Correction: A preliminary exploration of the mechanism of endometrial cancer development induced by intestinal flora diversity and its metabolites

Abstract

Correction on: Sun Y, Li J, Chen J, Peng H, Shao Y and Wu Z (2026) A preliminary exploration of the mechanism of endometrial cancer development induced by intestinal flora diversity and its metabolites. Front. Cell. Infect. Microbiol. 16:1815291. doi: 10.3389/fcimb.2026.1815291 Adding/removing text [The author affiliation superscript marker for Juan Li and Zhenzhen Wu were incorrect.]. A correction has been made to the section [Yuer Sun2 †, Juan Li1, Jiayu Chen2 †, Haochen Peng2, Yawen Shao1 and Zhenzhen Wu2*]: "[Yuer Sun 2 † , Juan Li 1 † , Jiayu Chen 2 † , Haochen Peng 2 , Yawen Shao 1 , Zhenzhen Wu 1, 2,* ]" The original version of this article has been updated.[Redundant proofreading text remained in the Introduction section.]. A correction has been made to the section [1 Introduction, Paragraph 1]: "[As one of the three primary malignancies affecting the female reproductive system, endometrial cancer (EC) imposes a substantial global health burden, with 2020 estimates indicating approximately 417, 000 new cases and 97, 000 deaths (Sung H et al.,2021). Epidemiological data from the American Cancer Society(ACS) demonstrate that the annual incidence of EC rose by 1.2% from 2005 to 2014, accompanied by a distinct shift toward younger patient populations (Smith RA et al., 2018). Geographically, China accounts for a disproportionate share of this global burden-representing 23.7% of incident cases and 30.2% of deaths (Sung H et al., 2021)-which underscores the disease's status as a major public health crisis. Pathologically, atypical endometrial hyperplasia (AEH) serves as a critical premalignant precursor, with up to 40% of cases concurrently presenting with or progressing to well-differentiated endometrioid adenocarcinoma (Abu-Rustum N et al., 2023). Given that these conditions share overlapping etiological risk factors, identifying precise biomarkers for the early detection and risk stratification of these precursor lesions remains paramount to improving clinical outcomes.]" The original version of this article has been updated.[Redundant proofreading text remained in the Introduction section.]. A correction has been made to the section [1 Introduction,Paragraph 2]: "[Over the past decade, the cancer biology paradigm has expanded to acknowledge the host-microbiome continuum, wherein gut microbiota dysbiosis critically modulates systemic oncogenesis (Gupta S et al., 2019;Wang J et al., 2018). Within gynecological oncology, a compelling body of evidence links the gut microbiome to endometrial cancer (EC) via the "estrobolome"-the aggregate of enteric bacterial genes specialized in estrogen metabolism (Sepich-Poore GD et al., 2021;Cuevas-Ramos G et al., 2010). Extensive replication cohorts have established that microbial-derived β-glucuronidases deconjugate excreted estrogens within the intestinal lumen, driving their reabsorption into the systemic circulation and thereby enhancing long-term estrogen bioavailability (Yu S et al., 2026;Mattavelli E et al., 2026). This persistent hyperestrogenic state fuels the uncontrolled proliferation of endometrial epithelial cells, anchoring the classic microbial-estrogen-carcinogenesis axis. Beyond this endocrine pathway, emerging evidence suggests that the gut microbiota exerts direct, multiomic regulatory control over the endometrial microenvironment. Recent framework integrations indicate that microbial-derived bioactive metabolites-specifically short-chain fatty acids (SCFAs) and secondary bile acids-transcend the gastrointestinal tract to induce localized epigenetic modifications and immune reprogramming within the endometrium (Wang J et al., 2018). Consequently, the scientific consensus has shifted from viewing the gut microbiota merely as a passive estrogen regulator to recognizing it as an active orchestrator of host metabolic and inflammatory homeostasis.]" The original version of this article has been updated.[Redundant proofreading text remained in the Introduction section.]. A correction has been made to the section [1 Introduction,Paragraph 3]: "[Paradoxically, translating these mechanistic insights into non-invasive diagnostic modalities remains heavily impeded by discordant data and geographical variations among existing cohorts. Although certain multi-ethnic studies highlight discrete taxonomic signatures associated with endometrial lesions (Taira T et al., 2015;Bernstein C et al., 2011), the directional shifts of pivotal candidate biomarkers-particularly Bacteroides and Lactobacillus species-exhibit pronounced inconsistency across diverse patient populations (Baker JM et al., 2017). These discordant findings stem largely from unadjusted interindividual confounders, such as localized dietary habits, body mass index (BMI), and ethnicity, which exacerbate microbial heterogeneity and obfuscate the definitive etiological trajectory of endometrial carcinogenesis. Consequently, a validated, universally applicable microbial-metabolic signature capable of robustly differentiating EC from healthy controls has yet to be established.]" The original version of this article has been updated.[Redundant proofreading text remained in the Introduction section.]. A correction has been made to the section [1 Introduction,Paragraph 4]: "[To address these pivotal knowledge gaps, the present study couples 16S rRNA gene sequencing with untargeted serum metabolomics to comprehensively delineate the gut microbiota-host metabolic interactome across a clinical continuum spanning healthy controls and endometrial cancer (EC) patients. By decoding the compositional and functional dynamics synchronized with endometrial lesion progression, we aimed to resolve the precise taxonomic perturbations and metabolic dysregulation landscapes underlying oncogenesis. Ultimately, this rigorous multi-omics integration seeks to reconcile existing cross-cohort inconsistencies, thereby offering robust, translational biomarkers for the early detection and risk stratification of endometrial cancer.]" The original version of this article has been updated.[Redundant proofreading text remained in the Introduction section.]. A correction has been made to the section [1 Introduction,Paragraph 5,6,7 and8]: "[These paragraphs should be removed.]" The original version of this article has been updated.[Redundant proofreading text remained in the Materials and methods section.]. A correction has been made to the section [2 Materials and methods, 2.1 Research subjects, Paragraph 1]: "[Following institutional review board (IRB) approval and the acquisition of written informed consent, a clinical cohort was prospectively recruited at Gansu Maternal and Child Health Hospital (Gansu Provincial Central Hospital) between February and August 2023. The disease cohort comprised patients diagnosed with pathologically confirmed atypical endometrial hyperplasia (AEH) or endometrioid adenocarcinoma in the Department of Gynecology. The control group consisted of healthy female volunteers undergoing routine health screenings during the same period. To ensure unbiased selection among the 82 eligible candidates, a computer-assisted stratified random sampling approach was executed utilizing the SPSS random number generator, stratified by clinical grouping (EC vs. healthy controls). The finalized study cohort comprised 41 female participants, consisting of 17 patients in the EC group and 24 in the healthy control group. Eligibility and Exclusion Criteria. To be eligible for inclusion in the disease group, patients were required to meet the following criteria: (1) age 18-75 years; (2) diagnosis of AEH according to the WHO classification, or pathologically confirmed endometrioid adenocarcinoma staged in accordance with the 2009 FIGO criteria post-standard hysterectomy. Healthy controls were required to have matching demographic profiles and no history of uterine or gynecological pathology. Candidates were excluded if they presented with any of the following confounding conditions: (1) a history of other concurrent malignancies or hormone-related systemic disorders; (2) current pregnancy or lactation; (3) prior history of oncological interventions, including chemotherapy, radiotherapy, or immunotherapy; (4) recent administration (within 3 months) of antibiotics, probiotics, immunosuppressants, or hormonal therapies; (5) active or recent gastrointestinal disorders (e.g., chronic diarrhea) within 3 months; or (6) specialized or restrictive dietary habits, such as strict vegetarianism.]" The original version of this article has been updated. "[Prior to surgery or during routine health examinations, fresh fecal specimens (≈ 1.5-3.0 g) were aseptically collected from patients and healthy volunteers using sterile instruments. These specimens were immediately transferred into 7.0-mL sterile centrifuge tubes and rapidly cryopreserved at -80°C. Concurrently, peripheral blood samples (2.5 mL) were obtained via venipuncture and delivered to the clinical laboratory for processing. To isolate serum, the whole blood was centrifuged at 4, 000 rpm (approximately 2, 000×g) for 5 min at room temperature. The resulting serum supernatant was subsequently aliquoted into duplicate labeled cryovials, cataloged with unique identification barcodes, and stored at -80°C pending downstream multi-omics analyses.]" The original version of this article has been updated. "[Library Preparation and High-Throughput Sequencing. Genomic DNA samples (30 ng) of verified quality were mixed with matching fusion primers to optimize PCR amplification parameters. Following amplification, the PCR products were purified using Agencourt AMPure XP magnetic beads and eluted in elution buffer to complete library construction. The fragment size distribution and concentration of the resulting libraries were assessed using an Agilent 2100 Bioanalyzer. Validated libraries that passed quality control based on their insert sizes were subsequently subjected to high-throughput sequencing. Sequence Quality Control and Filtration. Raw sequencing reads were filtered to remove low-quality sequences using a sliding-window approach. Specifically, a 25-bp window was established; if the average quality score within the window dropped below 20, the downstream bases were truncated from that position. Sequences were completely discarded if the truncated read length was less than 75% of the original length. Furthermore, reads contaminated with adapters (defined as a default adapter sequence overlapping by 15 bp with the reads, allowing up to 3 mismatches), reads containing ambiguous bases ("N"), and low-complexity sequences (defined as ≥ 10 consecutive identical bases) were excluded. Samples were demultiplexed based on unique primers and barcodes, permitting zero mismatches between barcode sequences and sequencing reads. Bioinformatics and Statistical Analysis. Taxa exhibiting significant shifts in relative abundance during treatment were identified using the DESeq2 package (version 1.12.4). Significantly differentially abundant taxa were screened based on False Discovery Rate (FDR) correction, with thresholds set at |log2fold-change| > 2.0 and P < 0.05. Subsequently, linear discriminant analysis effect size (LEfSe) software was employed to estimate the effect size of each microbial component on the differential variance, where a log-linear discriminant analysis (LDA) score > 3.0 was defined as statistically significant.]" The original version of this article has been updated. "[Metabolite Extraction and Mass Spectrometry Detection. Extracted metabolites were analyzed using a high-resolution mass spectrometry system operated in both positive and negative electrospray ionization modes. Peak extraction, based on the precursor mass-to-charge ratio (m/z) and retention time of the metabolites, was executed using XCMS software to identify specific metabolic features, with primary chromatographic peak areas integrated for relative quantification. Metabolite Identification and Annotation Confidence. Initial metabolite identification was performed using the open-source metaX software by matching experimental m/z values against the Human Metabolome Database (HMDB) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases. To address the inherent challenge of isobaric compounds and structural isomers within these repositories-which frequently result in a single m/z value corresponding to multiple candidate molecules-tentative annotations were rigorously validated. This structural confirmation was achieved by cross-referencing the secondary mass spectrometry (MS/MS) fragmentation patterns of the sample metabolites with reference MS/MS spectra from authentic chemical standards, thereby ensuring a high of and across and quality control samples were via Statistical quality control was using low-quality (defined as in > of samples or > of were filtered and The values were subsequently the was using to the metabolite identification and relative The original version of this article has been updated. data and downstream were executed a quality control metabolic with values in samples or in experimental samples were excluded. The values were using the To control for sample data was performed via utilizing the peak ratio of the samples as the reference identify differential metabolites, was with Specifically, metabolites were defined as if they concurrently the following strict criteria: a in the value > from the a < by the and a > or < functional and integration were subsequently using the Kyoto Encyclopedia of Genes and Genomes (KEGG) To the identified differential metabolites were within specific a was with a established at a < The original version of this article has been