Mapping the landscape of gastric cancer immunotherapy: Bibliometric insights into advances and hotspots

Abstract

BACKGROUND

Immunotherapy has surfaced as a promising therapeutic modality for gastric cancer (GC). A comprehensive review of advancements, current status, and research trends in GC immunotherapy is essential to inform future investigative efforts.

AIM

To delineate the trends, advancements, and focal points in immunotherapy for GC.

METHODS

We performed a bibliometric analysis of 2906 articles in English concerning GC immunotherapy published from 2000 to December 20, 2023, indexed in the Web of Science Core Collection. Data analysis and visualization were facilitated by CiteSpace (6.1.6R), VOSviewer v.1.6.17, and GraphPad Prism v8.0.2.

RESULTS

There has been an increase in the annual publication rate of GC immunotherapy research. China leads in publication volume, while the United States demonstrates the highest citation impact. Fudan University is notable for its citation frequency and publication output. Co-citation analysis and keyword frequency revealed and highlighted a focus on GC prognosis, the tumor microenvironment (TME), and integrative immunotherapy with targeted therapy. Emerging research areas include gastroesophageal junction cancer, adoptive immunotherapy, and the role of Treg cell in immunotherapy.

CONCLUSION

GC immunotherapy research is an expanding field attracting considerable scientific interest. With the clinical adoption of immunotherapy in GC, the primary goals are to enhance treatment efficacy and patient outcomes. Unlike hematological malignancies, GC's solid TME presents distinct immunological challenges that may attenuate the cytotoxic effects of immune cells on cancer cells. For instance, although CAR-T therapy is effective in hematological malignancies, it has underperformed in GC settings. Current research is centered on overcoming immunosuppression within the TME, with a focus on combinations of targeted therapy, adoptive immunotherapy, Treg cell dynamics, and precise prognosis prediction in immunotherapy. Additionally, immunotherapy's role in treating gastroesophageal junction cancer has become a novel research focus.

Key Words: Bibliometrics; Gastric cancer; Immunotherapy; Immune checkpoint inhibitor; Tumor microenvironment

Core Tip: This study employs bibliometric analysis to systematically elucidate the landscape of gastric cancer (GC) immunotherapy literature. Immunotherapy for GC has become a prominent research area, consistently attracting scientific scrutiny. Principal research topics include the prognostic evaluation of immunotherapy, immune microenvironment dynamics, synergistic strategies with targeted therapy, adoptive immunotherapy strategies, and the immunological treatment of gastroesophageal junction cancer. These topics continue to be at the forefront of current research initiatives.

INTRODUCTION

Gastric cancer (GC) ranks as the third leading cause of cancer-related mortality and sixth in global cancer incidence, posing a substantial health and healthcare burden[1]. Due to its nonspecific early symptoms, GC is frequently diagnosed in an advanced stage[2,3], with less than 10% of patients surviving beyond five years[4]. Hence, the development of innovative therapies to enhance both overall and progression-free survival (PFS) in GC patients, especially those with advanced disease, remains a critical therapeutic priority[5].

Immunotherapy exploits the immune system's defense mechanisms - its specificity, efficacy, and immunological memory - to modulate or suppress the body's immune response for treating various pathologies[6]. This therapeutic approach has been applied to a wide range of conditions, including viral diseases, autoimmune diseases, allergies, immune rejection responses, and neoplasms. Notably, cancer immunotherapy has witnessed notable advancements in recent years[7]. Clinical oncology applications encompass adoptive cell therapy (ACT), immune checkpoint inhibitors (ICIs), tumor vaccines, cytokine therapy, and oncolytic viral therapy[8].

ACT involves the isolation of tumor-infiltrating lymphocytes (TILs), predominantly T cells, from a patient. These lymphocytes are then subjected to ex vivo genetic modifications and proliferation before being reintroduced into the patient, with the objective of selectively eliminating neoplastic cells[9]. In contemporary clinical practice, chimeric antigen receptor T-cell therapy (CAR-T) has achieved significant therapeutic efficacy. CAR-T cell treatments targeting diverse tumor antigens, have shown promising results in clinical trials[10]. Researchers continue to refine CAR-T cells to specifically engage particular antigenic determinants, aiming to treat a broad range of tumors[11,12]. Additionally, novel therapies involving engineered CAR-NK cells and CAR-macrophages have been developed, exhibiting therapeutic potential[13,14]. Nonetheless, the implementation of CAR-T immunotherapy in GC confronts challenges, including the paucity of immune cell infiltration within the tumor microenvironment (TME), diminished cytotoxicity of immune cells in proximity to tumor cells, and the variability of adverse effects[15-17].

Immune checkpoints are critical molecular pathways that preserve self-tolerance within the immune system, yet can be exploited by tumors to evade immune surveillance[18]. ICIs function by targeting these pathways, thereby inhibiting tumor-mediated immune escape and revitalizing the immune system's capacity to eliminate cancer cells[19]. The primary targets for ICIs include programmed cell death ligand 1 (PD-L1), programmed cell death protein 1 (PD-1), and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4)[20,21]. Despite clinical studies demonstrating the efficacy of ICIs in GC[22], the efficacy of monotherapy is often limited, and combinations of PD-L1 monoclonal antibodies with chemotherapy are increasingly used as first-line therapy for advanced GC[15]. Furthermore, the integration of ICIs with other therapeutic strategies has been shown to augment the incidence of severe side effects[23,24]. A key area of current research focuses on combining ICIs with other modalities to overcome tumor-mediated immunosuppression and enhance the efficacy of immunotherapy.

Cytokines, a class of small proteins secreted by both immune and non-immune cells, which modulate immune cell interactions and communication[25]. They play a pivotal role in the TME, influencing immune cell-mediated tumor elimination[26,27]. Cytokines have become a significant therapeutic target in tumor immunotherapy, particularly for enhancing treatment efficacy. Traditional cytokines such as IL-2 and IFN-α have demonstrated clinical effectiveness[28,29]. However, due to poor patient tolerance and potential severe toxicity, cytokine monotherapy is no longer the standard for cancer treatment. Current research primarily focuses on combining cytokines with other immunotherapies to augment their therapeutic impact.

Tumor vaccines stimulate T cell-mediated immune responses against tumors by utilizing tumor-specific antigens. The focus in tumor vaccine has focused on identifying tumor antigens and developing effective vaccine delivery systems[30]. Initially, tumor-associated antigens, present in both normal and tumor cells, were targeted. The focus has since shifted towards neoantigens, exclusively expressed in tumor cells[31]. Various platforms for vaccine delivery, including DNA, RNA, peptides, and dendritic cells, have been extensively explored to enhance vaccine efficacy[32-35]. Despite significant progress in tumor vaccines in recent years, improving their effectiveness remains a challenge. The broad application of cancer vaccines faces numerous obstacles, such as identifying new tumor antigens, rapid artificial synthesis of vaccines, and monitoring immune responses with clinical significance. Advances in next-generation sequencing and increased computational power have enabled the identification of more valuable tumor antigens. However, there are currently few clinical studies on immunization vaccines, and the potential value of immunization vaccines in GC immunotherapy remains to be further explored.

Recent advancements in gene editing and viral transduction have propelled oncolytic virotherapy forward. These therapies employ genetically engineered viruses to target and infect cancer cells, thereby promoting a pro-inflammatory environment and bolstering immune surveillance against tumors[36]. Talimogene laherparepvec, an oncolytic virus, is recommended for metastatic melanoma due to its efficacy in advanced-stage disease[37]. However, in GC, the efficacy of virotherapy is constrained by insufficient antiviral immune responses, tumor microenvironmental immunosuppression, and the absence of predictive biomarkers for treatment response.

Bibliometrics involves quantitatively assessing literature from public databases, analyzing the trajectory of published research over time[38]. When integrated with graphing tools, bibliometrics visually represents research trends and focal points, enabling researchers to monitor the evolution of specific research areas and pinpoint ongoing challenges within specific fields. The sustained interest in certain topics can indicate unresolved questions in each research direction. By dissecting bibliometric analyses, researchers can identify critical underlying issues in their field. This study utilizes bibliometric techniques to review both historical and current landscape of GC immunotherapy, aiming to forecast future trends and anticipate emerging research areas.

MATERIALS AND METHODS

Data sources and search strategies

In the domain of literature analysis, the Web of Science Core Collection (WoSCC) is the preeminent database, recognized for its precision in literature categorization. We selected WoSCC as our primary data repository. Our search, performed on December 20, 2023, included all articles concerning GC immunotherapy from January 1, 2000, to December 20, 2023. The search strategy was as follows: TS = [(Gastric OR Stomach) AND (cancer OR tumor OR carcinoma OR neoplasm OR tumorous OR neoplastic)] AND TS = (Immunotherapy OR Immunotherapies OR immunotherapeutic). Inclusion criteria included: (1) Full-text articles on GC immunotherapy; (2) English-language articles and reviews; and (3) Publications from January 1, 2000, to December 20, 2023. Exclusions comprised: (1) Studies unrelated to GC immunotherapy; and (2) Non-peer-reviewed articles, such as news items, conference abstracts, and brief communications. Selected articles were exported in plain text, with the literature screening and search methodology depicted in Figure 1.

Figure 1 Flow chart illustrating the process for screening and searching literature.

Data analysis

For visualizing annual publication trends and national distribution, we employed Graphpad Prism version 8.0.2. Additionally, we also utilized VOSviewer (1.6.18) and CiteSpace (6.1.6R Advanced Edition) for a comprehensive analysis and to construct a scientific knowledge map. VOSviewer, a Java-based, open-source tool developed in 2009, is designed for managing and visualizing large bibliometric datasets. CiteSpace visualizes research trends by generating co-citation networks, aiding in the exploration of theories and assessment of technologies. This tool allows users to forecast research trajectories and gain insights into knowledge domains, research frontiers, and trends. By integrating these analytical tools, we can visually track research hotspots and uncover latent information concerning GC immunotherapy.

RESULTS

Our findings revealed a total of 2,906 publications on GC immunotherapy in the WoSCC from January 1, 2000, to December 20, 2023, comprising 2,132 articles (93.48%) and 774 reviews (6.52%). These publications spanned 80 countries and involved 2,905 institutions, with 14181 authors contributing. The annual publication count has shown a consistent rise since 2000, categorized into three phases: a period of slow growth from 2000 to 2015, with less than 50 papers per year, reflecting minimal research interest; a period of steady increase from 2016 to 2019, suggesting growing research attention; and a period of rapid growth post-2020, peaking in 2022, indicating a surge in research activity (Figure 2).

Figure 2 Publications by year distribution. The number of publications increased modestly from 2000 to 2015, then gradually from 2016 to 2019, and surged after 2020, reaching its peak in 2022 in this field.

Countries and institutions

The investigation of GC immunotherapy has been a global endeavor, with participation from 80 countries and regions. As depicted in Figure 3, the annual publication output from the top ten nations over the past decade reveals that China, the United States, Japan, Germany, and Italy are leading contributors. Notably, China's publications constitute 52.86% of the global total, significantly exceeding contributions from other countries.

Figure 3 The annual publication volume of the top ten nations for the previous years. A and B: The line graph (A) and heat map (B) of national publications show that the top 5 countries in this field are China, the United States, Japan, Germany, and Italy. China publishes many more articles than any other country.

The United States exhibited the highest citation/publication ratio at 56.81, with 28577 citations, outperforming all other top 10 publishing nations/regions (Table 1). China, with 1536 publications, ranked second in both publication volume and citation/publication ratio at 16.10, suggesting a relatively lower overall publication quality. Figure 4 illustrates robust collaboration between the United States and China, the leading publishers, as well as the United States's significant connections with France, Italy, and the United Kingdom. China's partnerships were notably more pronounced with Japan, Germany, Spain, and South Korea. Despite its high publication volume, China's frequent citations underscore its prominence in the field. Lately, nations like Germany and Japan have experienced a surge in publications, potentially linked to their collaborations with China.

Figure 4 The network of cooperation between China and the United States is demonstrated in the cooperative network. The United States maintains close collaborations with France, Italy, and the United Kingdom, while China's collaborations are closer with Japan, Germany, Spain, and South Korea.

Table 1 The table of national publication volume and citation situation.

Rank	Country/region	Article counts	Centrality	Percentage (%)	Citation	Citation per publication
1	China	1536	0.06	52.86	24723	16.10
2	United States	503	0.13	17.31	28577	56.81
3	Japan	291	0.03	10.01	9887	33.98
4	Germany	156	0.08	5.37	5726	36.71
5	Italy	153	0.18	5.26	4772	31.19
6	South Korea	134	0.08	4.61	3473	25.92
7	England	117	0.22	4.03	5191	44.37
8	France	69	0.07	2.37	3432	49.74
9	Spain	55	0.08	1.89	2083	37.87
10	Iran	47	0.08	1.62	1003	21.34

A total of 2905 institutions have published articles on GC immunotherapy. Among them, nine Chinese universities and one American university were in the top ten for publication volume (Table 2). Fudan University led with 108 publications, garnering 2695 citations, and an average of 24.95 citations per paper. Shanghai Jiao Tong University followed with 86 papers, 2079 citations, and 24.17 citations per paper, while Nanjing Medical University ranked third with 86 papers, 1385 citations, and 16.10 citations per paper. Analysis indicated a preference for domestic collaborations among institutions (Supplementary Figure 1). To enhance academic exchange, we advocate for increased international cooperation to overcome academic silos.

Table 2 Publication status of institutions.

Rank	Institution	Country	Number of studies	Total citations	Average citation
1	Fudan University	China	108	2695	24.95
2	Shanghai Jiao Tong University	China	86	2079	24.17
3	Nanjing Medical University	China	86	1385	16.10
4	Peking University	China	84	1530	18.21
5	Sun Yat Sen University	China	84	2465	29.35
6	Chinese Academy of Sciences	China	66	686	10.39
7	Chinese Academy of Medical Sciences - Peking Union Medical College	China	57	1257	22.05
8	Zhengzhou University	China	57	1262	22.14
9	University of Texas System	United States	56	7621	136.09
10	Zhejiang University	China	54	1616	29.93

Journals

Table 3 and Supplementary Figure 2 enumerate the top 10 journals by publication volume and citation frequency, respectively. Frontiers in Oncology led with 149 papers (5.13%), followed by Frontiers in Immunology with 127 papers (4.37%), Cancers with 99 papers (3.41%), and Journal for Immunotherapy of Cancer with 45 papers (1.55%; Table 3). Among these prolific journals, the Journal for Immunotherapy of Cancer reported the highest impact factor (IF) of 10.9, with all journals classified within the Q1 or Q2 quartile.

Table 3 The top ten journals by productivity and citations.

Rank	Journal	Article counts	Percentage (%)	IF	Quartile in category
1	Frontiers in Oncology	149	5.13	4.7	Q2
2	Frontiers in Immunology	127	4.37	7.3	Q2
3	Cancers	99	3.41	5.2	Q2
4	Journal for Immunotherapy of Cancer	45	1.55	10.9	Q1
5	Frontiers in Genetics	41	1.41	3.7	Q2
6	Cancer Immunology Immunotherapy	40	1.38	5.8	Q2
7	Frontiers in Cell and Developmental Biology	39	1.34	5.5	Q2
8	International Journal of Molecular Sciences	39	1.34	5.6	Q1
9	BMC Cancer	37	1.27	3.8	Q2
10	World Journal of Gastroenterology	36	1.24	4.3	Q2

IF: Impact factor.

The journal impact, assessed by co-citation frequency, reflects its influence within the scientific community. According to Figure 5A and Supplementary Table 1, Clinical Cancer Research was the most co-cited journal with 1721 citations, followed by Cancer Research with 1658 citations and Nature with 1561 citations. Among the top ten co-cited journals, Lancet recorded the highest citation count with 1434, and the most substantial IF at 168.9. All co-cited journals were classified within the Q1 category.

Figure 5 Visualization of journal-related statistical data. A: The co-citation network diagram of journals visualizes the connections between the most frequently co-cited journals; B: Co-overlay map of journals. The research published in the domains of molecular biology/immunology is primarily cited by journals in the field of molecular biology/genetics, while the research in the field of medicine/medical/clinical primarily receives citations from journals in the field of molecular biology/genetics.

The dual-map overlay delineates thematic dispersion in scholarly articles (Figure 5B). Colorful trajectories represent citation links, with cited journals on the left and citing journals on the right. Analysis reveals two principal color-coded citation pathways: Molecular biology/immunology research is primarily cited by journals within the molecular biology/genetics realm, whereas medicine/medical/clinical publications are predominantly referenced by the molecular biology/genetics field.

Authors

Table 4 and Supplementary Figure 3 enumerates the top ten authors by publication count in GC immunotherapy, accounting for 7.81% of the total with 227 papers. Lin Shen leads with 33 publications, followed by Hao Liu with 27, and Kyoung-Mee Kim with 24. Among these authors, two are from South Korean and eight are Chinese.

Table 4 The top 10 authors by total publications.

Rank	Author	Count	Location	Rank	Co-cited author	Citation
1	Lin Shen	33	China	1	Fuchs CS	570
2	Hao Liu	27	China	2	Jemal A	545
3	Kyoung-Mee Kim	24	Korea	3	Bang YJ	540
4	Jeeyun Lee	23	Korea	4	Kang YK	513
5	Baorui Liu	22	China	5	Janjigian YY	503
6	Hao Chen	21	China	6	Shitara K	479
7	Jia Wei	20	China	7	Bass AJ	437
8	Xiaotian Zhang	20	China	8	Smyth EC	401
9	He Li	19	China	9	Sung H	377
10	Chao Lin	18	China	10	Le DT	367

CiteSpace delineates the author collaboration network (Figure 6), while Table 4 identifies the top 10 authors with the highest citation and co-citation frequencies. A total of 46 authors have received over 50 citations each, reflecting significant scholarly recognition and impact. The network's prominent nodes that represent authors with the most co-citation, notably Fuchs CS (570 citations), Jemal A (545 citations), and Bang YJ (540 citations).

Figure 6 Author collaboration network diagram.

References

From 2000 to 2023, the reference network encompassed 1306 nodes and 4789 links over a one-year interval (Figure 7A). The most frequently co-cited article, according to Supplementary Table 2, was the JAMA Oncology publication by Fuchs et al[39], titled "Safety and Efficacy of Pembrolizumab Monotherapy in Patients with Previously Treated Advanced Gastric and Gastroesophageal Junction Cancer: Phase 2 Clinical KEYNOTE-059 Trial" (IF = 28.4). This study evaluated pembrolizumab's safety and efficacy in patients with advanced gastric or gastroesophageal junction cancer who had undergone prior treatments. The phase 2, open-label, single-arm, multi-cohort, global trial KEYNOTE-059, conducted from March 2, 2015, to May 26, 2016, included 259 patients from 16 countries. With an average follow-up of 5.8 months (range 0.5-21.6), patients were administered 200 mg of intravenous pembrolizumab every three weeks until disease progression or unacceptable toxicity was observed. The primary endpoints, safety and objective response rate (ORR), were assessed by central radiologic review according to RECIST v1.1. PD-L1 expression was evaluated through immunohistochemistry (IHC), with response duration as a secondary endpoint. Among the 259 patients, 76.4% were male, 77.2% Caucasian, with a median age of 62 years (range 24-89). The ORR was 11.6% (95%CI: 8.0%-16.1%), with a complete response rate of 2.3% (95%CI: 0.9%-5.0%). The median response duration was 8.4 months. Response rates for PD-L1-positive and PD-L1-negative tumors, were 15.5% and 6.4%, respectively, with durations of 16.3 and 6.9 months. Grade 3 to 5 treatment-related adverse events occurred in 17.8% of patients, leading to discontinuation in 0.8% due to adverse events and two treatment-related deaths. Pembrolizumab monotherapy demonstrated promising efficacy and tolerability in pretreated advanced gastric or gastroesophageal junction cancer patients, showing significant responses in both PD-L1-positive and PD-L1-negative tumors. Further research is warranted to confirm pembrolizumab's efficacy in this patient group. Co-citation and time clustering analyses (Figures 7B and 8) identified early research hotspots such as ok-432 (cluster4) and carcinoembryonic antigen (cluster 7), cytokine-induced killer cells (clusters 6 and 8), myeloid-derived suppressor cells (cluster 10), malignant ascites (cluster 11), and dendritic cells (cluster 12) as mid-term hotspots, with prognosis (cluster 0), TME (cluster 1), gastrointestinal cancer (cluster 2), Epstein-Barr virus (EBV; cluster 3), targeted therapy (cluster 5), gut microbiota (cluster 9), and CAR-T (cluster 13) as current hot topics and trends in the field.

Figure 7 Statistical feature analysis of co-cited literature. A: Co-cited literature network diagram. Using one year as a time slice, the period spans from 2000 to 2023, featuring a network with 1306 nodes and 4789 links; B: The co-citation clustering diagram conducted a cluster analysis of the topics of the cited literature, displaying 14 clusters.

Figure 8 Volcano plot of co-cited literature. Ok-432 (cluster 4) and carcinoembryonic antigen (cluster 7) are early research hotspots. Cytokine-induced killer cells (cluster 6), Cytokine induced killer cells (cluster 8), myeloid-derived suppressor cells (cluster 10), malignant ascites (cluster 11), and dendritic cells (cluster 12) are mid-term research hotspots. Prognosis (cluster 0), the tumor microenvironment (cluster 1), gastrointestinal cancer (cluster 2), Epstein-Barr virus (cluster 3), targeted therapy (cluster 5), gut microbiota (cluster 9), and CAR-T (cluster 13) are hot topics and trends in this field.

Keywords

Utilizing keyword co-occurrence analysis in VOS viewer reveals the current landscape and projected trajectory of the research field. The most frequent keywords were "expression" "chemotherapy", "prognosis" and "open-label" (Table 5, Figure 9A and Supplementary Figure 4). After excluding non-relevant terms, a network of 191 keywords with a minimum frequency of 26 occurrences was formed, delineating four distinct clusters. Cluster 1 (red) encompasses 66 keywords, such as "expression", "survival", "prognosis", "blockade", "biomarker", "microenvironment", "cells", "resistance", "angiogenesis", "receptor", "invasion", "infection", "immunity", "methylation", "subtypes", "landscape" and "antibody". Cluster 2 (green) comprises 58 keywords, including "T-cells", "breast cancer", "tumor-infiltrating lymphocytes", "dendritic cells", "stem cells", "up-regulation" and "induction". Cluster 3 (blue) contains 53 keywords, such as "chemotherapy", "open-label", "therapy", "efficacy", "surgery", "nivolumab", "oxaliplatin", "case report", "trial", "HER-2" and "trastuzumab". Cluster 4 (yellow) includes 14 keywords, such as "biomarkers", "non-small cell lung cancer", "checkpoint inhibitors", "CTLA-4 blockade", "microsatellite instability", "mismatch repair deficiency" and "PD-1". Additionally, a volcano map generated with CiteSpace illustrates the temporal evolution of research foci (Figure 9B).

Figure 9 Keyword Analysis. A: The high-frequency keyword network graph indicates that the most prevalent keyword is "expression", followed by "chemotherapy", "prognosis", and "open-label"; B: The keyword clustering volcano plot clearly illustrates shifts in research hotspots over time.

Table 5 The top 20 most popular keywords.

Rank	Keyword	Counts	Rank	Keyword	Counts
1	Expression	574	11	T-cells	205
2	Chemotherapy	454	12	Microsatellite instability	202
3	Prognosis	349	13	Blockade	184
4	Open-label	315	14	PD-L1	171
5	Survival	279	15	Colorectal-cancer	163
6	Nivolumab	268	16	Gastroesophageal junction	161
7	Tumor microenvironment	252	17	Dendritic cells	150
8	Cells	238	18	Biomarker	149
9	Double-blind	226	19	Breast-cancer	139
10	Therapy	206	20	Regulatory t-cells	128

Burst analysis of references and keywords

Using CiteSpace, we identified the 50 most influential citations in the field of GC immunotherapy research (Figure 10A). Among these, the publication titled "Comprehensive Molecular Characterization of Gastric Adenocarcinoma" by Cancer Genome Atlas Research Network[40], with the highest citation impact (74.38), confirmed GC's substantial role in cancer mortality. Despite GC's histological and etiological heterogeneity complicating molecular and clinical profiling, Cancer Genome Atlas (TCGA)'s examination of 295 primary gastric adenocarcinomas revealed four distinct molecular subtypes. These include EBV-positive tumors with PIK3CA mutations, hypermethylation, and JAK2, CD274 (PD-L1), and PDCD1 LG2 (PD-L2) amplifications; microsatellite unstable tumors with elevated mutation rates affecting targetable oncogenic signaling proteins; genomically stable tumors characterized by diffuse histological characteristics, RHOA mutations, or RHO family GTPase activating protein fusions; and chromosomally unstable tumors characterized by aneuploidy and receptor tyrosine kinase amplifications. These classifications provide a strategic framework for patient stratification and targeted therapy development. The dataset of 50 references spans from 2000 to 2023, reflecting sustained citation frequency over two decades, with 10 papers currently experiencing a peak in citations frequency, indicating the vibrant research interest in GC immunotherapy.

Figure 10  Cited references and keyword burst analysis. A: Cited references burst visualization. All 50 references were published between 2000 and 2023, indicating their frequent citation over the past two decades, with 10 of these papers currently experiencing peak in terms of citations; B: The keyword burst visualization indicates that "gastroesophageal junction", " adoptive immunotherapy" and "regulation of T cells" are three current research hotspots.

We analyzed the 455 most significant burst keywords, focusing on the top 50 with the strongest activity (Figure 10B). These keywords underscore emerging research directions and current focal points within the GC immunotherapy domain.

DISCUSSION

Recent comprehensive research has focused on immunotherapy for GC, uncovering a plethora of potential therapeutic targets and corresponding agents, which are currently in various stages of clinical trials and increasingly integrated into GC treatment protocols. The proliferation of immunotherapy studies in GC has been particularly notable over the past few years. Unlike traditional reviews and meta-analyses, bibliometric analysis provides a more direct visualization of data, facilitating the identification of emerging research trends and focal points within the field. This study employed bibliometric methods to assess all publications on immunotherapy indexed by the WoSCC from 2000 to December 20, 2023. The analysis delineated key statistical characteristics of the publications, encompassing citations, authorship, journals, countries, institutions, and keywords, and provided an overview of research trends and the current landscape of the discipline.

Country and institution

In this study involved 2906 articles, with 1536 originated from China, representing 52.86% of the total corpus (Table 1). Nine Chinese academic institutions ranked among the top 10 in terms of publication volume in this domain. China's dominance in global publication output can be attributed to its large population, extensive research infrastructure, and the high prevalence of GC[1]. However, despite its leading publication volume, China's citation rate per paper is the lowest among the top ten countries, suggesting a general concern regarding the quality of immunotherapy-related articles emanating from China. Conversely, the United States, follows with 17.31% of the total publications, has achieved a significant total citation count of 28577, with a citation-to-publication ratio of 56.81, indicating a higher quality compared to other countries. Japan, ranks third in publication volume, constituting 10.01% of the total. The high incidence of GC in East Asia, potentially linked to high salt intake in the region's diet, is reflected in the substantial publication output from China and Japan[1].

Among the top 10 institutions excelling in publication output in GC immunotherapy, nine are based in Chinese, with the University of Texas System representing the United States. The University of Texas System, despite ranking ninth in publication volume, boasts the highest average citations per publication, at 136.09, significantly outpacing the next institution. This fact underscores the superior quality of United States-published articles, consistent with national-level findings. The geographic distribution of leading nations in GC immunotherapy research is predominantly in economically developed regions, such as East Asia and North America. This distribution highlights the pivotal role of financial support in medical research. It is advocated that medical institutions in developed regions enhance collaborative efforts with those in less developed areas to elevate the global standard of GC immunotherapy research.

References

The top 10 co-cited references, detailed in Supplementary Table 3, include a preeminent article from JAMA Oncology titled "Safety and Efficacy of Pembrolizumab Monotherapy in Patients With Previously Treated Advanced Gastric and Gastroesophageal Junction Cancer: Phase 2 Clinical KEYNOTE-059 Trial"[39] (IF = 28.4). This study assesses the therapeutic efficacy and safety of pembrolizumab in patients with advanced GC or gastroesophageal junction cancer who had undergone prior treatment. Notably, six of the top 10 co-cited references pertain to clinical trials of immunotherapeutic agents for GC, underscoring the significant clinical impact of immunotherapy in this context. Investigators are currently prioritizing strategies to enhance the efficacy of immunotherapy in clinical settings[41]. Given the pronounced heterogeneity of GC, tailoring therapeutics to specific GC subtypes is a burgeoning area of interest[42]. The co-cited reference volcano plot (Figure 8) highlights that GC prognosis and TME are burgeoning research foci in the realm of GC immunotherapy. These two aspects are pivotal for guiding the therapeutic application of immunotherapy in GC.

Prognosis: Immunotherapy, particularly ICIs, holds significant promise for advancing cancer treatment and has seen extensive clinical application[43]. However, ORRs to ICIs vary considerably across studies, ranging from 10% to 26%[44]. Predicting the response of GC patients to ICIs and their prognosis is crucial for enhancing the clinical efficacy of these therapies, aligning with the goals of precision medicine[45]. Numerous biomarkers, validated by clinical studies, are currently in use, including PD-L1 expression levels, EBV infection status, tumor mutational burden (TMB), TME, and microsatellite instability (MSI)/mismatch repair (MMR) status [41]. Although IHC is a standard practice for measuring PD-L1 expression levels and forecasting cancer prognosis in most solid tumors, clinical trials yield varying predictive outcomes due to factors such as inconsistent endpoint selection, evaluation systems, and the heterogeneity of PD-L1 expression levels in tumors[46]. PD-L1 expression within a tumor can exhibit significant intra-tumor heterogeneity and is subject to temporal changes, especially post-treatment[47]. Consequently, relying solely on PD-L1 expression levels for predicting immunotherapy outcomes is insufficient for the precise treatment needs of GC patients. For MSI and MMR, both TCGA and the Asian Cancer Research Group have incorporated MSI into their molecular subtyping frameworks for GC[48], with evidence suggesting that MSI-high GC has a more favorable prognosis compared to microsatellite stable (MSS) GC[49] and demonstrates greater sensitivity to immunotherapy[50]. However, the predictive utility of MSI in immunotherapy for GC with a low MSI phenotype necessitates further extensive clinical trials to establish its efficacy. The concurrent assessment of multiple biomarkers to gauge immunotherapy responsiveness is gaining traction in the medical community. A notable example is the PD-L1/MSI/TMB biomarker panel, which integrates PD-L1 expression with MSI and TMB to provide a more comprehensive view of a patient's immune profile, thereby enhancing the precision of immunotherapy response predictions[51]. The integration of genomic, transcriptomic, and proteomic analyses has proven instrumental in deepening our understanding of a tumor's immunotherapy response, as evidenced by studies on head and neck squamous cell carcinoma's response to chemotherapy, immunotherapy, and combined treatments[52].

Advancements in immune checkpoint research have unveiled novel biomarkers that predict the efficacy of ICIs. He et al[53] retrospectively established CD73 as a prognostic biomarker for immunotherapy. Li et al[54] identified the rs17718883 PD-L1 polymorphism, which induces a P146R mutation, disrupting the PD-1/PD-L1 interaction and suggesting an improved GC prognosis but diminished responsiveness to PD-1/PD-L1 blockade therapy. Zhang et al[55] developed an extracellular vesicle protein expression-based score to prognosticate ICI outcomes in 112 treated patients. The challenge for researchers is now to identify and integrate these emerging biomarkers with established ones to enhance the precision of immunotherapy prognosis.

TME: TME encompasses a complex array of non-neoplastic cells and extracellular components within tumors, including fibroblasts, endothelial cells, neurons, adipocytes, immune cells (both innate and adaptive), extracellular matrix, cytokines, chemokines, growth factors, and extracellular vesicles[56]. The proliferation of malignant cells and their evasion of immune surveillance are heavily influenced by the TME[57]. Immunosuppressive elements within the TME, including regulatory cells and tumor-derived immunosuppressive cytokines, diminish the efficacy of ICIs by dampening the activity of TILs[58,59]. Consequently, the TME is a focal point in GC immunotherapy research, as reflected in the co-cited literature and keyword cluster volcano plots (Figures 8 and 9B). Given the prominence of ICIs in immunotherapy, studies are increasingly examining their interaction with the TME. Zhang et al[60] demonstrated that MFSD2A suppresses the COX2-prostaglandin synthesis pathway in GC cells, reducing TNFβ1 secretion and enhancing CD8+ T cell activity, thereby potentiating the antitumor effects of PD-L1 inhibitors. Akiyama et al[61] discovered that PDGFs secreted by tumor cells can induce CAFs to augment chemokine secretion, thereby recruiting PMN-MDSCs and potentially enhancing cancer cell resistance to PD-1 blockade. The heterogeneity of the TME is a determinant of cancer cell susceptibility to immunotherapy, underscoring the utility of TME analysis in the prognosis of immunotherapeutic responses. Chen et al[62] applied digital image analysis, m-IHC, and machine learning to tumor samples from eighty anti-PD-1/PD-L1 treated patients to assess immune cell infiltration and predict treatment efficacy. Jiang et al[63] utilized non-invasive computed tomography imaging and machine learning to evaluate the TME and forecast the therapeutic response of GC to PD-1/PD-L1 inhibitors.

Recognizing the pivotal role of TME in modulating antitumor immunity and its inhibitory effects on ICIs, researchers are increasingly focusing on TME regulation and reprogramming as a therapeutic strategy in immuno-oncology[64]. Jeong et al[65] identified spatial reprogramming within the infiltrative GC TME and proposed CCL2 as a potential immunotherapeutic target, as evidenced by single-cell RNA sequencing of tumor and adjacent normal tissues, corroborated by in situ hybridization and IHC. Jin et al[66] provided significant evidence for the role of the inhibin β subunit (INHBB) in tumor cell proliferation, migration, and invasion, and identified activin B as a critical factor in the differentiation of GC-associated fibroblasts. These insights suggest new potential targets for GC immunotherapy. Given the TME's high heterogeneity, research into TME and immunotherapy has garnered considerable interest (Figure 7B); however, most studies have been confined to cellular or animal models. Consequently, further clinical trials are essential to validate the feasibility of these immunotherapeutic targets.

The concurrent administration of chemotherapy and immunotherapy has become the standard initial treatment for advanced GC, supported by evidence demonstrating that TME modulation contributes to the efficacy of chemoimmunotherapy. An et al[67] reported findings from a phase II, single-arm trial assessing the impact of combining 5-FU/platinum chemotherapy with pembrolizumab on the TME in advanced GC patients. Using sequential tumor biopsies and single-cell RNA sequencing, significant alterations in TME cellularity and transcriptomic profiles were noted, potentially enhancing post-chemotherapy immune responsiveness in a subset of patients. Further large-scale clinical trials are warranted to determine how these insights can inform clinical practice and facilitate the development of TME-based personalized therapeutic strategies.

Keywords

The keyword clustering volcano plot (Figure 9B) identifies "expression", "tumor microenvironment", and "target therapy" as dominant themes. While "expression" was a significant topic from 2000 to 2012, its prominence has declined relative to "tumor microenvironment" and "target therapy," indicating a shift in research priorities from the direct effects of gene expression on GC immunotherapy to enhancing the efficacy of such treatments[41]. The keyword burstness graph (Figure 10B) highlights "gastroesophageal junction", "adoptive immunotherapy" and "regulatory T cells" as keywords with significant burst strength, suggesting their potential to remain focal points in upcoming research.

Target therapy: The suboptimal immune response, inadequate intra-tumoral drug delivery, and TME-mediated immunosuppression limit the clinical efficacy of monotherapy immunotherapies. Consequently, combination therapies that include immunotherapeutic agents are gaining traction in oncological practice[68]. Notably, the synergy of immunotherapy and targeted agents is a burgeoning area of interest. A phase II trial investigated the frontline regimen of pembrolizumab combined with trastuzumab for HER2-positive esophageal, gastric, and gastroesophageal junction cancers, employing an open-label, single-arm design[69]. In a non-randomized, single-center phase II trial Kwon et al[70] evaluated the effects of treating 31 patients with advanced GC with a combination of durvalumab and ceralasertib, featuring a disease control rate, median PFS, overall ORR, and overall survival of 22.6% (95%CI: 9.6% to 41.1%), 58.1% (95%CI: 39.1% to 75.5%), 3.0 months (95%CI: 2.1 to 3.9), and 6.7 months (95%CI: 3.8 to 9.6), respectively. The CAMILLA phase 1b study conducted in 2023 appraised the therapeutic potential of combining cabozantinib with durvalumab in treating pMMR/MSS gastroesophageal and other gastrointestinal tumors[71]. The interaction between targeted therapy and immunotherapy extends beyond the clinical assessment of drug interactions. Immune responses are modulated by numerous signaling pathways, suggesting that the targeting of specific pathways could be a potential immunotherapeutic strategy[72]. Xing et al[73], found that GC cell lines with elevated human leukocyte antigen I (HLA-I) expression were refractory to natural killer (NK) cell-mediated lysis. Downregulation of HLA-I expression was observed to significantly enhance NK cell recruitment, thereby reducing tumor burden. You et al[74] engineered CD47-high expressing vesicles modified with cyclic arginine-glycine-aspartic acid (cRGDyK) to deliver small interfering RNA targeting YTH N6-methyladenosine RNA binding protein 1 (YTHDF1), effectively regulating YTHDF1 expression in tumor cells and regulating immune responses. Miao et al[75] demonstrated that miR-375 mediates PD-L1 phosphorylation in GC cells via hsa_circ_0136666, thereby facilitating immune evasion, highlighting the role of circRNA and miRNA in immune checkpoint modulation. Wang et al[76] identified that PUM1, an RNA-binding protein, directly interacts with NPM3 mRNA, enhancing its interaction with NPM1 and promoting NPM1 nuclear localization, which in turn upregulates PD-L1 transcription. This body of research underscores the multifaceted regulation of GC cell immune responses, including transcriptional, translational, and epigenetic mechanisms. Despite the experimental nature of current findings, there is a pressing need to translate such insights into clinically relevant, targeted therapeutic strategies.

Gastroesophageal junction: The global incidence and mortality rates of GC have decreased over the past decade, largely due to the widespread application of Helicobacter pylori eradication therapy, public health measures, and improvements in therapeutic and diagnostic technologies[1]. Nevertheless, the incidence of gastroesophageal junction adenocarcinoma (GEJA) is on the rise[77]. The unique anatomical particularity of GEJA has spurred ongoing debates about its optimal management, especially concerning surgical approaches[78]. Precision medicine's evolution has made the accurate classification of cancer subtypes and the tailored selection of therapeutic strategies crucial in clinical oncology[45]. Given GEJA's status as a distinct subtype of gastric and esophageal cancers, with increasing research focus on its immunotherapy (Figure 10B).

According to the American Joint Committee on Cancer, GEJA is managed similarly to esophageal or GC[79]. Specifically, tumors within 2 cm (including 2 cm) below the esophagogastric junction (EGJ) that involve the EGJ, are staged using the esophageal adenocarcinoma TNM criteria, while those more than 2 cm below are classified under GC staging guidelines. Consequently, GEJA is often included in clinical trials for both esophageal and GCs. For example, the phase 3 KEYNOTE-062 trial[80] evaluated 763 treatment-naive patients with either GEJA or GC and PD-L1 CPS ≥ 1, comparing pembrolizumab monotherapy, pembrolizumab combined with oxaliplatin, and oxaliplatin alone. Another phase 2 multicenter trial[81] explored the use of sintilimab with chemoradiotherapy as a neoadjuvant treatment for locally advanced gastric or GEJA cancers. With the recent classification of GEJA under both gastric and esophageal cancer categories, research is pivoting towards the efficacy of treatments traditionally used for these cancers in the context of GEJA. There is an anticipation that novel histological and genomic subtyping will further the development of immunotherapies specifically for GEJA.

Adoptive immunotherapy: Known also as ACT, adoptive immunotherapy entails extracting a patient's immune cells, expanding and modifying them outside of their body, and then reintroducing them to the patient's system. Specific targeting and elimination of malignant cells is possible by the modified immune cells, thus achieving the goal of treating tumors[82]. Three main types of ACT include TILs, TCR-T, and CAR-T[82]. In recent years, propelled by the notable success of CAR-T cells in addressing hematologic malignancies[83], researchers have shifted their attention towards adapting CAR-engineered technology for the modification of various immune cells, such as macrophages, NK cells, γδ T cells, and NK T cells[84,85]. Accompanying this is the use of ACT treatment for solid tumors, such as GC, which has emerged as a focal point of current research. Naoyuki Sakamoto et al[86] progressively expanded NK cells ex vivo using recombinant human fibronectin fragment-induced T-cells (RN-T cells), and the enlarged cells demonstrated cytotoxicity in vitro. Additionally, human peritoneal macrophages were engineered by Dong et al[87] to express HER2-FcεR1γ-CAR (HF-CAR), and in a mouse model of HER2-positive peritoneal cancer, these HF-CAR macrophages showed inhibitory effects on tumor formation, providing a possible means of extending the survival of those suffering from peritoneal recurrence of HER2-positive GC. In a prospective experiment, sixty-three patients with advanced GC were examined by Qiao et al[88] for both overall survival and disease-free survival. Compared to patients treated with S-1 plus oxaliplatin alone, those treated with S-1 plus oxaliplatin combined with dendritic cell-cytokine-induced killer cells demonstrated a longer overall lifespan and survival without illness.

While adoptive immunotherapy has demonstrated advantages in exerting anti-tumor effects, its clinical efficacy encounters challenges, particularly concerning tumor infiltration. This is notably apparent in the context of limited effectiveness observed in solid tumors[10]. Subsequent investigations ought to center on devising more efficient methods for overcoming immunological inhibition found in the surrounding TME. Enhancing the ability of adoptive cells to combat cancers in solid forms and support the feasibility of adoptive immunotherapy in clinical trials.

Regulatory T cells: Foxp3-expressing regulatory T cells (Tregs), characterized by CD25 and CD4, are pivotal in constraining excessive immune responses, preserving immunological equilibrium, and averting autoimmune pathogenesis[89,90]. These cells subdue immune reactions through several pathways; notably, CTLA4 on Tregs outcompetes CD28 for binding to CD80/CD86 on antigen-presenting cells (APCs), thereby dampening APC-induced activation of effector T cells[91]. Treg cells contain a high-affinity IL-2 receptor comprised of subunits α (CD25), β (CD122), and γ (CD132), which enables them to bind IL-2 more efficiently than effector T cells, leading to a TME that is more concentrated in Treg cells than in effector T cells[92]. They also secrete immunosuppressive cytokines including TGF-β, IL-10, and IL-35[93-95]. They attenuate effector T cell activity by converting ATP/ADP to adenosine by ectoenzymes CD39 and CD73[96]. Through these mechanisms, Tregs foster an immunosuppressive TME, shielding tumors from immunologic assault[97]. Consequently, strategies to counteract Treg-mediated immunosuppression are gaining prominence in advanced GC immunotherapy[98].

Key signaling pathways are implicated in the recruitment of regulatory Tregs into the TME. Kumagai et al[99] found that mutations in the RHOA gene activate the PI3K-AKT-mTOR pathway in GC cells, increasing free fatty acid production. Tregs preferentially metabolize these free fatty acids over effector T cells, leading to a higher Treg/effector T cell ratio within the TME. Combining PI3Kβ inhibitors with ICIs can potentiate the anti-tumor effects of ICIs. Additionally, the Wnt/β-catenin pathway influences the GC TME; CCL28, a β-catenin target gene, recruits Tregs and promotes GC progression when β-catenin is activated. Downregulating β-catenin or CCL28 can impede Treg infiltration and tumor growth[100]. Song et al[101] discovered that miR-192-5p facilitates Tregs differentiation, highlighting the regulatory role of miRNAs in Treg function.

The majority of patients exhibit limited responses to anti-PD-1 therapy, sparking a surge in interest towards enhancing immunotherapy efficacy. Current research is intensively focused on the role of Tregs and other immune constituents within TME that modulate the effectiveness of effector T lymphocytes in tumor immunity. Given the pivotal position of Tregs in shaping immune responses, future studies on these cells could unveil novel therapeutic avenues for GC immunotherapy.

Our study acknowledges inherent limitations. Firstly, the literature review was restricted to the WoSCC database, potentially constraining the scope of data sources and introducing a bias towards the literature indexed therein. Secondly, the selection and evaluation of articles were predominantly influenced by their citation impact, potentially overlooking other significant factors that could affect the analysis. Thirdly, the assessment of article impact favors older publications, which have had more time to accumulate citations, potentially skewing the impact evaluation when based solely on citation metrics. Future research should employ a comprehensive and multifaceted approach to synthesize, ensuring a more accurate representation of the landscape of GC immunotherapy research.

CONCLUSION

Conclusively, this investigation employed bibliometric techniques to provide a comprehensive and lucid overview of advancements in GC immunotherapy. The field is moving towards clinical application, with a heightened focus on optimizing immunotherapeutic efficacy. As insights into tumor immune mechanisms deepen, the tumor immune microenvironment has emerged as a prominent area of research. Current research is concentrated on strategies to surmount the immunosuppressive TME, integrating immunotherapy with targeted therapies, adaptive immunity, the immunomodulatory functions of regulatory T cells (Tregs), and the precise prognostication of immunotherapy outcomes. Moreover, the application of immunotherapy in gastroesophageal junction cancer has gained significant research attention in recent years.