Global research trends in m 6 A demethylases and cancer: a bibliometric and visualization analysis

1 Introduction

Over 170 distinct chemical modifications have been identified in RNA, which broadly regulate post-transcriptional RNA behavior and biological functions (Li et al., 2022). Among these, N6-methyladenosine (m6A) was first discovered in eukaryotic messenger RNA and viral nuclear RNA in the 1970s, and has since been established as one of the most prevalent and abundant RNA modifications (Desrosiers et al., 1974; Wang et al., 2020). m⁶A plays a pivotal role in regulating RNA splicing, translation, stability, and degradation (Sun et al., 2023). The dynamic equilibrium of m6A modification is orchestrated by three classes of proteins: writers (methyltransferases, e.g., METTL3/METTL14), erasers (demethylases, e.g., FTO and ALKBH5), and readers (m⁶A-binding proteins, e.g., the YTHDF family) (Zeng et al., 2020; Guan et al., 2022). Together, these regulators constitute a complex m⁶A regulatory network that influences diverse cellular physiological and pathological processes. Among the various RNA modifications, m⁶A has attracted considerable attention due to its extensive functional repertoire and dynamic reversibility. In particular, the discovery of m⁶A demethylases has revealed the dynamic and reversible nature of m⁶A modification in gene expression regulation. By reversing m⁶A modification, demethylases exert distinct regulatory functions in tumor biological behaviors, including tumor cell proliferation, resistance to apoptosis, invasion and metastasis, maintenance of cancer stemness, and remodeling of the tumor microenvironment.

Although substantial progress has been made in elucidating the biological functions and molecular mechanisms of m⁶A demethylases in cancer, the overall developmental trajectory, international collaboration landscape, distribution of core research institutions, and the evolution of scientific themes in this field remain to be systematically characterized. Traditional narrative reviews are often constrained by authors’ subjective preferences, making it difficult to provide a comprehensive and objective depiction of a discipline at a macro level. Bibliometrics, an interdisciplinary field grounded in mathematics and statistics, enables a systematic and impartial evaluation of knowledge domains. By mining latent information from large-scale literature datasets, it can objectively reveal the developmental trajectory, Frontier hotspots, and future trends within a specific research field. This study aims to delineate the key milestones in the development of m⁶A demethylase research in oncology and to identify current research hotspots and emerging trends. Through this panoramic analysis, we seek to provide researchers with insights into international collaboration, topic selection, and disciplinary integration, thereby facilitating the clinical translation of m⁶A demethylase-related findings and offering a new perspective for the innovation of cancer treatment strategies.

2 Materials and methods

2.1 Retrieval strategy

The Web of Science (WoS) is a comprehensive academic database platform and one of the most authoritative tools for literature retrieval and analysis worldwide. It encompasses high-quality academic literature across the natural sciences, social sciences, and arts and humanities, and provides robust bibliometric analysis capabilities. We retrieved relevant publications from the WoSCC database and established a dataset for bibliometric analysis. Detailed Medical Subject Headings (MeSH) terms for tumors are provided in Supplementary Table S1; notably, no MeSH term currently exists for m⁶A demethylase. To minimize the risk of misclassification and improve data purity, the 761 records found in the initial search were all filtered. Two reviewers (R1 and R2) independently reviewed the titles and abstracts of these 761 records. The records with differences are resolved by R1 and R2 through discussion; If no agreement can be reached, the third reviewer (R3) will make a decision. All the records that passed the title/abstract screening, as well as those whose qualifications were in doubt, were evaluated in full. After full-text evaluation, 761 records all meet the requirements. There is no so-called’ validation’ subset because no separate validation samples are used (Pitr et al., 2023).To prevent errors arising from database updates, all data were searched and exported on 25 August 2026. To minimize potential analytical bias, the following inclusion criteria were applied: (1) publications written in English; (2) document types limited to articles and reviews; and (3) publication date ranging from 1 January 2015 to 31 December 2025. After applying these criteria, a total of 761 publications were ultimately obtained for in-depth analysis. The detailed literature screening flowchart is shown in Figure 1 and Supplementary Table S2. In addition, the PubMed database was supplemented and cross-database sensitivity analysis was carried out. See Supplementary Table S2 for specific retrieval. Unlike WoSCC, PubMed offers more granular indexing of clinical study types and trial designs, enabling more precise identification of clinical research (Cheng et al., 2026).

2.2 Data analysis

Multidimensional quantitative analyses were performed using established bibliometric tools. The bibliometrix package (version 4.6.1) for R is an open-source package developed by Massimo Aria and Corrado Cuccurullo for bibliometric analysis. It offers comprehensive functionalities to process and analyze literature data retrieved from major scientific databases (e.g., Web of Science, Scopus, PubMed) and to generate various statistical and visual outputs. Additional analyses were conducted using the online platform Bibliometric.com, Scimago Graphica, CiteSpace (version 6.4.R1), and VOSviewer (version 1.6.20). CiteSpace, a Java-based citation visualization software developed by Chaomei Chen, enables statistical analysis and transforms raw data into visual representations of literature networks (Chen, 2004); it was used to analyze keywords and references with significant burst strength. VOSviewer, a bibliometric mapping software developed by Nees Jan van Eck and Ludo Waltman in 2009 (van Eck and Waltman, 2010), was used to visualize co-authorship and institutional collaborations. By integrating the above software tools, we systematically analyzed the research field (Lei et al., 2025).

3 Results

3.1 Publishing trends and regional distribution characteristics

Ultimately, 761 publications pertaining to m⁶A demethylases and tumors were included in this study, comprising 538 articles and 223 reviews. This field is characterized by substantial interdisciplinary integration, with an average annual growth rate of 50.92% (Table 1). Over the study period, annual publication output exhibited a consistently upward trajectory. Notably, the literature output was limited between 2015 and 2019, whereas 709 publications appeared from 2020 onward, accounting for 93.2% of the total corpus, indicating that this field has emerged as a prominent research focus in recent years. Specifically, only two articles were published in 2015, the lowest annual output, while 2022 and 2023 each saw the highest output, with 137 articles published in each year (Figure 2A).

TABLE 1. General characteristics of the WoSCC database.Compound Annual Growth Rate (CAGR) calculated as: [(N_final/N_initial)^(1/years) - 1] × 100.

Twenty-seven countries or regions have contributed publications to the field of m⁶A demethylases and tumors. China ranked first with 644 articles, accounting for 84.63% of the total 761 publications. Among these Chinese articles, 578 were single-country publications (SCP), i.e., articles produced solely by authors from China without international collaboration, while 66 were multiple-country publications (MCP), i.e., articles co-authored by authors from multiple countries, representing 10.25% of China’s total output (Table 2). The global geographical distribution map further reveals that Asia, North America, and Europe are active research areas in this field, with close inter-regional collaboration (Figure 2B). Although China and the United States dominate in publication output, the United Kingdom (75%) and Belgium (75%) lead globally in terms of international collaboration activity, as measured by the proportion of MCP articles (Figure 2C).

TABLE 2. Countries/regions with Top10 posts and their collaboration rate.

The top three journals in terms of publication output were Frontiers in Oncology (31 articles), Molecular Cancer (26 articles), and Cell Death and Disease (15 articles) (Figure 2D). These leading journals are classified as Q1 or Q2 in the Journal Citation Reports categories and are primarily situated within the fields of oncology and cell biology. The top three institutions with the highest publication output in this field were Sun Yat-sen University (134 articles), Nanjing Medical University (115 articles), and the Chinese Academy of Sciences (85 articles) (Figure 2E).

3.2 Research trajectory

The essence of a dual-map overlay lies in the relationship between the citing and cited domains, and the knowledge flow between disciplines at the journal level can be visualized through the overlay graph. In Figure 3, the left panel represents the primary distribution of citing journals in the field of m⁶A demethylase and tumor research from the WoSCC database, while the right panel shows the main distribution of cited journals. The research in this field is primarily concentrated in journal clusters related to molecular biology, immunology, clinical medicine, physics, and materials science. Analysis of the dual-map overlay reveals three distinct citation pathways in the knowledge dissemination within this field. The first pathway extends from molecular biology and immunology to molecular biology and genetics (yellow track), reflecting the continuity and deepening of knowledge within the discipline, as well as the accumulation of theoretical and mechanistic research. The second pathway extends from clinical medicine and neurology to drug toxicology and treatment (green track), embodying the translational medicine research paradigm driven by clinical questions. The third pathway originates from physics, materials, and chemistry to nursing and medicine (pink track), which highlights the interdisciplinary methodological innovations contributed by the physical sciences and materials engineering to this field, demonstrating the momentum of multidisciplinary collaboration.

3.3 Analysis on the annual publishing output of authors, journals and institutions

From 2015 to 2025, a total of 4,160 scholars worldwide contributed to the field of m6A demethylase and tumor research. According to Lotka’s law, 3,140 scholars (75.5% of the total) published only one paper, while 607 scholars (14.6%) published two or more articles. This distribution indicates a large but highly mobile research community, suggesting that the field remains in a relatively early stage of development, with overall productivity being highly skewed (Figure 4B). Following 2019, the annual count of the top 10 most productive scholars in the field exhibited a marked increase (Figure 4A). LI YANG topped the list with 34 publications, and their H-index (20) and G-index (34) were the highest among all authors, underscoring their substantial academic influence in this field (Table 3).

TABLE 3. The top 10 authors by number of publications.

At the journal level, the annual publication output has increased steadily year by year (Figure 4C). Frontiers in Oncology (the line indicated by the blue arrow) is particularly prominent, ranking first with 31 articles and recording an H-index of 17, a G-index of 29, and 862 total citations. It is followed by Molecular Cancer with 26 articles and the highest H-index (25) (Table 4). Regarding research institutions, Sun Yat-sen University and Nanjing Medical University are among the most productive. The annual publication volume of these institutions has maintained steady growth, especially since 2019 (Figure 4D).

TABLE 4. The top ten core journals in which m⁶A demethylase published papers in the field of tumor research.

3.4 Collaboration network among authors, institutions and countries

Based on publication counts, this study constructed collaboration networks across three dimensions: authors, institutions, and countries/regions, using Total Link Strength (TLS) as a quantitative metric to assess collaboration intensity among these entities. The author collaboration network (Figure 5A) visualizes co-authorship patterns and publication outputs. Applying a minimum publication threshold of five articles per author, 38 authors were included in the network. The visualization reveals several core author collaboration clusters, with Wang Hongsheng, Li Jiexin, and Chen Feng serving as central nodes, thereby constituting key hubs within the author collaboration network in this field.The institutional collaboration network (Figure 5B) illustrates the collaborative connections and the distribution of TLS among core research institutions. Applying a minimum publication threshold of 14 articles per institution, 23 institutions were included in the network. In the network, node sizes represent publication volumes, line thickness indicates collaboration intensity, and different colors distinguish institutional collaboration clusters. The top three institutions in terms of TLS were Sun Yat-sen University (44), the Chinese Academy of Sciences (27), and Nanjing Medical University (23).The country/region co-authorship network (Figure 5C) reflects the landscape of international scientific collaboration. Applying a minimum publication threshold of three articles per country/region, 15 countries were included in the network. In this network, the top two countries ranked by TLS-China (97) and the United States (92)-serve as the two core nodes of the international collaboration network in this field. The most prominent line thickness between these two countries reflects their strong collaborative ties and high frequency of interaction within this research area.

3.5 Research hotspots and Frontier evolution based on keywords

Keywords serve as the core summary of a research article. The top 20 high-frequency keywords are listed in Table 5, including subject keywords such as “RNA Methylation” (45), “m⁶A Modification” (27), and “N6-methyladenosine” (23); disease keywords such as “gastric cancer” (21), “hepatocellular cancer” (21), “colorectal cancer” (20), and “breast cancer” (19); and other keywords such as “tumor microenvironment” (14) and “cancer therapy” (9). These results indicate that gastric cancer, liver cancer, and colorectal cancer are the most extensively studied cancer types in the context of m⁶A demethylase research, with their immune microenvironments and therapeutic interventions representing the primary research foci in this field. The keyword co-occurrence network further corroborates the above findings (Figure 6E).

TABLE 5. The 20 most commonly used keywords.

Keyword co-occurrence analysis (Figure 6A) facilitates the identification of semantic relationships among keywords and extracts latent information from textual data, thereby enhancing the understanding of research content and associated topics. To ensure analytical accuracy, keywords with closely related semantics were clustered and integrated, ultimately yielding 13 major clusters (Figure 6B). These clusters were primarily categorized into three groups: thematic clusters, cancer-type clusters, and epigenetic modification clusters. Based on these clusters, a timeline graph of keywords was constructed (Figure 6C), which illustrates the evolutionary trajectory and temporal distribution of keywords within each cluster. The modularity Q value (Q) and the weighted mean silhouette value (S) were used to evaluate the rationality of the network structure. In this study, Q = 0.8699 (>0.3) and S = 0.969 (>0.7), indicating a significant clustering structure and reliable results. As shown in the visualization map (Figure 6D) depicting the top 20 keywords with the strongest citation bursts, “non-small cell lung cancer” exhibited the highest burst strength (2.3). The temporal evolution of this research field can be divided into distinct developmental stages (Figure 6F), shifting from targeted therapy across various cancer types in the early phase to the identification of diverse m⁶A demethylases in the later phase, with research foci gradually transitioning toward tumor metastasis and prognosis evaluation.

3.6 Citation dynamics and paradigm shift in this field

Citation analysis reveals the evolving knowledge landscape in this field. The most frequently cited document is the review “The Role of m⁶A Modification in Biological Functions and Diseases” published by Xiulin Jiang’s team in Signal Transduction and Targeted Therapy in 2021. The top 20 cited publications were published in high-impact journals, and their total citation counts all exceeded 680 (Figure 7A; Table 6). In the citation burst analysis of the top 30 references (Figure 7B), the article exhibiting the highest burst intensity (25.26) discusses how, under hypoxic conditions, HIF-dependent signaling induces the expression of the m⁶A demethylase ALKBH5, which demethylates NANOG mRNA and promotes its stable expression, thereby facilitating the enrichment of the breast cancer stem cell phenotype (Zhang et al., 2016).

TABLE 6. The top 10 most cited publications.

The journal co-citation analysis revealed a core clustering structure of authoritative journals characterized by interdisciplinary integration. The co-cited journals were mainly concentrated in leading international comprehensive and specialized journals such as Cell, Nature, and Molecular Cancer (Figure 7C), underscoring the high authority of research outputs in this field and the trend toward interdisciplinary convergence.

3.7 The evolution track of the research topic

To systematically analyze the evolution of research hotspots regarding m⁶A demethylases in the field of tumor research, this study employed multiple correspondence analysis to generate a conceptual structure map based on author keywords and indexed keywords, and plotted keyword clusters within a two-dimensional diagram (Figure 8A). The first cluster, marked in blue, mainly comprises common solid tumors such as gastric cancer and lung cancer, and is associated with the tumor microenvironment, immunotherapy, and prognosis. This reflects the heterogeneity of m⁶A modification across different cancer types and its critical role in immune microenvironment remodeling and clinical prognosis. The second cluster, marked in green, focuses on RNA modifications and epitranscriptomics, representing current and future research frontiers. The third cluster, marked in red, shows methyltransferases and demethylases plotted separately in the lower right corner, highlighting the significance of targeting m⁶A regulatory enzymes as emerging anticancer strategies. Topic analysis was performed using author keywords and their interrelationships to identify research themes (Figure 8B). Immunotherapy and the tumor microenvironment were identified as Motor Themes, indicating that these topics are both central and well-developed.

The topic evolution diagram illustrates the temporal distribution and evolutionary trajectory of research themes from 2015 to 2025 (Figure 8C). From 2015 to 2019, research primarily focused on RNA methylation, FTO, m6A, METTL3, demethylases, genetics, as well as cancer progression and metastasis. This indicates that early studies were mainly devoted to elucidating the “writer-eraser” mechanism of m⁶A and its preliminary regulatory effects on the malignant phenotypes of tumor cells. From 2020 to 2023, with the deepening of research, m⁶A and FTO remained central, and were closely associated with N⁶-methyladenosine, genetics, ALKBH5, and immunotherapy. Notably, prognosis and drug resistance emerged as key focal points during this stage, reflecting a transition in research focus from single basic mechanisms to clinicopathologic features and oncologic treatment challenges (e.g., immune escape and drug resistance). From 2024 to 2025, Frontier trends indicate that research will further advance toward clinical translation and precision medicine. Most prominently, the sustained emergence and increasing prominence of immunotherapy suggest that future m⁶A research will prioritize combined strategies involving targeted regulatory enzymes, tumor immunotherapy, and epigenetic precision medicine.

3.8 Cross-database sensitivity analysis based on PubMed data

PubMed and WoSCC are not independent samples, so this part does not constitute independent verification, but only serves as a cross-database sensitivity analysis to test whether the main results of WoSCC change with the database replacement. The results show that the annual number of publications in this field has been increasing, which is largely consistent with the overall growth trend observed in the WoSCC dataset (Figure 9A). At the national level, China and the United States remained the top two contributors (Figure 9B). In terms of journal distribution, Frontiers in Oncology ranked first with 24 articles (Figure 9C). Multiple correspondence analysis (Figure 9D) revealed three distinct keyword clusters. The first cluster, marked in red, focuses on basic experimental research. Its core nodes include mice, animals, nude mice, and cell lines, indicating that in vitro and in vivo experimental models are the primary research approaches in this field. The second cluster, marked in blue, is oriented toward clinical samples and epigenetic regulation. Core nodes include humans, methylation, epigenetics, and messenger RNA. The third cluster, marked in green, features an independent prognosis node, reflecting an independent thematic direction that ultimately extends toward clinical translation and prognosis evaluation. To sum up, PubMed sensitivity analysis is consistent with the results of WoSCC in terms of annual trend, national distribution and theme structure. However, in view of the high degree of overlap between the two databases and the fact that this study has not been included in Scopus, Embase and other databases, the results should be used as a reminder of consistency.

4 Discussion

In this study, bibliometric and visual analysis methods were employed to systematically map the global knowledge landscape of m⁶A demethylases and tumors. The results demonstrate that this field has entered a phase of rapid growth since 2020, with an average annual growth rate exceeding 50%. It has emerged as an active area of investigation at the intersection of epigenetics and cancer research, exhibiting a developmental trend characterized by in-depth exploration of basic mechanisms, accelerating clinical translation, and deep multidisciplinary integration. In terms of the research landscape, China ranks first globally in total publication output, with institutions such as Sun Yat-sen University and Nanjing Medical University forming prominent research clusters, highlighting their substantial research capabilities and domestic innovation potential. Meanwhile, countries such as the United Kingdom and Belgium occupy central positions in the international collaboration network and the output of high-impact publications, contributing to a global research system marked by complementary cooperation between Eastern and Western regions. A stable collaborative network of highly productive scholars and core institutions has been established in this field; however, the overall distribution of researchers remains dispersed, leaving considerable room for further expansion. Analysis of the research trajectory and thematic evolution reveals that knowledge flow in this field radiates from molecular biology to genetics, clinical medicine, and materials science. The research focus has evolved from the early characterization of basic mechanisms involving FTO and ALKBH5 to the clinical translation level, encompassing the tumor immune microenvironment, therapeutic resistance, and prognosis evaluation. Notably, immunotherapy and the tumor microenvironment were identified as Motor Themes, suggesting that precision medicine strategies targeting m⁶A regulatory enzymes in combination with immune checkpoint blockade will represent a prominent focus of future research.

4.1 Concept traceability and foundation period

FTO was first identified as a obesity-susceptibility gene identified through genome-wide association studies, and later confirmed as a member of the Fe(II) and 2- ketoglutarate-dependent dioxygenase family, which can erase the N6 methyl group of RNA adenine, and is the first identified m⁶A demethylase (Azzam et al., 2022; Marcinkowski et al., 2021). This discovery redefines m⁶A from a static apparent marker to a reversible dynamic regulatory layer. From 2015 to 2020, the concept of reversible regulation of m⁶A rapidly extended to oncology, and FTO promoted the progress of acute myeloid leukemia by removing m⁶A modification on MYC, CEBPA and other transcripts (Li et al., 2017); On the other hand, ALKBH5 showed a selective cancer-promoting effect in AML, and its high expression suggested a poor prognosis (Shen et al., 2020)^. However, the functional verification in solid tumors is gradually spread out: ALKBH5 maintains the tumorigenicity of glioblastoma stem cells (Zhang et al., 2017); FTO not only enhances the tumorigenicity of melanoma, mediates the resistance to PD-1 therapy (Yang et al., 2019), but also drives the progression of breast cancer by inhibiting BNIP3 (Niu et al., 2019). In the aspect of treatment transformation: FTO targeted intervention represented by R-2HG has shown anti-tumor activity (Su et al., 2018), suggesting that “eraser” has the potential to become a drug. Overall, the amount of papers published at this stage only accounts for 6.83%. The research content is mainly about molecular identification and function of demethylase.

4.2 Peak period: prognostic value map of m⁶A demethylases based on cancer types with high incidence

Since 2020, the number of publications in this field has increased rapidly, and the research focus has gradually shifted from exploring the basic functions of FTO to the regulation of tumor prognosis, invasion, and metastasis. Gastric cancer, hepatocellular carcinoma, and colorectal cancer have emerged as the core research subjects, which correspond to the medium- and high-frequency disease keywords identified in the analysis. During this stage, studies revealed remarkable cancer-type-specific effects of m⁶A demethylases. In gastric cancer, FTO is highly expressed and promotes tumorigenesis and metastasis by mediating m⁶A modification and mitochondrial dynamic reprogramming of caveolin-1 (Zhou et al., 2022). In hepatocellular carcinoma, downregulation of FTO expression reduces m⁶A demethylase activity, leading to an abnormal increase in m⁶A modification levels of specific oncogenes, which subsequently enhances the invasion and metastasis of hepatocellular carcinoma cells by regulating related signaling pathways (Zhou et al., 2025). Additionally, hypoxia microenvironment-mediated downregulation of FTO has been found to promote tumor metastasis in colorectal cancer (Ruan et al., 2021). Meanwhile, ALKBH5 also exhibits dual regulatory effects. In hepatocellular carcinoma, ALKBH5 post-transcriptionally inhibits the oncogene LYPD1 via m⁶A demethylation, thereby suppressing the proliferation, invasion, and metastasis of hepatocellular carcinoma cells, suggesting that ALKBH5 is an independent prognostic factor for this cancer (Chen et al., 2020). In gastric cancer, ALKBH5 inhibits cancer cell invasion by mediating m⁶A demethylation of PKMYT1 (Hu et al., 2022). In contrast, in colorectal cancer, ALKBH5 stabilizes RAB5A mRNA through a YTHDF2-dependent pathway, subsequently activating RAB5A post-transcriptionally to promote tumor progression (Shen et al., 2023). During this phase, the majority of publications were contributed by Chinese scholars, and international collaborative research remained limited. This is consistent with the quantitative analysis showing that 89.75% of publications from China were SCP. This pattern may be related to factors such as domestic funding priorities, self-sufficiency of large research teams, and language barriers.

4.3 Frontier hotspots

4.3.1 m⁶A demethylase in reshaping the tumor-immunosuppressive microenvironment

As a core epigenetic “molecular switch” reshaping the immunosuppressive tumor microenvironment, m⁶A demethylases have emerged as a prominent research focus in tumor immunology. In colorectal cancer, ALKBH5 is highly expressed and activates the Wnt/DKK1 signaling axis by mediating m⁶A demethylation and degradation of AXIN2 mRNA, thereby inducing an immunosuppressive microenvironment that is closely associated with poor patient prognosis. Targeting this pathway effectively reverses immune escape and enhances the efficacy of anti-PD-1 therapy (Zhai et al., 2023). ALKBH5 also regulates paracrine interactions between tumor cells and tumor-associated macrophages, driving the progression of non-small cell lung cancer and modulating sensitivity to anti-PD-L1 therapy (Hua et al., 2024). Meanwhile, fat mass and obesity-associated protein exerts key regulatory functions. Targeting the FTO/m6A/GPNMB axis inhibits the growth and metastasis of hepatocellular carcinoma and reshapes the immune activation state, highlighting the clinical translational potential of FTO small-molecule inhibitors (Chen et al., 2024). In breast cancer, the hypoxic microenvironment upregulates PD-L1 expression through an FTO-dependent m⁶A/YTHDF3/PDK1 pathway, ultimately mediating an immunosuppressive phenotype (Wang et al., 2025). Current research has clearly moved beyond the independent regulation of single tumor cells to focus on immune-related m⁶A modifications and remodeling, which is consistent with the molecular biology and immunology knowledge domain identified in the dual-map overlay analysis of this study.

4.3.2 m⁶A demethylase-driven therapy for resistance

Overcoming acquired drug resistance remains a major challenge in the precision treatment and clinical translation of oncology. RNA demethylases play a pivotal role in the multi-level regulation of tumor targeted therapy resistance by reshaping the m⁶A modification landscape. FTO diminishes the sensitivity of colorectal cancer cells to 5-FU via the FTO–SIVA1 axis (Lin Z. et al., 2023), and promotes gemcitabine resistance in pancreatic cancer and osimertinib resistance in EGFR-mutant lung adenocarcinoma by reducing m⁶A modification levels (Lin K. et al., 2023; Chen J. et al., 2025). Concurrently, ALKBH5 mediates docetaxel resistance in prostate cancer by regulating CLSPN expression via the ALKBH5–IGF2BP2 axis (Chen Y. et al., 2025), and drives cetuximab resistance in KRAS-mutant colorectal cancer by activating FOXM1 through demethylation (Han et al., 2025); the resulting accumulation of 27HC due to its deletion serves as a microenvironmental basis for lenvatinib resistance and immune escape in hepatocellular carcinoma (Wang et al., 2026). Targeting m⁶A demethylases and their downstream effector molecules represents a potential strategy to reverse tumor drug resistance.

4.4 Future prospects

Based on the evolutionary trends of this field, the global research landscape, and existing knowledge gaps-and integrating the latest advances in translational medicine from the past three years-future research should prioritize three strategic directions: clinical translation, international collaboration, and multidisciplinary integration. First, efforts should be directed toward promoting multi-level clinical translation and establishing a comprehensive precision diagnosis and treatment system. There is an urgent need to conduct multicenter prospective cohort studies to construct tumor recurrence prediction models based on m⁶A modification profiles (Xu X. et al., 2025), thereby providing systematic evidence-based support for RNA methylation as a liquid biopsy biomarker and bridging the gap from basic discovery to clinical application (Xu C. et al., 2025). Second, given the imbalanced pattern in which China leads in publication volume but the international collaboration rate remains low at only 10.25%, it is essential to strengthen global multicenter research cooperation, balance regional research development, and establish standardized public multi-omics data repositories. Third, deepening interdisciplinary integration will expand the boundaries of innovation. By moving beyond a single-discipline paradigm through the integration of materials science (targeted nano-delivery), artificial intelligence (AI; m⁶A prognosis algorithms), and immunology (single-cell microenvironment mapping), we can advance the core objective of serving precision oncology.

4.5 Advantages and limitations

In this study, bibliometric analysis was employed to examine the knowledge structure in this field from a quantitative and macroscopic perspective, thereby revealing the developmental trends and research hotspots of m⁶A demethylases in tumor research. Through multidimensional systematic analysis, this study provides in-depth references for related research. However, several limitations should be acknowledged. The discrepancies observed between the WoSCC and PubMed datasets in terms of journal counts and national publication outputs are primarily attributable to differences in their collection scopes and statistical criteria. WoSCC encompasses all types of academic publications (e.g., research articles and reviews), whereas PubMed places greater emphasis on the screening of clinical medical literature, resulting in partial heterogeneity in the included documents. Because PubMed and WoSCC are highly overlapped, there is no formal cross-database consistency statistical test in this study, and the results of sensitivity analysis should be interpreted descriptively. Furthermore, this study has several limitations, including language restrictions; the exclusion of other databases such as Scopus; citation-timing biases; database indexing discrepancies; dependence on software parameters; potential inaccuracies arising from keyword synonyms; the influence of self-citations; limitations of aggregate statistical methods; and the inability of bibliometric approaches to evaluate experimental quality. Additionally, potential biases inherent in the bibliometric algorithms based on natural language processing may also affect the analytical results.

5 Conclusion

In this study, the research landscape of m⁶A demethylases in the field of oncology was delineated through a bibliometric panorama. This field has experienced rapid growth since 2020. Although China holds a dominant position in publication output, the breadth of international collaboration remains to be further expanded. The research focus has evolved from the early identification of basic mechanisms to the tumor immune microenvironment, therapeutic resistance, and prognosis evaluation. In the future, efforts should be directed toward breaking down disciplinary barriers, deepening international cooperation, and facilitating advances in the precision diagnosis and treatment of tumors.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fgene.2026.1943884/full#supplementary-material

Supplementary material Supplementary material at the source

Supplementary material Supplementary material at the source

Footnotes

Edited by: Qiang Tang, Zhejiang University, China

Reviewed by: Mingliang Liu, Affiliated Hospital of Jining Medical University, China

Fengyu Xu, Zhejiang University, China

Shafee Ur Rehman, International Atatürk-Alatoo University, Kyrgyzstan