By combining the overlay visualization map of keyword co-occurrence and the burst chart, we found that molecular docking technology has recently become a widely used method. Since its inception in the mid-1970s, molecular docking has proven to be an essential tool for efficiently understanding the interactions between chemical compounds and their molecular targets, and for drug discovery and development (59). The ambiguity of active ingredients and therapeutic targets remained a limiting factor for the broader clinical application of TCM in cancer immunotherapy. Keyword co-occurrence analysis showed that research on TCM anti-tumor immunity has gradually shifted from in vivo and in vitro experimental studies to the development of clinically targeted drugs. Additionally, research has increasingly expanded from compound TCM preparations to individual pharmacological ingredients (58, 60). Therefore, molecular docking combined with multi-omics analysis was a popular and trending research method for TCM in cancer immunotherapy. For example, techniques of network pharmacology, molecular docking, and experimental validation were applied by Lu to demonstrate the mechanism of BZD for anti-CRC. As a result, quercetin, kaempferol, licochalcone A, naringenin, and formaronetin were more highly predictive components related to the T cell activation in colorectal cancer mice by Molecular docking and experimental validation (54). Gao determined the top 10 key proteins in the upregulated KEGG pathways of PMN-MDSCs in melanoma tumour-bearing mice through proteomics and Cytoscape analysis. Then PMN-MDSC inhibitor prim-O-glucosylcimifugin was screened by molecular docking from the TCM Library. Finally, the effect of the inhibitor was verified through proteomics and metabolomics analysis in melanoma and triple-negative breast cancer mouse tumour models (61). Nevertheless, the results of molecular docking are often accompanied by significant uncertainty, which is easy to produce false positive results. Next, molecular docking usually relies on server resources, which requires high computing resources, and the calculation process is time-consuming. In addition, the technology requires high quality of input data, requiring that the input protein structure needs to undergo strict pretreatment, including structural repair, hydroprotolation, etc., to ensure the accuracy of docking. In summary, although molecular docking technology has some shortcomings, its ability to efficiently predict binding patterns and affinity makes it of great value in the field of drug design and biological research. By combining the experimental data and the optimized calculation method, the accuracy and reliability can be further improved (62).

The third hotspot and trend in TCM for cancer immunotherapy involved molecular mechanisms. Anti-tumor immune responses could be initiated through innate and adaptive immune systems. TCM has also been reported to exert immunoregulatory effects by upregulating immune responses, particularly in the TME. Studies have shown that TCM exerts its immunomodulatory function by regulating the differentiation and cytokine secretion of macrophages, dendritic cells, NK cells, MDSCs, T cells, and B cells, thereby enhancing the ability of immune cells to kill malignant cells or present antigens (48, 63–66). Currently, there were few clinical studies on TCM affecting immune checkpoints. Keyword time-zone map analysis showed that recent studies have focused on the inhibitory effects of TCM on immune checkpoints. Researches showed that TCM could directly reduce PD-1 and PDL-1 expression in tumor cells. In addition, TCM could remodel the gut microbiota to block PD-1 expression in cancer (67, 68).

Fourth, to clarify the timing of TCM use as an immunomodulatory agent, we included reports of cancers and inflammatory diseases highly associated with cancer, such as chronic liver disease and ulcerative colitis. Wang conducted a meta-analysis concluding that patients with stage III/IV non-small cell lung cancer (NSCLC) could safely receive elemene injections alongside platinum-based chemotherapy, which enhanced clinical efficacy, improved cellular immune function, and reduced chemotherapy toxicity (69). Relevant clinical investigations have shown that combining Qing Gong Detoxification Soup with TP (paclitaxel-cisplatin) chemotherapy significantly restored immune homeostasis, inhibited tumor markers, and reduced the incidence of adverse reactions in patients with cervical cancer after surgery (70). Chronic HBV infection could ultimately lead to liver cirrhosis and hepatocellular carcinoma, which were major complications and leading causes of mortality worldwide (2). Reports indicated that curcumin’s antiviral activity, which targeted and modified regulatory T cells into T helper cells, could be enhanced by blue laser photobiomodulation, facilitating the elimination of HBV-infected hepatocytes (71). According to the results of keyword co-occurrence analysis, “inflammatory bowel disease” (IBD) was the most studied inflammatory disease at present. Natural products and herbal medicines have demonstrated efficacy for IBD in experimental models and clinical trials by modulating innate and adaptive immune responses, regulating macrophage activation, and inhibiting TNF-α activity (72). In summary, immunotherapy using TCM could contribute to effective maintenance or adjuvant therapy throughout the whole course of cancer instead of only in the late stages, enabling more patients to live with cancer for longer periods.

We conducted a systematic visualization analysis of the articles and trends in TCM for cancer immunotherapy in an intuitive way for the first time, providing new research directions for scholars in this field. However, our research also had some limitations. For example, due to limitations in the current mainstream bibliometric software, the data from different database cannot be integrated, such as PubMed, CNKI, and patent information in both Chinese and English. Therefore, we will use more Chinese database solely to supply this research and more powerful software are supposed to be used in the future. In addition, the input data was extracted using carefully designed search strings, but some irrelevant articles were still retrieved, and relevant publications with unclear titles might have been missed. Furthermore, the literature analysis software used (CiteSpace, VOSviewer, and Bibliometrix Package) might produce slightly different statistical results due to version discrepancies. As with any statistical endeavor, bibliometric analyses have the potential to generate misleading and biased results. Finally, recently published high-quality articles might not have received attention because of their temporarily low citation counts. Despite these limitations, our study provided insights into the characteristics of research and citations in the field of TCM for cancer immunotherapy.