大腸がんにおける歯周病と口腔病原体：系統的レビュー、メタアナリシス、およびトライアル逐次解析

Background: Periodontitis and oral dysbiosis have been linked to systemic inflammation and carcinogenesis. Among oral pathogens, Porphyromonas gingivalis (Pg) and Fusobacterium nucleatum (Fn) are biologically plausible contributors to colorectal cancer (CRC) via inflammatory and immunomodulatory pathways. However, the magnitude and consistency of these associations remain uncertain. Objective: To evaluate whether periodontitis and key oral pathogens are associated with CRC risk and prognosis through a systematic review, meta-analysis, and trial sequential analysis (TSA). Methods: We searched PubMed, Scopus, and Web of Science from inception to December 2024 following PRISMA 2020. Eligible observational studies assessed periodontitis exposure or detection of oral bacteria in relation to CRC incidence or survival. Effect estimates (RRs/HRs) were log-transformed and pooled using random-effects models; heterogeneity was quantified with I². TSA was conducted to appraise information size and the stability of the primary association. Risk of bias was evaluated with ROBINS-I/QUIPS as appropriate. PROSPERO: CRD420251168522. Results: Five studies evaluating periodontitis/oral-pathogen exposure and CRC incidence yielded a 70% higher risk (HR = 1.70; 95% CI: 1.33-2.19; I² = 0%). Detection of Fn was associated with approximately threefold higher risk of CRC (RR = 3.20; 95% CI: 1.76-5.82; p < 0.001). Pg presence was linked to worse overall survival (HR ≈ 2.4; p < 0.01). TSA suggested that the accrued evidence for the primary incidence association is likely sufficient to reduce random errors; nevertheless, interpretability is constrained by the small number of observational studies and between-study differences in exposure and outcome ascertainment. Conclusions: Current evidence indicates that periodontitis and oral pathogens-particularly Fn and Pg-are significantly associated with CRC development and progression. These findings support the clinical relevance of the oral-gut axis and underscore oral health as a potentially modifiable factor in cancer prevention. Further large, well-designed prospective cohorts and mechanistic studies are warranted to strengthen causal inference.