The clinical and translational perspectives on the lung microbiome in interstitial lung diseases: a bibliometric review Authors: Mengshi Hang, Yuchen Liu, Xiao Shen, Yuhui Zhao, Yongdong Xu, Xinyi Gong, Linlin Xu, Na Li, Lei Dong Published: 2026 ABSTRACT BackgroundIncreasing evidence suggests that microbiota plays important roles in the pathogenesis and progression of interstitial lung diseases (ILDs). However, the global research landscape and emerging trends in this field remain insufficiently characterized. This study aimed to systematically characterize the research landscape, evolving hotspots, and future trends in the field of host microbiota and ILDs using bibliometric and visualization approaches, and to further explore the progress of related clinical studies.MethodsPublications up to November 8, 2025 were retrieved from the Web of Science Core Collection. Concurrently, clinical trials within the same timeframe were extracted from PubMed to assess advancements in the field. Bibliometric and visual analyses were conducted using VOSviewer, CiteSpace, SCImago Graphica, and Microsoft Excel.ResultsA total of 295 publications were included, showing a marked increase in research output since 2012. China and the United States were the leading contributors, with the United States demonstrating higher academic impact and stronger international collaboration. Core institutions and authors were mainly concentrated in North America and Europe. Keyword analysis revealed a clear evolution of research focus, shifting from early exposure-related studies and hypersensitivity pneumonitis to lung microbiome dysbiosis, the gut-lung axis, and metagenomic approaches. Recent hotspots emphasize microbiome-based clinical applications, with increasing attention to host-microbiome interactions and immune regulatory mechanisms.ConclusionsResearch on microbiota and ILDs has expanded rapidly and shows increasing interdisciplinary integration. Future studies should enhance international collaboration, clarify underlying mechanisms, and promote clinical translation of microbiome-based biomarkers and personalized therapeutic strategies. Highlight box Key findings • This study provides the first comprehensive bibliometric review of microbiome research in interstitial lung diseases (ILDs). Since 2012, the field has expanded rapidly, with research hotspots shifting toward lung microbiome dysbiosis, the gut-lung axis, metagenomics, and microbiome-related clinical translation. What is known and what is new? • Alterations in lung and gut microbiota are increasingly associated with ILD pathogenesis, immune dysregulation, disease progression, and prognosis, particularly in idiopathic pulmonary fibrosis (IPF). Advances in sequencing technologies have challenged the concept of the “sterile lung” and highlighted the importance of host-microbe interactions. • This study provides the first systematic, data-driven overview of global microbiome research in ILDs. It maps publication trends, leading contributors, collaboration patterns, and emerging research frontiers, while integrating bibliometric findings with available clinical evidence. The analysis highlights the growing importance of lung microbiome ecology, the gut-lung axis, metagenomic approaches, and microbiome-based biomarkers. What are the implications, and what should change now? • Future research should move beyond predominantly cross-sectional and descriptive designs toward longitudinal and mechanistic studies. Integrating pulmonary and intestinal microbiome data with host immune parameters may clarify causal roles in disease progression, improve patient stratification, and support microbiome-guided precision therapies in ILDs. Introduction Interstitial lung diseases (ILDs) comprise a heterogeneous group of pulmonary disorders characterized by inflammation and/or fibrosis of the lung parenchyma. Their etiology is complex and encompasses multiple subtypes, including idiopathic interstitial pneumonia, sarcoidosis, and hypersensitivity pneumonitis. Patients commonly present with progressive dyspnea, diffuse infiltrates on chest radiography, declining pulmonary function, and hypoxemia, and those with end-stage disease often die from respiratory failure (1). Approximately 30–40% of patients with ILDs develop progressive pulmonary fibrosis, for whom the median survival is only 2.5–3.5 years, indicating a generally poor prognosis (2). To date, the pathogenic mechanisms underlying ILDs have not been fully elucidated. Genetic mutations and polymorphisms (3), dysregulated immune and inflammatory responses (4,5), and fibroblast proliferation (6) have all been implicated in the initiation and progression of ILDs. In recent years, with the rapid development and widespread application of high-throughput technologies, host-associated microbiota has emerged as a highly active and rapidly expanding research area within the ILDs field. A growing body of evidence indicates that dysbiosis of the lung (7,8) and gut microbiota (9,10), viral or bacterial infections (11,12), and environmental exposures (13,14), as exogenous or endogenous microbial-related factors, play important roles in the development and progression of ILDs, including idiopathic pulmonary fibrosis (IPF) (15), hypersensitivity pneumonitis (16), and connective tissue disease-associated ILDs (17,18). Given the rapid expansion of literature in this interdisciplinary field, a comprehensive synthesis of existing research is essential to clarify its intellectual structure, identify influential contributors, and delineate evolving research priorities. Such an overview is also of practical importance for guiding future mechanistic studies and facilitating clinical translation of microbiome-related discoveries. Bibliometric analysis applies quantitative methods to evaluate academic literature and its associated metadata, enabling systematic assessment of publication trends, collaboration patterns, and knowledge evolution within a research domain (19). Although studies examining the microbiome in ILDs have increased markedly over the past decade, a structured bibliometric evaluation of this field remains limited. Therefore, the present study employs bibliometric and visualization techniques to comprehensively map the research landscape, collaborative networks, and emerging hotspots in studies of the host microbiome and ILDs. We present this article in accordance with the BIBLIO reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0369/rc). Methods Data collection and processing A comprehensive literature search was conducted on November 8, 2025, using the Web of Science Core Collection (WoSCC) and PubMed databases. The detailed search strategy is shown in Figure 1 and Appendix 1. In the WoSCC database, a total of 704 records were initially retrieved. To ensure data quality and comparability, non-English publications, non-English reviews, and articles irrelevant to the research topic were excluded. Only documents classified as article or review were included. Specifically, the following document types were excluded: book chapters (n=3), early access articles (n=3), letters (n=9), editorial materials (n=23), meeting abstracts (n=50), proceedings papers (n=9), notes (n=2), non-English publications (n=25), and topic-irrelevant records (n=285). After screening, 295 publications were retained for analysis, including 212 original articles and 83 review articles. All records were exported in full record and cited references format. In the PubMed database, five microbiome-related clinical studies conducted during the same period were identified. One study was excluded due to irrelevance to the research topic, and the remaining four clinical studies were included to illustrate the progress of microbiome-related clinical research in ILDs. These records were exported in PubMed format. Flowchart of data acquisition and processing. COVID-19, coronavirus disease 2019; TS, topic. The literature screening process was independently performed by two researchers. Any discrepancies were resolved through discussion, and when necessary, a third investigator was consulted to reach a final decision. Bibliometric analysis and visualization Bibliometric data processing and visualization were performed using Microsoft Excel 2024, VOSviewer (version 1.6.20), CiteSpace (version 6.3 R3), and SCImago Graphica, to comprehensively depict the research landscape of this field. Microsoft Excel 2024 was used to summarize the annual publication output up to November 8, 2025, revealing long-term publication trends. VOSviewer, developed by van Eck and Waltman (20), was employed to construct network maps of international collaboration, co-authorship, journal co-citation, and keyword co-occurrence. Using CiteSpace, developed by Synnestvedt and colleagues (21), analyses of institutional collaboration, keyword burst detection, and keyword clustering with timeline visualization were conducted, and the results were integrated into visual knowledge maps. SCImago Graphica was applied to illustrate the annual publication output of the most productive countries. Results Overview of publication trends A total of 295 eligible publications related to the microbiome and ILDs were retrieved from the Web of Science Core Collection up to November 8, 2025. As shown in Figure 2, research in this field dates back to 1980, with minimal output before 2012 (≤5 articles annually), indicating that research during this period was largely exploratory. Publication trends in studies on the association between ILDs and the microbiome. Bars represent the annual number of publications, and the line represents the cumulative number. The x-axis indicates the year; the left y-axis shows annual publications, and the right y-axis shows cumulative publications. ILD, interstitial lung disease. Over the past decade, the field has entered a phase of rapid expansion. Publication volume has increased markedly, accounting for 75.59% of all studies. This rapid growth is likely associated with the widespread adoption of high-throughput sequencing technologies and the paradigm shift away from the long-held concept of the “sterile lung” (22,23). In the most recent five years, annual publication output has remained consistently high, reaching a peak of 41 articles in 2024—1.52 times higher than in 2023. Although data for 2025 include only the first 11 months, 39 articles have already been published, suggesting sustained academic interest and continued growth. Overall, research focus has evolved from early descriptive studies to investigations of functional roles, host-microbiome interactions, and disease progression, highlighting the lung microbiome as an active participant in ILDs rather than a passive colonizer. Analysis of countries A total of 38 countries contributed to research on the relationship between the microbiome and ILDs. Table 1 presents the top 10 countries ranked by publication volume. China ranked first in publication volume (n=96), followed by the United States (n=78), while the United States led in academic impact (4,536 citations; 58.15 per article), far exceeding China. European countries such as the United Kingdom (n=26) and Switzerland (n=8) produced fewer studies but achieved higher average citations, with the United Kingdom reaching 77.85 per article. Table 1Top 10 countries by number of publicationsRankCountryDocumentsCitationsAverage citationsTotal link strength1China961,41614.75102USA784,53658.15383Japan301,00633.53124United Kingdom262,02477.85245France1433523.9336Italy1119717.9197Poland1149945.3668Spain1153248.3689Germany936941.00410Switzerland841051.259 Top 10 countries by number of publications Figure 3 shows that country-level collaboration forms three major clusters in North America, East Asia, and Western Europe. The United States and United Kingdom occupy central collaborative positions, whereas China, despite its high output, shows a less dense network. In contrast, Western European countries—including Germany, France, Italy, the Netherlands, and Spain—demonstrate strong regional collaboration and form closely connected research clusters. These patterns reflect early technological and infrastructural advantages in North America and Western Europe, while China’s rapid growth but limited collaboration suggests the need for stronger international partnerships to enhance research quality and global visibility. Country collaboration network map. Each node represents a country; the size of the node reflects the publication volume, and thicker links between nodes indicate more frequent collaboration. Institutional distribution Based on CiteSpace analysis, research on the microbiome in ILDs involves 313 institutions worldwide. The top 10 institutions by publication output are listed in Table 2, and Figure 4 illustrates the institutional collaboration network. Institutions from Europe and North America entered the field earlier and remain dominant. Notably, Imperial College London (n=18), Royal Brompton & Harefield NHS Foundation Trust (n=17), and the University of Michigan (n=11) demonstrate both high research productivity and strong inter-institutional collaboration, reflecting well-established research networks. Table 2Top 10 institutions by number of publicationsRankInstitutionsCountCentralityYear1Imperial College London180.0120032Royal Brompton & Harefield NHS Foundation Trust170.0220033University of Michigan11020144CHU Besancon11020015Universite Marie et Louis Pasteur11020016National Jewish Health80.0120067Chinese Academy of Sciences70.0620178Institute of Science Tokyo7020099University of California70.1201110University of Colorado70.012006 Top 10 institutions by number of publications Institutional collaboration network map. In contrast, institutions from the Asia-Pacific region have increased their output in recent years. The Chinese Academy of Sciences, which began publishing in 2017, has produced seven studies; however, its collaborations remain largely domestic. Given that microbiome research in ILDs is an emerging and highly interdisciplinary field, robust global collaboration is essential for integrating multi-omics data, clinical resources, and standardized methodologies. Therefore, strengthening international partnerships may be a critical step for institutions in the Asia-Pacific region to bridge existing gaps in research integration, promote cross-cohort data sharing, and enhance their overall contribution to the global research landscape. Author collaboration analysis Based on the VOSviewer analysis of the author collaboration network, a total of 1,861 authors have published research on the microbiome in ILDs, among whom 57 authors with three or more publications constitute the core contributor group (Figure 5). As shown in Table 3, the most productive authors are mainly affiliated with internationally active research teams, including Philip L. Molyneaux (n=16) and Toby M. Maher (n=12) from Imperial College London, and David N. O’Dwyer (n=8) from the University of Michigan. These authors demonstrate both high productivity and strong international collaboration. In contrast, many prolific authors from France, Japan, and China, such as Laurence Millon, Yoshinobu Eishi, and Qiyue Jiang, tend to collaborate primarily within their own countries, forming relatively concentrated national networks, possibly due to language, research systems, or regional funding priorities. Notably, Professor Gary B. Huffnagle from the University of Michigan has an average of 205 citations per publication, indicating exceptional academic influence and a foundational role in this field. Taken together, these leading author groups have played a central role in driving the transition of ILD microbiome research from descriptive profiling toward mechanism-oriented and translational studies. Author collaboration network of authors with ≥3 publications. Table 3Authors with the largest publication volumeRankAuthorDocumentsCitationsAverage citationsTotal link strength1Molyneaux, Philip L.161,13270.75402Maher, Toby M.121,210100.83363Millon, Laurence811414.25274O'Dwyer, David N.858873.50215Reboux, Gabriel812115.13286Eishi, Yoshinobu735851.1467Huffnagle, Gary B.61,232205.33268Invernizzi, Rachele633756.17199Martinez, Fernando J.61,079179.8331 Authors with the largest publication volume Analysis of journals Table 4 summarizes the most productive journals in microbiome research on ILDs. A total of 76 articles (25.76%) were published in these journals, most of which are ranked in Journal Citation Reports (JCR) Q1. Leading respiratory journals, including the American Journal of Respiratory and Critical Care Medicine (IF =19.4) and the European Respiratory Journal (IF =21.2), published 10 and 4 articles, respectively, reflecting strong interest from high-impact platforms. The American Journal of Respiratory and Critical Care Medicine had the highest output and citation counts, highlighting its leading role and influence. Although The Lancet and The Lancet Respiratory Medicine each published only one article, they showed the highest average citations, indicating their importance as key knowledge sources. The bibliographic coupling network (Figure 6) included 53 journals and formed four clusters, with strong links among core journals such as American Journal of Respiratory and Critical Care Medicine, Frontiers in Medicine, and Respiratory Research, suggesting a concentrated knowledge structure. The journal distribution demonstrates a multidisciplinary field spanning respiratory medicine, immunology, and microbiology, and reflects the field’s evolution from early pathogen-centered studies toward contemporary investigations of microbial dysbiosis, host-microbe interactions, and gut-lung axis mechanisms. Table 4Journals with the largest publication volumeRankSourceDocumentsCitationsAverage citationsTotal link strengthJCRIF1 American Journal of Respiratory and Critical Care Medicine 101,243124.32,235Q119.42 Frontiers in Cellular and Infection Microbiology 9717.891,570Q24.83 Frontiers in Immunology 7392561,292Q15.94 Frontiers in Medicine 714620.862,125Q13.05 Frontiers in Microbiology 710214.57813Q14.56 International Journal of Molecular Sciences 618230.332,077Q14.97 Respiratory Research 527454.81,392Q15.08 Scientific Reports 555111,285Q13.99 American Journal of Industrial Medicine 413934.75444Q13.110 European Respiratory Journal 429373.251,060Q121.211 Frontiers in Pharmacology 44511.2529Q14.812 Journal of Immunology 4431107.75833Q23.413 Sarcoidosis Vasculitis and Diffuse Lung Diseases 45413.5261Q31.8IF, impact factor; JCR, Journal Citation Reports. Journals with the largest publication volume IF, impact factor; JCR, Journal Citation Reports. Journal bibliographic coupling network map. Node size represents the publication volume, and link thickness reflects the coupling strength between journals. Keyword analysis Keyword burst analysis This study employs the bibliometric analysis tool CiteSpace to perform burst detection on keywords. The method is based on Kleinberg’s burst detection algorithm, which analyzes the frequency variations of keywords across different time slices to identify the evolution of research hotspots. The time span is set from 1991 to 2025, with a time slice of 1 year. The node type is set to “Keyword”, and the node selection follows the Top N strategy (Top N=20). Burst detection is used to identify significant growth in keywords during specific time periods, while burst strength is employed to measure their level of activity. As shown in Figure 7, early burst keywords, including hypersensitivity pneumonitis, farmer’s lung, extrinsic allergic alveolitis, and exposure, were mainly active before 2010. This indicates that early studies on the ILD microbiome were primarily exposure- and pathogen-oriented. During this stage, microorganisms were mainly conceptualized as external antigens or pathogenic factors, and hypersensitivity pneumonitis, due to its relatively clear exposure-disease relationship, became a typical model. This reflects an “exposure-driven” research paradigm, in which microbial factors were studied mainly in the context of environmental exposures rather than as intrinsic components of disease biology. Top 20 keywords with the strongest citation bursts. The burst keywords are listed on the left, and the timeline (1991–2025) is displayed on the right. The blue line represents the overall time span, while the red segments indicate the periods during which keywords exhibit bursts. Burst strength reflects the degree of increase in attention to a keyword during a given period; the higher the value, the stronger the burst. Between 2009 and 2017, the emergence of keywords such as infection, pathogenesis, IPF, and terms related to immunity and genetics (e.g., T cells, MUC5B promoter polymorphism) reflected a shift toward mechanism-oriented research. During this period, the interactions among microbial factors, host immune responses, and genetic susceptibility received increasing attention, especially in IPF. This stage marks a key transition from descriptive associations to hypothesis-driven studies, with the microbiome being explored as a potential regulator of disease onset and progression rather than merely a coexisting factor. In recent years, emerging keywords such as lung microbiome, gut-lung axis, and gut microbiome have further promoted a paradigm shift toward a systems-level understanding of ILD. These emerging topics reflect the growing recognition that microbial communities not only affect the local lung environment but also participate in systemic immune regulation via the gut-lung axis. This shift indicates that ILD research is moving from organ-specific mechanisms toward an integrated network of host-microbiome interactions. Importantly, these frontier topics are closely linked to translational applications, including microbiome-based biomarkers, risk stratification, and potential therapeutic interventions, suggesting a shift from mechanistic exploration to clinically actionable research. Overall, the development of burst keywords shows a clear trajectory from exposure-driven research to mechanistic studies, and more recently, to integrative and translational research. This progress highlights the increasingly important role of the microbiome as a biological driver and potential clinical target in ILD. Keyword co-occurrence analysis Keyword co-occurrence analysis was conducted using VOSviewer. Among 626 extracted keywords, 68 keywords with a minimum co-occurrence frequency of three were included to construct the co-occurrence network (Figure 8). The top 10 most frequent keywords (Table 5) were primarily related to ILD subtypes—such as IPF, sarcoidosis, and hypersensitivity pneumonitis—as well as microbiome-related concepts, including gut microbiota, lung microbiota, and gut-lung axis. Top 20 keywords with the strongest citation bursts. Table 5The top 10 most frequently occurring keywordsRankKeywordOccurrencesTotal link strength1Idiopathic pulmonary fibrosis41652Pulmonary fibrosis33453Microbiome32574Gut microbiota31375Sarcoidosis24346Interstitial lung disease23397Hypersensitivity pneumonitis21108Microbiota18289Lung microbiota172610Gut-lung axis1632 The top 10 most frequently occurring keywords These high-frequency keywords exhibited strong interconnections within the network, indicating a tightly integrated knowledge structure. The results demonstrate that current research has expanded beyond descriptive associations to focus on the role of pulmonary and intestinal microbiota across different ILD subtypes. In particular, immune modulation mediated by microbial communities and its impact on disease progression has emerged as a central theme within this research domain. Keyword timeline analysis Keyword clustering and temporal evolution were further analyzed using CiteSpace, yielding a modularity value of Q =0.5705 and a mean silhouette score of S =0.8049, indicating robust and reliable clustering results. The timeline visualization (Figure 9) illustrates the developmental trajectory of major research themes. The largest cluster was labeled IPF, confirming IPF as the core research focus within the field. Other prominent clusters included Swedish farmers and gut-lung axis. From a temporal perspective, respiratory microbiome represented the most recently formed cluster, followed by metagenomic and polymorphisms, suggesting increasing adoption of metagenomic approaches and host genetic analyses in recent years. Importantly, clusters such as IPF, gut-lung axis, and metagenomic displayed dense node distributions in the later stages of the timeline, reflecting sustained research activity and rising academic attention. These findings indicate that these themes not only represent current hotspots but are also likely to remain key drivers of future research and theoretical advancement in the microbiome-ILD field. Keyword timeline visualization. Colored horizontal lines indicate keyword clusters; nodes represent individual keywords. The horizontal position of each node indicates the year when publications containing the keyword first appeared, and curve length reflects the duration of keyword occurrence. Cluster labels are shown at the right end of each line. Clinical progress analysis Four clinical trials retrieved from PubMed (Table S1) reflect the evolving understanding of microbiome-related mechanisms in ILD, from host-microbe interactions to clinical outcome-oriented studies. Early studies, such as those on sarcoidosis, have shown that polymorphisms in the TLR10-TLR1-TLR6 gene cluster are associated with disease progression, and the loss of protective haplotypes increases the risk of chronic progression, suggesting that the host’s response to microbial signals may affect the prognosis of ILD. With the advent of high-throughput sequencing technologies, a prospective cohort study employing 16S rRNA sequencing analyzed bronchoalveolar lavage fluid (BALF) from 55 patients with IPF to explore associations between the lung microbiome and disease progression. The results showed that Streptococcus operational taxonomic unit (OTU) 1345 (>3.9%) and Staphylococcus OTU 1348 (>1.8%) were significantly associated with disease progression (hazard ratio 10.19 and 5.06, respectively). These findings suggest that specific microbial taxa may serve as biomarkers of IPF progression and may contribute to disease pathogenesis. This provides a potential rationale for antibiotic-based therapeutic strategies. However, causality remains to be established in larger interventional studies. Subsequently, several small randomized trials suggested potential benefits of antibiotic therapy in IPF. However, a phase III randomized controlled trial, which randomly assigned 342 patients to receive oral co-trimoxazole (960 mg twice daily) or placebo, failed to confirm these findings. The primary composite endpoint—time to all-cause mortality, lung transplantation, or first non-elective hospitalization—showed no significant difference between groups. Secondary outcomes, including lung function and quality of life, also showed no significant improvement. Notably, this study did not assess the lung microbiome, and the effects of co-trimoxazole on its composition and ecology were not quantified. Therefore, despite observed associations between microbiota and ILD, translating these findings into effective clinical interventions remains challenging. In addition, an exploratory clinical study in patients with locally advanced non-small cell lung cancer undergoing concurrent chemoradiotherapy evaluated the effects of thymosin α1 on radiation pneumonitis and gut microbiota dynamics. Although dynamic changes in gut microbial diversity were observed during treatment in patients who developed radiation pneumonitis, no significant differences were detected in overall community structure or genus-level composition. The study indicates potential crosstalk among thymosin α1, the gut microbiota, and host immunity, which needs to be validated in larger-scale studies. Collectively, these studies demonstrate a clear association between microbiota-related factors and disease progression and prognosis in ILDs; however, the clinical benefits of antibiotic or other microbiome-targeted interventions remain to be further validated in well-designed trials. Discussion General Information This study provides a comprehensive overview of the global research landscape on microbiome and ILDs. Overall, the field has undergone a clear transition from a slow-growth phase to rapid expansion after 2012, largely driven by advances in high-throughput sequencing technologies and the evolving recognition of the lung microbiome as an active participant in disease processes rather than a passive bystander. From a global perspective, research activity is characterized by geographic imbalance and clustered collaboration patterns. North America and Western Europe occupy central positions in terms of academic impact and international collaboration, whereas countries such as China have demonstrated rapid growth in publication output but relatively limited global integration. This disparity suggests that strengthening cross-national collaboration may be essential for improving research quality and influence. At the institutional and author levels, the field is shaped by several highly collaborative research groups that have driven its conceptual evolution from descriptive studies toward mechanistic and translational research. Similarly, the concentration of publications in high-impact respiratory and multidisciplinary journals reflects both the clinical relevance and the interdisciplinary nature of this field. Collectively, these patterns indicate that microbiome research in ILDs is transitioning toward a more integrated, collaborative, and clinically oriented research paradigm. Hotspots and frontiers By integrating keyword co-occurrence, burst detection, thematic evolution, and clinical analysis, this study identified key hotspots and trends in host microbiome and ILD research. Early studies, limited by technology, focused on infections, environmental exposure, and specific pathogens, lacking community-level insights. With advances in sequencing, research has shifted toward microbiome dysbiosis, emphasizing ecological complexity and greater translational potential. Respiratory microbiome in ILD progression: from dysbiosis to mechanistic and prognostic insights Bibliometric analysis in the present study shows that research on ILD and the microbiome has entered a phase of rapid expansion since 2012. Concurrently, keywords have shifted from “infection” and “exposure” toward “lung microbiome”, “idiopathic pulmonary fibrosis”, and “disease progression”, indicating a fundamental change in the research paradigm. Respiratory microbes are no longer viewed as isolated pathogens but as a dynamic ecosystem involved in disease initiation and progression. This shift has been primarily driven by technological advances. The application of high-throughput sequencing, particularly 16S rRNA sequencing, to BALF samples has enabled culture-independent characterization of lung microbial communities, directly challenging the long-standing concept of the “sterile lung”. BALF-based studies consistently show that while different ILD subtypes share similarities at the phylum level, they differ markedly at the genus level. For instance, enrichment of Veillonella and Actinomyces has been reported in IPF, increased Corynebacterium in granulomatous diseases, and dominance of Staphylococcus in chronic hypersensitivity pneumonitis (CHP) (24-26). These findings suggest that microbiome alterations in ILD reflect ecological restructuring rather than simple infection, thereby shifting the focus from classical pathogen-centered models to a microbial ecology framework. Building on this, several clinical cohort studies have examined associations between microbiome features and disease progression. In the COMET-IPF study, 16S rRNA sequencing of BALF demonstrated that enrichment of Streptococcus and Staphylococcus was significantly associated with disease progression and independently predictive of outcomes (27). Similarly, Huang et al. integrated BALF microbiome data with peripheral blood transcriptomics and found that reduced microbial diversity and enrichment of specific taxa, including Streptococcus and Staphylococcus aureus, were associated with shorter progression-free survival (28). Beyond community composition, microbial burden has emerged as another key dimension (29). Molyneaux et al. reported a significantly higher bacterial load in BALF from patients with IPF compared with healthy controls (12). In contrast, Invernizzi et al., analyzing BALF samples from 110 patients with CHP, 45 with IPF, and 28 controls using 16S rRNA sequencing and quantitative PCR, found that bacterial burden was associated with survival only in IPF but not in CHP (25). Previous studies have also shown that bacterial load is significantly higher during acute exacerbations of IPF than in stable disease (30). In a prospective cohort of 193 patients with IPF, lower airway bacterial burden quantified by 16S rRNA qPCR was significantly associated with disease progression and mortality, independent of radiologic fibrosis extent, suggesting that bacterial burden may serve as an independent predictor of disease progression and death (31). Together, these findings have established quantitative microbial metrics as an emerging focus in the field. Keyword co-occurrence analysis further indicates a growing emphasis on mechanistic studies in recent years. Evidence from animal models supports the clinical associations described above. In bleomycin-induced pulmonary fibrosis models, germ-free mice exhibited significantly lower mortality than conventional mice, accompanied by a reduction in pulmonary Th1 cells and an increase in regulatory T cells, suggesting that the lung microbiome contributes to disease progression through modulation of local immune responses rather than merely acting as a bystander (32). Notably, microbial dysbiosis precedes the peak of lung injury and persists thereafter, supporting a potential causal role in disease progression (32). In addition, specific microbes and their products can directly drive tissue injury and fibrosis. For example, Staphylococcus aureus secretes the pro-apoptotic peptide corisin, which induces alveolar epithelial cell apoptosis and promotes collagen deposition (33). Klebsiella-associated microbes can induce mitophagy in macrophages and enhance TGF-β1 release, thereby promoting fibroblast activation and excessive collagen accumulation (34). In summary, the emergence of the respiratory microbiome as a central focus in ILD research reflects the combined influence of technological innovation, clinical demand, and a shift in conceptual frameworks. The field has moved from demonstrating microbial presence to elucidating functional relevance and prognostic value. Regulatory role of the gut-lung axis in ILDs: from microbial dysbiosis to intervenable signaling networks Keyword analysis in this study shows that “gut microbiome” and the “gut-lung axis” have gained sustained prominence in recent years and formed relatively distinct research clusters, indicating a shift from peripheral topics to a recognized research frontier. This trend reflects a broader transition in ILD research, from a lung-centered perspective toward an integrated, system-level regulatory framework. Clinically, accumulating evidence suggests that gut microbial dysbiosis is common in ILD. In IPF, sarcoidosis, and connective tissue disease-associated ILD, reductions in beneficial taxa (e.g., Lactobacillus) and enrichment of potentially pathogenic bacteria (e.g., Corynebacterium, Enterobacteriaceae) have been reported (35-38). Epidemiological data further show that approximately one quarter of patients with IPF have coexisting gastrointestinal disorders (39), providing additional clinical support for gut-lung interactions. The emergence of this research focus has also been driven by advances in both conceptual frameworks and methodologies. The “common mucosal immune system” concept offers a mechanistic basis for cross-organ immune regulation, whereby intestinal immune status can influence pulmonary inflammation (40-43). Experimental studies further demonstrate that reduced gut microbial diversity exacerbates pulmonary fibrosis via IL-6/STAT3/IL-17A signaling, whereas a more diverse microbiota—particularly with enrichment of beneficial taxa such as lactobacilli—attenuates these pathways and confers protective effects (44). Recent work has increasingly linked the gut-lung axis to microbial metabolites and inflammatory regulation. Gut microbiota-derived metabolites, especially short-chain fatty acids (SCFAs), play key roles in anti-inflammatory and antifibrotic processes. Mechanistically, SCFAs such as butyrate suppress inflammatory macrophage differentiation and infiltration, inhibit HDAC3 activity while promoting histone acetylation, and downregulate profibrotic gene expression. They may also exert antifibrotic effects through modulation of YBX1 ubiquitination (45,46). In contrast, dysbiosis can reduce SCFA production and impair intestinal barrier integrity, facilitating lipopolysaccharide (LPS) translocation, which promotes fibroblast activation and excessive extracellular matrix deposition (47,48). Circulating LPS can further exacerbate pulmonary inflammation via the IFN-γ/STAT1/STAT3 signaling pathway (49). Together, these findings position the gut microbiome not merely as an associated factor, but as a potential therapeutic target. From a bibliometric perspective, the rise of the gut-lung axis as a research frontier underscores a paradigm shift in ILD research—from single-organ models toward systems-level, integrative biology. Clinical translation progress in ILDs microbiome research: challenges from association to precision intervention With growing mechanistic evidence, microbiome research is moving from basic exploration toward clinical translation. Bibliometric analyses highlight increasing interest in the clinical value of microbiome features in ILDs, particularly for risk prediction and therapeutic decisions. Prospective cohort studies linking lung microbiome characteristics with disease progression and adverse outcomes suggest potential value for prognostic stratification (27). However, most evidence remains observational, supporting associations rather than causality, and the mechanisms by which microbial alterations drive ILD onset and progression require further validation. In this context, some studies have attempted to incorporate microbiome information into clinical stratification and decision-support frameworks. A dysbiosis index (DI) constructed based on whole-genome sequencing can partially differentiate ILD patients from healthy individuals and may provide a potential stratification basis for unclassifiable ILDs (8). Furthermore, metagenomic next-generation sequencing (mNGS) applied to BALF is considered capable of providing novel insights for the clinical management of acute exacerbations in ILD and may offer information beyond conventional pathogen detection, supporting individualized therapeutic decisions (50). These efforts suggest that microbiome research could complement traditional imaging and pulmonary function metrics in early risk assessment and disease progression prediction. On the therapeutic front, given the limited efficacy and notable side effects of existing antifibrotic drugs (51), and based on the hypothesis that dysbiosis influences ILD progression, increasing research has attempted to intervene in the microbiome via antibacterial, probiotics, and other approaches aimed at restoring pulmonary or gut microbial homeostasis to improve ILD outcomes (52). The role of antibiotics, however, remains controversial. While earlier studies suggested that agents such as azithromycin or co-trimoxazole may improve quality of life, enhance forced vital capacity (FVC), and reduce disease burden in fibrotic ILD (53-55), more recent high-quality randomized controlled trials have failed to confirm these benefits. A multicenter randomized trial including 513 IPF patients showed that adding co-trimoxazole or doxycycline to standard care did not reduce the risk of respiratory-related hospitalization or mortality, and adverse event rates increased (56). Similarly, a double-blind, placebo-controlled study found no significant benefit of long-term co-trimoxazole on survival, hospitalization, lung function, or patient-reported outcomes (57). Taken together, although early studies suggested that antimicrobials may improve some metrics by reducing infections, untargeted broad-spectrum antibiotics without microbiome stratification are unlikely to offer sustained benefit. Therefore, future research should prioritize microbiome-guided precision interventions to evaluate antimicrobial or immunomodulatory therapies in ILDs (29). In contrast, precision approaches may offer greater therapeutic potential. Patients with higher bacterial burden or enrichment of specific pathogenic taxa may represent more suitable candidates for microbiome-targeted interventions (58). For example, neutralizing antibodies against corisin, a pro-apoptotic peptide secreted by Staphylococcus aureus, have shown potential effects in alleviating pulmonary fibrosis and acute lung injury in animal models (59). Along the gut-lung axis, modulation of the gut microbiota through probiotics or fecal microbiota transplantation (FMT) has also been suggested as a novel strategy to mitigate pulmonary fibrosis (44,60). Moreover, evidence suggests that whether microbial stimuli translate into persistent inflammation and fibrotic responses largely depends on host immune response patterns (61). Wang et al., in a prospective study of IPF patients, investigated the effects of inhaled interferon-γ on the microbiome and host immunity. However, they did not observe a significant impact of inhaled interferon-γ on lower respiratory tract microbial diversity. Notably, the relative abundance of oral-derived taxa (e.g., Prevotella, Veillonella, Streptococcus, Rothia) correlated with local and systemic Th1, Th2, Th17, and fibrosis-related cytokines. This association remained significant even after excluding the influence of inhaled interferon-γ treatment, highlighting the critical regulatory role of host immune tone in linking microbial exposure to fibrotic outcomes. These findings suggest that the lung microbiome may independently influence host immune status, thereby potentially affecting disease progression in IPF and serving as a promising therapeutic target (62). In summary, bibliometric trends indicate a shift from descriptive associations to mechanism-oriented and clinically translational research in ILDs; however, significant challenges remain in bridging microbiome-related insights with effective therapies. Future studies should prioritize causal inference, multi-omics integration of microbiome and host immunity, and microbiome-guided precision interventions to improve patient stratification and therapeutic outcomes in heterogeneous diseases such as IPF. Limitations Although this study was conducted with methodological rigor, several limitations should be acknowledged. First, all data were derived exclusively from the WoSCC and the PubMed database. While this approach ensures data quality, it may have resulted in the omission of relevant publications indexed in other databases, thereby limiting the comprehensiveness of the literature coverage. Second, only English-language publications were included, excluding potentially relevant studies published in other languages, which may introduce language bias and affect the representativeness of the findings. Third, as only a limited number of eligible clinical studies were identified from PubMed, the analysis of clinical progress was restricted in scope. Finally, as both the WoSCC and PubMed are continuously updated, citation counts and available publications may change over time. Therefore, the bibliometric results presented in this study reflect a specific time point. Conclusions Microbiome research in ILDs has grown rapidly since 2012, with the US and UK leading in influence, China producing the most publications, and key institutions including Imperial College London and the University of Michigan. Influential authors such as Molyneaux and Huffnagle, and leading journals like American Journal of Respiratory and Critical Care Medicine, have shaped the field. Research hotspots have shifted from environmental exposure to lung microbiome dysbiosis, the gut-lung axis, and microbiome-based clinical applications. Although accumulating evidence links microbiota to disease progression and prognosis, effective microbiome-targeted interventions remain to be established. Future studies should focus on mechanistic validation, international collaboration, and clinical translation. Supplementary The article’s supplementary files as Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0369/prf