DOI: 10.1155/ijog/7538825 ISSN: 2314-436X

Cross‐Cohort Transcriptomic and Oral Microbiome Analyses Identify Ferroptosis‐Associated Signatures in Periodontitis and Oral Squamous Cell Carcinoma

Chenguang Qiu, Peng She, Lin Wu

Background

Periodontitis (PD) and oral squamous cell carcinoma (OSCC) have been investigated in relation to inflammation, oral microbial alterations, and iron/redox biology. Whether ferroptosis‐associated expression patterns overlap across independently sampled periodontal and cancer cohorts remains uncertain. We characterized these patterns without presuming that the available datasets represent matched individuals or a temporal disease sequence.

Methods

We analyzed gingival single‐cell RNA‐sequencing data (GSE171213), an OSCC discovery expression cohort (GSE30784), and an independent oral tongue squamous cell carcinoma validation cohort (GSE13601). Periodontal microbial functional gene profiles (GSE54290) and a cross‐study OSCC 16S abundance resource were examined separately. Analyses included cell annotation, KEGG ferroptosis gene‐set scoring, sample‐ and cell type‐level comparisons, differential expression, external transcriptomic signature validation, microbial diversity and abundance analyses, and LPS‐stimulated cell‐line RT‐qPCR and ELISA measurements. The transcriptomic and microbial cohorts were not linked at the individual‐patient level.

Results

Quality control retained 34,342 cells from 12 gingival samples (four healthy controls [HCs], five PD, and three treated periodontitis [PDT]), assigned to 11 annotated cell populations. Mean sample‐level ferroptosis gene‐set scores were 0.081 in HC, 0.114 in PD, and 0.076 in PDT; these descriptive scores do not measure ferroptotic cell death. SAT1 showed a positive PD‐versus‐HC association in cell type pseudobulk analysis (log2 fold change, 0.546; false discovery rate [FDR], 0.012) and positive effects across all three transcriptomic cohorts. A discovery‐derived ferroptosis‐associated OSCC signature achieved an area under the receiver operating characteristic curve of 0.933 in GSE13601. Periodontal functional profiling identified 25 advanced PD‐associated annotations (FDR < 0.05), including lower ferrochelatase signals. Cross‐study OSCC microbiome analysis found 293 diagnosis‐associated features after project adjustment; project‐held‐out classification yielded an AUC of 0.674. Following lipopolysaccharide stimulation, GPX4, SLC7A11, and FTH1 increased and ACSL4 decreased at the measured mRNA and protein levels in gingival fibroblasts and THP‐1‐derived macrophages.

Conclusions

Independently sampled periodontal and OSCC cohorts show ferroptosis‐associated transcriptomic changes, with SAT1 emerging as a cross‐cohort candidate. Parallel microbial datasets identify disease‐associated functional gene and taxonomic differences but do not demonstrate within‐patient host–microbe interactions or causal mediation. LPS‐exposed cell‐line measurements show expression changes in selected ferroptosis‐associated markers, not ferroptotic cell death or functional resistance. The results are exploratory associations; in the absence of direct ferroptosis assays, paired longitudinal cohorts, and prospective clinical validation, they cannot establish a disease‐progression mechanism, clinically deployable biomarker or therapeutic efficacy.

More from our Archive