Abstract
Background
Non-invasive molecular tools that can both screen for severe periodontitis and quantify treatment-associated pathogen changes remain an unmet need. We evaluated a non-invasive multiplex quantitative polymerase chain reaction (qPCR) mouthwash panel providing absolute colony-forming-unit (CFU) quantification of five periodontal pathogens (
Porphyromonas gingivalis, Treponema denticola, Tannerella forsythia, Prevotella intermedia, Fusobacterium nucleatum
), and asked whether a single test can both screen for severe disease and track each patient’s species-level treatment response.
Methods
Treatment-response monitoring was assessed in a real-world retrospective cohort (349 patients, 703 visits; 235 paired pre/post-treatment samples) using paired Wilcoxon signed-rank tests and Jonckheere–Terpstra trend tests across baseline stage. Screening performance was independently evaluated in a prospective cross-sectional cohort (n=87; 31 healthy, 56 Stage III/IV) with leave-one-out cross-validation (LOOCV) and a pre-specified sensitivity analysis.
Results
For screening, LOOCV AUC was 0.96 for
P. gingivalis
and 0.97 for the combined panel, preserved under inclusion of antibiotic- and treatment-exposed subjects. Beyond screening, the panel resolved each patient’s species-level response to non-surgical therapy: all five species decreased significantly, with median within-patient reductions of −0.37 to −0.72 log
10
CFU (paired Wilcoxon, all pF. nucleatum
reached or fell below the Healthy reference median in 61% of evaluable patients, whereas
P. gingivalis
reached it in only 5% and remained approximately 500-fold above the Healthy reference at the median final visit, identifying residual keystone-pathogen burden not captured by clinical or radiographic assessment.
Conclusion
In this two-cohort evaluation, a single non-invasive mouthwash qPCR test provided two linked clinical outputs: high-accuracy screening for severe periodontitis and quantitative individual-level tracking of species-specific microbial response to therapy. These findings are preliminary, derive from a single-center screening cohort and a retrospective monitoring cohort, and require external validation before clinical deployment. By placing pathogen loads on the same absolute CFU scale, the assay supports both cross-sectional disease detection and longitudinal residual-burden monitoring as an adjunct to the 2018 AAP/EFP framework. Clinical Research Information Service (CRIS) registration: KCT0011511 (
https://cris.nih.go.kr
).