Knowledge mapping of immunotherapy for breast cancer: A bibliometric analysis from 2013 to 2022

ABSTRACT

Breast cancer is the leading cause of cancer-related death among women globally. Immunotherapy has emerged as a major milestone in contemporary oncology. This study aims to conduct a bibliometric analysis in the field of immunotherapy for breast cancer, providing a comprehensive overview of the current research status, identifying trends and hotspots in research topics. We searched and retrieved data from the Web of Science Core Collection, and performed a bibliometric analysis of publications on immunotherapy for breast cancer from 2013 to 2022. Current status and hotspots were evaluated by co-occurrence analysis using VOSviewer. Evolution and bursts of knowledge base were assessed by co-citation analysis using CiteSpace. Thematic evolution by bibliometrix package was used to discover keywords trends. The attribution and collaboration of countries/regions, institutions and authors were also explored. A total of 7,975 publications were included. In co-occurrence analysis of keywords, 6 major clusters were revealed: tumor microenvironment, prognosis biomarker, immune checkpoints, novel drug delivery methods, immune cells and therapeutic approaches. The top three most frequently mentioned keywords were tumor microenvironment, triple-negative breast cancer, and programmed cell death ligand 1. The most productive country, institution and author were the USA (2926 publications), the University of Texas MD Anderson Cancer Center (219 publications), and Sherene Loi (28 publications), respectively. There has been a rapid growth in studies on immunotherapy for breast cancer worldwide. This research area has gained increasing attention from different countries and institutions. With the rising incidence of breast cancer, immunotherapy represents a research field of significant clinical value and potential.

KEYWORDS:

• Breast cancer
• immunotherapy
• bibliometric
• immune checkpoint
• tumor microenvironment

View responses to this article:

Knowledge mapping of immunotherapy for breast cancer: A bibliometric analysis from 2013 to 2022: A correspondence

Introduction

Breast cancer is the most common malignant tumor in women. According to statistics from the World Health Organization, there were approximately 2.3 million new cases of breast cancer in 2020, accounting for 11.7% of all cancer cases.¹ It is the leading cause of cancer-related death among women globally, imposing a significant burden on both health and economies. Breast cancer is a heterogeneous disease, with different molecular subtypes exhibiting distinct biological and clinical characteristics. Over the past three decades, significant advancements in targeted and adjuvant therapies have greatly improved the prognosis of early-stage disease. However, treatment options for advanced metastatic disease remain limited, with a median overall survival of 4–5 years for luminal subtype and 1 year for triple-negative subtype.² Therefore, there is an urgent need to improve treatment modalities. Immunotherapy has emerged as a major milestone in contemporary oncology, revolutionizing the management of various solid tumors. This includes immune checkpoint inhibitors, cancer vaccines, antibody-drug conjugates (ADCs), oncolytic viruses, and adoptive cell therapy. The most extensively studied immunotherapeutic agents for breast cancer are immune checkpoint inhibitors.³ In normal circumstances, the immune system employs inhibitory checkpoint pathways to regulate immune responses against pathogens, preventing excessive responses and limiting tissue damage and autoimmune activity. This mechanism is mediated by several immune checkpoint molecules such as programmed cell death protein 1 (PD-1) and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), which downregulate and inhibit T cell responses by binding to their ligands: programmed cell death ligand 1 (PD-L1), programmed cell death ligand 2 (PD-L2), and CD80/CD86.⁴ Tumor cells exploit this mechanism to create an immunosuppressive microenvironment. Monoclonal antibodies targeting PD-1, PD-L1, and CTLA-4 circumvent this immune suppression, providing cell-mediated anti-tumor activity.³ Immune checkpoint blockade removes the brakes on an activated immune system, leading to sustained antigen-specific recognition and response, and has greatly transformed the treatment landscape for various tumors, such as lung cancer and melanoma.^5,6 Immunotherapy may represent an important new concept and a promising future prospect in breast cancer research.

In recent years, a framework called “research weaving” has been proposed, which utilizes systematic knowledge mapping and bibliometrics to visualize the current hot topics, inherent connections, and changes in research trends in a research field. This framework is one of the important techniques for assessing the quality, credibility, and impact of academic work.^7,8 Bibliometrics, as a quantitative analysis method, offers the advantages of analyzing large quantities of highly heterogeneous literature and objectively and visually demonstrating past academic research activities and achievements. It helps reduce biases in paper evaluation caused by human factors.⁸ Currently, bibliometrics analysis and visualization have been widely used in various fields to analyze research status, hot topics, and trends.^9–11

With the advent of immune checkpoint inhibitors, tumor immunotherapy has been increasingly recognized. The number of scientific publications in the field of breast cancer immunotherapy has dramatically increased. However, there is still a lack of objective quantitative studies on the research status and trends in the face of massive scientific research results. This study is based on the Web of Science (WoS) database, which is one of the most important data sources for academic research in health/medical science-related categories and one of the most widely used databases in bibliometric analysis.^12,13 It statistically analyzes the most contributing countries, institutions, and authors in this field and summarizes the top 10 cited publications in this field. Furthermore, it analyzes the relationship network and hotspots of relevant literature published between 2013 and 2022, identifies important subfields through clustering, and presents them visually. This study provides a comprehensive understanding of the research status, hot topics, evolution, and development trends in the field of breast cancer immunotherapy for researchers in this field.

Materials and methods

Searching strategy and data collection

This study was designed as a cross-sectional study and was conducted on June 11, 2023, using the WOS Core Collection (WoSCC). WoS is widely regarded as the most influential scientific research database, as it includes journals that are typically considered to have the highest quality. It is commonly used for bibliometric analysis of scientific research in the field of health.¹⁴ All relevant literature was retrieved and downloaded in “Plain text” format to avoid biases introduced by updates.

The retrieval and limiting strategy for the literature is as follows: 1. The literature was limited using the TS (“topic,” including title, abstract, author’s keywords, and keywords Plus) keyword search strategy. The search strategy used was TS = (Breast Neoplasm* OR Breast Tumor* OR Breast Cancer* OR Cancer of the Breast OR Cancer of Breast OR Breast Carcinoma*) AND TS = (Immunotherapy OR Immunotherapies OR Immunotherapeutic); 2. The publication date range was set from January 1, 2013, to December 31, 2022; 3. Available records in the WoSCC. The following relevant information from the publications was collected: title, abstract, keywords, authors, institutions, countries/regions, and references. The publications were limited to those written in English. The publication type was limited to articles or reviews, and other types of publications such as conference papers, comments, and editorials were excluded.

Data analysis and mapping

Retrieved data was imported into Citespace (version 6.1.R3), VOSviewer (version 1.6.18), and the bibliometrix package (version 3.2.1) in R (4.0.3, https://www.r-project.org/) to analyze the published research in terms of country/region, research institutions, researchers, citations, keywords, and other relevant information to construct a knowledge map.

Co-authorship analysis is performed to reflect scientific collaboration. When different authors, institutions, or countries/regions coexist in a publication, it is considered as an indication of collaboration.¹⁵ To identify the research’s hot topics and thematic distribution, a co-occurrence analysis was conducted on the author’s keywords. This analysis involved counting the number of times a pair of keywords appeared together in the same document, which served as a measure of their relationship. To uncover the knowledge foundation and its evolution in the field, a co-citation analysis was carried out on the reference list. Co-citation analysis is defined as the situation where two articles appear together in the reference list of a third citing article, indicating a co-citation relationship between the two articles.¹⁶

Co-authorship analysis, co-occurrence analysis, and co-citation analysis were conducted using VOSviewer. It utilizes the VOS mapping technique to construct distance-based maps using similarity matrices. VOSviewer enables the visualization of knowledge maps for large-scale literature data, representing authors, journals, and other relevant information. It has been widely used in bibliometric analysis research.¹⁷ The visualization map consists of nodes and links. The size of a node’s circle is proportional to the frequency of the displayed indicator. In cluster analysis, the color of the circle is determined by its category. The thickness and length of the links between nodes represent their connection strength and relevance.

To discover the knowledge base of the domain and its evolution, the journal co-authorship analysis and reference co-citation analysis were conducted using Citespace, a software for bibliometric analysis and data visualization.¹⁸ The results were clustered based on keywords or disciplines, with Modularity Q > 0.3 and mean silhouette >0.5 indicating stable and reliable clustering results. Additionally, Citespace was used to generate dual-map visualizations of journal connections based on citation and co-citation relationships.¹⁹ For the reference co-citation analysis, the parameters used were time slicing (2013–2022), years per slice,¹ node type (cited reference), selection criteria (top N = 50), and no clipping. Furthermore, burst detection was performed by evaluating the burstiness using the sum appearance weighted by the time window, where a significant increase in the frequency of references within a specific time period is considered an indication of an emerging research topic’s rapid attention.^20,21

To understand the research evolution in the field of immunotherapy for breast cancer, thematic evolution and thematic maps were generated using the Bibliometrix package. Bibliometrix package is a comprehensive scientific mapping analysis tool based on R.²² Thematic evolution and thematic maps involve clustering topics based on keywords from publications and mapping them in a low-dimensional space to reflect the trends in topic changes. Thematic evolution is presented using a Sankey diagram to illustrate the evolution of topics across different time slices. Thematic maps use the density index as the y-axis and the centrality index as the x-axis. Density represents the strength of internal connections between keywords within a topic, while centrality refers to the strength of connections between the topic and other external topics. The maps are divided into four quadrants: Q1: Motor Themes representing significant and well-developed topics; Q2: Niche Themes representing topics highly developed but less connected to other topics; Q3: Emerging or Declining Themes representing topics with low internal and external associations, suggesting emerging or declining trends; Q4: Basic Themes considered as fundamental and transversal topics in the field.²³ Changes and trajectories across different time slices are identified to determine the emergence or decline of topics.²² To compare the contributions and impact of authors and journals in the field, widely used productivity quantification indices such as the H-index²⁴ and G-index²⁵ were also calculated.

All the original data used in the study were obtained from publicly available databases, and there was no involvement of participant data, therefore ethical review was not required.

Results

Temporal distribution of publications

A total of 7,975 publications related to immunotherapy research in breast cancer were retrieved between January 1, 2013, and December 31, 2022. Among these, 5,380 (67.5%) were articles, and 2,595 (32.5%) were reviews. As shown in Figure 1, the flowchart illustrating the literature search and selection process for excluded publications. During this period, the number of studies on immunotherapy for breast cancer gradually increased from 290 publications in 2013 to 1,688 publications in 2022, representing a growth of 482.1% (Figure 2(a)). The total number of citations for all publications was 283,310, with an average of 35.52 citations per publication.

Figure 1. The flowchart of searching and selection process.

Figure 2. The hotspots and burst of references co-citation of publications in the field of immune therapy for breast cancer. (a) The changes in the numbers of global publications over time; (b) The co-occurrence analysis of keywords, nodes are proportional in size to the frequency of keyword occurrence and the color of the nodes is determined by their category in cluster analysis; (c) The co-citation analysis of references in timeline manner. Nodes are proportional in size to the number of reference co-citations and the thickness and color of the node’s rings reflect the number of citations an article receives in a given year. Nodes with purple rings indicate high betweenness centrality, which are essential to connecting conceptual clusters that exist in different time periods. The connections between references are shown by the density of links and a unique color is assigned to each year; (d) The top 20 references with the strong citation bursts. A burst is a surge of the frequency of the citation of an article. The red bar indicates the time interval when the reference co-citation burst started and ended.

Hotspots of keywords

To understand the core content and hotspots in the field, we analyzed the author’s keywords in the retrieved literature. The top 20 most frequently occurring author’s keywords are listed in Table 1. The main research keywords in this field include tumor microenvironment, PD-L1, chemotherapy, metastasis, targeted therapy, PD-1, human epidermal growth factor receptor-2 (HER2), and immune checkpoint inhibitors. We conducted keyword co-occurrence analysis using VOSviewer based on the author’s keywords, as shown in Figure 2(b). A total of 261 keywords were identified that had been used more than 15 times, and these keywords were clustered into six clusters. The red cluster, centered on tumor microenvironment and cancer immunotherapy, indicates that the research direction mainly focuses on components such as cytokines, extracellular matrix, myeloid-derived suppressor cells, and cancer stem cells. It is also associated with factors such as inflammation, angiogenesis, autophagy, apoptosis, immune evasion, and immune suppression. The blue cluster highlights biomarkers related to the prognosis of immunotherapy for breast cancer. The yellow cluster represents the current progress in targeting different immune checkpoints in several lines of treatment for breast cancer. The purple cluster focuses on novel drug delivery methods and technologies related to immunotherapy. The green cluster highlights the role of immune cells such as natural killer cells, T cells, as well as monoclonal antibodies and bispecific antibodies in immunotherapy. The blue-green cluster represents the research area of therapeutic approaches related to immunotherapy in breast cancer, including chemotherapy, radiotherapy, targeted therapy, and combination therapy.

Table 1. Top 20 authors’ keywords of immunotherapy for breast cancer.

Rank	Keywords	Records	Total links	Rank	Keywords	Records	Total links
1	Immunotherapy	2154	5495	11	Metastasis	185	508
2	Breast cancer	1474	3467	12	Targeted therapy	175	489
3	Cancer	536	1358	13	PD-1	166	621
4	Tumor microenvironment	484	1348	14	HER2	132	354
5	Triple-negative breast cancer	419	1107	15	Immune checkpoint inhibitors	130	448
6	Cancer immunotherapy	345	673	16	Radiotherapy	128	362
7	PD-L1	298	986	17	Melanoma	121	338
8	Prognosis	264	689	18	Dendritic cells	105	268
9	Biomarker(s)	261	803	19	Immunogenic cell death	94	221
10	Chemotherapy	241	697	20	Combination therapy	89	304

The information of the top 10 most cited publications is shown in Supplementary Table S1. The most highly cited publication among these is a review by Quail DF et al., published in 2013. This article discusses the contradictory roles of the tumor microenvironment in the progression and metastasis of malignant tumors. It also summarizes various therapeutic attempts to reeducate stromal cells within the tumor microenvironment to exert anti-tumor effects.²⁶ Additionally, this article has the highest average citation rate per year.

Evolution and burst of knowledge base

The compilation of reference citations in publications can be considered as a scientific knowledge repository. We conducted co-citation analysis of the reference citations and presented them in a timeline format. Figure 2(c) (Reference Co-citation Timeline) illustrates eight major clusters identified through the co-citation analysis. During the period around 2007–2013, the co-cited references primarily clustered around the PD-L1 pathway, therapeutic applicability, myeloid-derived suppressor cells, and mucin-1 protein, indicating the early explorations in immunotherapy for breast cancer. Subsequently, around 2012–2016, a cluster of co-cited references related to tumor-infiltrating lymphocytes and abscopal responses emerged, highlighting the significant impact of the tumor immune microenvironment and immune escape in breast cancer immunotherapy. The most recent co-citation cluster, starting from 2016, focuses on metastatic TNBC, indicating it as a recent research hotspot.

The co-citation burst reflects the time periods of rapid changes in the intensity of referenced citations. Based on the start and end times, the top 20 references with strong citation bursts are listed in order (Figure 2(d) and Supplementary Table S2). Among the top 20 most cited references, 16 are articles and 4 are reviews. One article included a phase I clinical trial with 296 patients with advanced cancer, demonstrating objective responses to anti-PD-1 antibodies in approximately one-fourth to one-fifth of patients with non-small cell lung cancer, melanoma, or renal cell carcinoma.²⁷ These two articles ranked first and third, respectively, in terms of co-citation burst strength among all references, indicating that blockade of PD-1 or PD-L1 may become a new benchmark in immune therapy against tumors. The second-ranked reference is a review published in 2012, which summarizes several immune checkpoints with promising clinical applications in cancer immunotherapy.²⁸ Among the frequently cited top 20 references, the field of breast cancer is mainly associated with TNBC, tumor-infiltrating lymphocytes (TILs), and anti-PD-L1 antibody.

Trends of themes

In this section, we conducted an analysis of high-frequency keywords and their temporal variations (Figure 3(a)). To enhance readability, we excluded highly synonymous keywords from the analysis. We constructed topic evolution graphs and topic maps to illustrate the trends in different topics. The topic evolution from 2013 to 2017 and from 2018 to 2022 was visualized using Sankey diagrams (Figure 3(b)). The topic maps, based on centrality and density, depicted the evolution of topics across different time periods (Figure 3(c,d)). When comparing the periods of 2013–2017 and 2018–2022, we observed a significant increase in both centrality and density of the “metastasis” topic, indicating its significance and well-developed nature, as it entered the motor themes. Additionally, from 2011–2015 to 2016–2021, there was an increased density of the “PD-L1” and “PD-1” topics. Notably, compared to the period of 2013–2017, the years of 2018–2022 exhibited a new topic of high centrality and density, namely “tumor microenvironment,” suggesting its emergence as a rapidly developing and emerging topic.

Figure 3. The trends of publications in the field of immune therapy for breast cancer. (a) The changes of high-frequency keywords over time; (b) The thematic evolution of publications in the last decade; (c,d) Thematic maps during 2013–2017 and 2018–2022. Thematic maps were divided into four quadrants: I: Motor themes with high density and centrality; II: Niche themes with high density but low centrality; III: Emerging or declining themes with low density and centrality; IV: Basic themes with low density but high centrality.

Attribution and collaboration of countries/regions, authors and institutions

A total of 103 countries/regions have contributed to this field. Publication information categorized by country/region was collected and calculated (Figure 4(a) and Table 2). The United States had the highest number of publications at 2926. Belgium had the highest number of publications per trillion gross domestic product (GDP), with a value of 274.36. The collaboration relationships between countries/regions are shown in Figure 4(b). The countries/regions were divided into three colored clusters, with collaboration indicated by the thickness of the links, represented by the Total Link Strength (TLS). The top three countries/regions with the highest TLS were the United States (TLS = 1238), China (TLS = 572), and Germany (TLS = 316). The green cluster was centered on China and the United States and included several countries/regions, such as Iran, India, Singapore, Russia, and Egypt. The blue cluster was centered on Japan and Australia. The red cluster mainly consisted of European countries/regions, with Germany, the United Kingdom, and Italy being the prominent centers.

Figure 4. The attribution sources, collaboration networks on exercise and physical activity in older adults. (a) The top 10 productive countries/regions in this field; (b) The co-authorship relationships of countries/regions, the size of the nodes indicates the number of publications, and the thickness and length of the links between the nodes indicate the strength and relevance of the connections between the nodes; (c) The co-authorship relationships of authors; (d) The co-authorship relationships of institutions; (e) The distribution of publications across disciplines; (f) The cumulative growth pattern of publications in the top 10 productive journals; (g) The dual-map overlay of journals, the left label in the figure represents citing journals, the right label represents cited journals, and the colored paths represent the citation relationships between them.

Table 2. The top 10 productive countries/regions in the field of immunotherapy for breast cancer.

Rank	Country/region	Numbers of Publications	Publications per million people*	Publications per trillion GDP*	Numbers of citations	Average citations per publication	Co-authorship total link strength
1	USA	2926	8.82	125.50	147679	50.47	1238
2	PEOPLES R CHINA	2420	1.71	136.46	62335	25.76	572
3	GERMANY	497	5.97	116.67	23945	48.18	316
4	ITALY	551	9.32	261.42	24875	45.15	277
5	ENGLAND	361	5.36	115.28	20540	56.90	271
6	FRANCE	372	5.49	125.77	23359	62.79	237
7	AUSTRALIA	309	12.03	199.01	20006	64.74	194
8	CANADA	295	7.71	148.37	17063	57.84	177
9	JAPAN	266	2.12	53.84	12702	47.75	131
10	BELGIUM	163	14.06	274.36	10484	64.32	130

*Calculations based on 2021 population and GDP data from world bank (https://databank.worldbank.org/). GDP is calculated using GDP (current US$).

A summary of the information regarding the top 10 most productive authors is presented in Table 3. The co-authorship relationships between scholars are illustrated in Figure 4(c), divided into 33 clusters. Sherene Loi was identified as the author with the highest citation count (4007), total link strength (686), and H-index,²⁹ while Giuseppe Curigliano had the highest number of publications³⁰ and G-index.³¹ Table 4 presents the top 10 publishing institutions, with The University of Texas MD Anderson Cancer Center having the highest number of publications (219), Harvard Medical School receiving the highest number of citations (14490), and Dana-Farber Cancer Institute having the highest average citation count per publication (112.86). The collaboration relationships among institutions are divided into eight clusters, as shown in Figure 4(d). The institution with the highest TLS score was The University of Texas MD Anderson Cancer Center (TLS = 576).

Table 3. The top 10 productive authors in the field of immunotherapy for breast cancer.

Rank	Author	Affiliation	Numbers of publications	Numbers of citations	Average citation per publication	Co-authorship total link strength	H-index	G-index
1	Sherene Loi	The University of Melbourne	28	4007	143.11	686	29	40
2	Lajos Pusztai	Yale University	19	3280	172.63	478	17	23
3	Giuseppe Curigliano	European Institute of Oncology IRCCS/University of Milan	44	1576	35.82	469	23	52
4	Roberto Salgado	GZA-ZNA Hospitals	13	1982	152.46	411	17	23
5	Leisha A Emens	University of Pittsburgh	18	1455	80.83	355	14	19
6	Elizabeth A Mittendorf	Brigham and Women’s Hospital/Dana-Farber Brigham Cancer Center	30	978	32.60	331	22	36
7	Phillip K Darcy	University of Melbourne	19	2295	120.79	316	18	21
8	Rita Nanda	The University of Chicago	11	1588	144.36	301	9	12
9	Sylvia Adams	New York University	11	1617	147.00	300	16	19
10	Paul A Beavis	The University of Melbourne	16	2162	135.13	297	16	16

Table 4. The top 10 productive institutions in the field of immunotherapy for breast cancer.

Rank	Institution	Country/region	Numbers of Publications	Numbers of Citations	Average citation per publication	Co-authorship total link strength
1	The University of Texas MD Anderson Cancer Center	USA	219	9671	44.16	576
2	Harvard Medical School	USA	169	14490	85.74	562
3	National Cancer Institute	USA	142	9934	69.96	307
4	Sun Yat-sen University	China	137	4337	31.66	195
5	Chinese Academy of Sciences	China	133	6327	47.57	293
6	Sichuan University	China	113	2803	24.81	95
7	Shanghai jiao tong University	China	112	3694	32.98	167
8	Fudan University	China	111	3978	35.84	123
9	Memorial Sloan-Kettering Cancer Center	USA	105	11589	110.37	341
10	Dana-Farber Cancer Institute	USA	100	11286	112.86	437

Distribution across disciplines and journals

We conducted a statistical analysis of the top 10 subject categories defined by the WoS classification in the publications (Figure 4(e)). The three main subject categories within this field are Oncology, Immunology, and Pharmacology Pharmacy, accounting for approximately 71.12% of the publications. The cumulative growth pattern of annual publications and information on the top 10 productive journals are displayed in Figure 4(f) and Supplementary Table S3. The journal CANCERS had the highest number of publications with 308 articles. The dual-map overlay of journals depicted four major citation relationships between citing journals and cited journals in this field (Figure 4(g)). The labels on the left side of the map represent the citing journals, while those on the right side represent the cited journals. Referential links originate from a journal on the left side of the map and point to a journal on the right side of the map. The color of a link indicates the discipline of the source journal. The most prominent paths, represented by two distinct colors, reflect the three main citation categories. The orange path illustrates that journals in the field of molecular/biology/immunology cite literature from journals in the fields of molecular/biology/genetics as well as health/nursing/medicine. The green path demonstrates that journals in the field of medicine/medical/clinical cite literature from journals in the field of molecular/biology/genetics.

Discussion

Currently, the application of immunotherapy in cancer treatment is increasing. Immunotherapy has become a research hotspot. This study aims to investigate research hotspots, bursts of knowledge base, and trends of themes in immunotherapy for breast cancer over past decade. Literatures focusing on this field published from 2013–2022 were analyzed and presented in a visualized manner. After excluding studies that did not meet the selection criteria, our analysis included 7,975 English papers published in 1,188 journals from 6,209 institutions in 103 countries/regions. Through a bibliometric analysis of research on immunotherapy in the field of breast cancer, this study provides researchers with a general understanding of immunotherapy for breast cancer.

In recent years, there has been an increasing trend in publications related to immunotherapy for breast cancer, indicating growing interest in this topic. The number of publications showed significant growth in 2016, 2020, and 2021, which may be attributed to key events in the field of breast cancer immunotherapy. In 2014, the US Food and Drug Administration (FDA) approved the PD-1 inhibitors nivolumab and pembrolizumab for advanced melanoma, heralding a new era in cancer immunotherapy. In November 2018, the results of the Impassion130 study showed that Atezolizumab plus nab-paclitaxel prolonged progression-free survival (PFS) in patients with metastatic TNBC, both in the intention-to-treat population and the PD-L1-positive subgroup.²⁹ Based on these results, in March of the following year, the FDA approved the PD-L1 inhibitor Atezolizumab in combination with nab-paclitaxel as a first-line treatment for advanced or metastatic TNBC with PD-L1 expression ≥1%, making it the first immune checkpoint inhibitor approved for the treatment of breast cancer. The KEYNOTE-355 trial demonstrated that pembrolizumab, in combination with standard chemotherapy, improved PFS compared to chemotherapy alone in patients with metastatic TNBC.³² Based on these findings, in November 2020, pembrolizumab received accelerated approval from the FDA for the treatment of patients with metastatic TNBC with a baseline combined positive score of 10 or more for PD-L1 expression. With significant clinical research progress in immunotherapy for breast cancer, bringing great hope for breast cancer treatment, immunotherapy for breast cancer has become a research hotspot in recent years.

Tumor microenvironment is currently the research hotspots, which includes tumor cells, immune cells, stromal cells, extracellular matrix, vessels, soluble factors, and physical properties.³³ Non-cancerous cells within the TME play a crucial role in cancer development by promoting tumor survival, progression, metastasis, and treatment resistance. Interactions between tumor cells and stromal cells in the TME are also associated with tumor growth and immunotherapy resistance.³⁴ T cells are key effector cells for cancer control, and different components in the TME have been shown to impact their function. Defects in antigen-presenting cells’ processing and presentation of tumor antigens, as well as downregulation of major histocompatibility complex I molecules in breast cancer cells, can lead to the failure of anti-tumor T cell activation, causing immunotherapy resistance.³⁵ B cells mediate breast cancer regression by secreting immunoglobulin G in response to tumor cells and activating T cells to exert cytotoxic effects and produce cytokines (such as IFN-γ and granulocyte-macrophage colony-stimulating factor), enhancing anti-tumor immunity.³⁶ Myeloid-derived suppressor cells primarily exert T cell inhibitory effects leading to immunotherapy resistance through the upregulation of ARG1, iNOS, and ROS.³⁷ Anti-angiogenic therapy induces tumor vascular normalization, improves blood perfusion, and promotes immune cell recruitment and dendritic cell maturation. Combining anti-angiogenic therapy with immunotherapy produces better clinical responses and reduces adverse events.³⁸

China ranks second in the top 10 most productive countries/regions, second only to the United States, and is the only developing country among them. This may be attributed to the larger number of breast cancer patients in China. There were 2,261,419 newly diagnosed cases of breast cancer worldwide in 2020, with 416,371 cases occurring in China, accounting for approximately 18.4%.³⁹ Consistent with the distribution of countries, China and the United States occupy five institutions in the top 10. This indicates that both China and the United States have played important roles in the academic development in this field. However, it is worth noting that although China has a considerable number of publications, the average citations per publication are much lower compared to other countries, indicating a lack of highly cited papers.

A total of 1,601 papers, accounting for 20.08% of all publications, were published in the top ten academic journals. CANCERS ranked first in terms of total publications, followed by FRONTIERS IN IMMUNOLOGY and FRONTIERS IN ONCOLOGY, indicating a strong interest in breast cancer immunotherapy research in these journals. CLINICAL CANCER RESEARCH had the highest H-index and G-index, indicating better publication quality. These journals are mostly related to oncology, immunology, and biology, and the dual-map overlay showed similar results, indicating that research on breast cancer immunotherapy primarily focuses on clinical and basic research. The reciprocal citation between articles from these two aspects helps explore the mechanisms of breast cancer immunotherapy from clinical problems and ultimately bridge the gap between basic research and clinical practice. The most cited articles were published in NEW ENGLAND JOURNAL OF MEDICINE. In fact, 8 out of the top 20 references with the highest citation burst were from NEW ENGLAND JOURNAL OF MEDICINE, indicating that this influential and prestigious journal is more likely to publish high-quality research in the future.

By analyzing the clustering timeline of highly cited papers and co-cited references, we gained a rough understanding of the development of breast cancer immunotherapy. The early clusters in highly cited papers include the PD-L1 pathway and myeloid-derived suppressor cells. These basic experimental studies laid the foundation for subsequent research. The PD-L1/PD-1 axis is an important immune checkpoint signaling pathway that can downregulate the magnitude of inflammatory responses and maintains immune homeostasis.⁴⁰ During the process of immune evolution, the PD-L1/PD-1 axis is an indispensable pathway for maintaining immune tolerance and preventing autoimmune diseases. However, the PD-L1/PD-1 axis also affects the balance between tumor immune surveillance and immune resistance. Increased PD-L1 expression on tumor cells or TILs leads to T-cell exhaustion, thereby weakening tumor-specific immune responses and promoting tumor progression.⁴¹ Myeloid-derived suppressor cells are a heterogeneous population of immature myeloid cells with immunosuppressive effects, which undergo significant expansion during tumor progression. These cells not only directly support immune escape but also promote tumor invasion through various nonimmune activities. Additionally, they have been shown to impair the efficacy of chemotherapy, radiotherapy, and immunotherapy. Therefore, myeloid-derived suppressor cells are considered potential therapeutic targets in cancer treatment.⁴² Besides, the timeline includes breast cancer, tumor-infiltrating lymphocytes, and abscopal responses.

With a foundation laid by basic experimental research, clinical studies was initiated. First, single-agent immunotherapy was investigated. In a multicenter, nonrandomized phase Ib trial called KEYNOTE-012, Nanda et al. evaluated the efficacy of single-agent pembrolizumab in recurrent or metastatic breast cancer patients with PD-L1 positivity, and the overall response rate was 18.5%.⁴³ Further research by Adams et al. in the KEYNOTE-086 study demonstrated durable antitumor activity of pembrolizumab monotherapy in both previously untreated and treated metastatic TNBC patients.⁴⁴ These results suggest the potential safety and antitumor activity of immunotherapy in advanced TNBC, although the benefit of monotherapy in clinical practice remains limited. In addition to monotherapy, combination therapy of immunotherapy with chemotherapy has been investigated. The IMpassion130 study showed that the combination of atezolizumab and albumin-bound paclitaxel as first-line treatment for advanced TNBC improved the PFS and overall survival (OS) in the intention-to-treat population and PD-L1-positive patients, with a median OS of 25 months in PD-L1-positive advanced TNBC patients compared to 12–18 months with chemotherapy alone.²⁹ IMpassion130 was the first phase III clinical trial to demonstrate the efficacy of immunotherapy as first-line treatment in advanced TNBC, signifying a milestone in the field. Furthermore, the KEYNOTE-355 study conducted by Schmid et al. investigated the clinical efficacy of adding pembrolizumab to chemotherapy compared to placebo plus chemotherapy in patients with unresectable locally recurrent or metastatic TNBC as initial treatment.³² In the first-line treatment of PD-L1-positive (combined positive score ≥10) metastatic TNBC, the combination of pembrolizumab with chemotherapy significantly improved PFS and reduced the risk of recurrence by 35.0% compared to placebo plus chemotherapy. Combination therapy of immunotherapy with targeted therapy is also an area of current interest. Studies have explored combination therapies with PARP inhibitors, AKT inhibitors, and VEGFR2 inhibitors for metastatic TNBC.^30,45,46

TNBC demonstrates a better response to immunotherapy compared to other breast cancer subtypes due to several key characteristics. First, TNBC is characterized by a higher presence of TILs,⁴⁷ which contributes to a more favorable response to immune checkpoint inhibitors. High expression of TILs in TNBC is also associated with improved prognosis.⁴⁸ Additionally, TNBC displays elevated levels of PD-1 and its ligand PD-L1 on both tumor and immune cells, providing a direct target for immune checkpoint inhibitors.⁴⁹ Moreover, TNBC exhibits a high tumor mutational burden (TMB), resulting in the production of tumor-specific neoantigens that activate tumor-specific T cells and elicit an anti-tumor immune response. Immune checkpoint inhibitors can enhance this immune response.⁴⁹ However, it is important to note that breast cancer is not typically a highly immunogenic disease, leading to a relatively poor response to immunotherapy overall.⁵⁰ Not all breast cancer patients benefit from immunotherapy or combination therapy, and unresolved issues remain.

Due to the distinct response patterns of tumors to immune checkpoint inhibitors compared to chemotherapy, the Immune Response Evaluation Criteria in Solid Tumors have been developed to better assess the benefits of immunotherapy. However, most current trials still utilize traditional Response Evaluation Criteria in Solid Tumors.⁵¹ Pathological complete response (pCR) after neoadjuvant chemotherapy is considered an alternative endpoint for long-term clinical outcomes, which might be less appropriate to evaluate long-term immune memory responses that could sustain therapeutic effects and prevent relapses.³ Addressing current challenges, future efforts should focus on using more appropriate methods to evaluate immune therapy responses and developing suitable endpoints for immunotherapy research. It is important to note that breast cancer is not typically highly immunogenic, resulting in a relatively poor overall response to immunotherapy.⁵⁰ Not all breast cancer patients benefit from immunotherapy or combination therapy, and unresolved issues persist. Previous studies explored biomarkers related to the efficacy of breast cancer immunotherapy. Early increases in TILs two weeks after immunotherapy were correlated with an increased pCR rate.⁵² In studies such as IMpassion-031 and KEYNOTE-522, PD-L1-positive tumor patients were associated with higher pCR rates.³¹ Further analysis of the GeparNuevo study revealed TMB and immune infiltration as independent predictors of response to neoadjuvant immune checkpoint inhibition in early triple-negative breast cancer.⁵³ Continued development of rational and precise immunotherapy biomarkers is crucial for accurately selecting breast cancer patients who will benefit from immunotherapy. Additionally, optimizing clinical trial design, promoting interdisciplinary collaboration, and integrating multi-omics big data are necessary to identify the best treatment combinations for immunotherapy. This approach aims to minimize immune-related toxicities, improve patient outcomes, and achieve the maximum potential of immunotherapy. In addition, there is a need to focus on exploring more promising applications of immunotherapy methods, such as developing novel immune checkpoints, chimeric antigen receptor T cells, and bispecific T cell engager antibodies.

A limitation of this study is that only literature retrieved from the WoSCC was used. While WoSCC is an important source for bibliometric analysis due to its consistent and standardized record formats and wide coverage,⁵⁴ the retrieval data for our publication may be incomplete. Currently, effectively integrating large-scale literature indexing records across databases is challenging, and this limitation is common in bibliometric research.^11,55 In the future, appropriate tools will be needed to address this limitation by facilitating the integration of large-scale cross-platform literature data, in order to minimize selection bias.

In conclusion, there has been a rapid increase in global research on immune therapy for breast cancer over the past decade. The number of studies in this field has skyrocketed, and there has been a growing emphasis from countries and institutions. Research in this area encompasses eight major interdisciplinary topics. Among them, Tumor microenvironment is currently the research hotspots. The immune therapy for metastatic TNBC represents a crucial direction for future development.

Author contributions

FQ conceived the study, conducted most of the data analysis and drafted the manuscript. GW participated in the data analysis and made detailed revisions to the manuscript. PW participated in the data analysis and the figure production. XL and XZ guided the entire analysis process and determined the direction of the research for each section. All the authors read and approved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2024.2335728.

Additional information

Funding

This work was supported by Open Fund of Chongqing Key Laboratory for Intelligent Oncology in Breast Cancer [Grant No. 2023rxaznzlzdsys01], Chongqing Municipal Health and Health Commission [Grant No.2019NLTS005], Chongqing Research Institute Performance Incentive Guide Special Project and Beijing Science and Technology Innovation Medical Development Foundation [Grant No. KC2021-JF-0167-05].