Pyroptosis-Ferroptosis Crosstalk Suggests Candidate Molecular Clusters and Immune Remodeling in Peri-Implantitis

Background: Peri-implantitis is a major biological complication for the long-term stability of dental implants but its molecular heterogeneity and mechanism of programmed cell death are unknown. The present study aimed to elucidate the molecular characteristics and alterations in the immune microenvironment of peri-implantitis from the perspective of pyroptosis-ferroptosis crosstalk. Methods: We retrospectively integrated three public GEO transcriptomic datasets (GSE178351, GSE33774, and GSE57631), comprising 29 peri-implant tissue samples, including 16 peri-implantitis samples and 13 healthy controls. Batch effects were corrected, followed by differential expression analysis, GSEA, GSVA, consensus clustering, machine learning-based exploratory feature prioritization, immune-infiltration estimation, and predicted ceRNA network construction. Results: A total of 2450 disease-associated candidate differentially expressed genes were identified, among which 41 genes were associated with both pyroptosis and ferroptosis. Pathway analysis indicated significant upregulation of inflammatory responses, complement activation, TNF-α/NF-κB, IL-6/JAK/STAT3, reactive oxygen species (ROS) pathways and tissue remodeling-related processes in peri-implantitis tissues. Based on these 41 overlap genes, unsupervised clustering suggested two candidate expression clusters, C1 and C2, in the integrated cohort. C1 was predominantly composed of peri-implantitis samples and showed stronger inflammatory and cellular-stress-related pathway activity. BRAF and TRPV1 were prioritized as exploratory candidate genes and showed associations with estimated immune-infiltration patterns. Conclusions: This exploratory analysis suggests that pyroptosis-ferroptosis crosstalk-related signatures may be associated with immune remodeling in peri-implantitis. BRAF and TRPV1 may serve as candidate genes for future validation, while the C1/C2 clusters should be interpreted as preliminary expression patterns rather than established disease subtypes.