By Dr. Benjamin Caplan, MD | Board-Certified Family Physician, CMO at CED Clinic | Evidence Watch
A 2025 narrative review proposes that periodontitis and diabetes share epigenetic pathways involving DNA methylation and histone modification that worsen both diseases. The authors suggest CBD, metformin, and plant compounds could theoretically reset these molecular switches, but the evidence is almost entirely preclinical, and no clinical trial data supports these therapeutic proposals.
Can Epigenetic Therapies Break the Vicious Cycle Between Gum Disease and Diabetes?
A new narrative review maps the molecular overlap between periodontitis and diabetes and proposes CBD, metformin, and plant compounds as candidate epigenetic regulators, though the evidence remains largely preclinical and speculative, with no clinical trial data supporting the proposed therapeutic strategies in this specific disease context.
#38
Limited Direct Relevance
Mechanistic hypotheses are plausible but rest on preclinical data; no clinical evidence supports the proposed CBD or combination strategies for this indication.
Diabetes Mellitus
Epigenetics
CBD
Metformin
Periodontitis affects roughly 45% of young adults and up to 60% of those over age 65, and its bidirectional relationship with diabetes places an enormous burden on both dental and metabolic health systems. Patients with periodontal disease face a 33% increased risk of hyperglycemia and a 1.24-fold higher risk of developing diabetes within five years. If the shared epigenetic mechanisms proposed in this review prove amenable to pharmacological intervention, it could open a genuinely novel therapeutic axis targeting the root of both conditions simultaneously. Understanding where the science actually stands on this question matters greatly for clinicians weighing emerging claims about CBD and other agents in this space.
The relationship between periodontitis and type 2 diabetes has long been recognized as bidirectional: hyperglycemia worsens periodontal tissue integrity and immune function, while chronic oral inflammation promotes insulin resistance and impaired glycemic control. This narrative review, published in the International Journal of Molecular Sciences in 2025, attempts to identify the shared epigenetic mechanisms that underlie this cycle. The authors focus on DNA methylation enzymes (particularly DNMT1 and DNMT3b), histone-modifying enzymes (HATs, HDACs, SIRT6), and NF-kappaB-driven inflammatory signaling as molecular nodes where periodontitis-associated oral pathogens, especially Porphyromonas gingivalis, can trigger systemic epigenetic reprogramming. A particularly notable mechanistic claim is that oral pathogen dissemination into the gut suppresses DNMT-1, leading to downregulation of beta-cell transcription factors PDX1 and NKX6.1, which produces dysfunctional bi-hormonal beta cells incapable of proper insulin secretion.
Having mapped these pathways, the review proposes CBD, metformin, and various natural plant-derived compounds as candidate epigenetic modulators that might interrupt the inflammatory cascade at key points. However, the authors themselves explicitly acknowledge that CBD’s epigenetic mechanisms in this specific context have not been elucidated. The evidence base draws primarily from mouse models (C57BL/6 ligature-induced periodontitis), mini-pig streptozotocin diabetes models, and in vitro human osteoblast and gingival cell lines. No clinical trial data for any of the proposed combination strategies are presented, and the review did not follow a systematic search methodology, meaning the literature coverage may reflect selective rather than comprehensive retrieval. The authors conclude that empirical validation through rigorous clinical studies is needed before any of these proposals can inform treatment decisions.
The mechanistic framework here is genuinely interesting. The idea that a single oral pathogen like P. gingivalis could trigger epigenetic changes cascading all the way to pancreatic beta-cell dysfunction is the kind of systems-level thinking that medicine needs more of. And to the authors’ credit, they are transparent about the enormous gap between the molecular hypotheses they assemble and anything approaching clinical evidence. What concerns me is how easily a review like this gets condensed into a headline claiming CBD can treat diabetes through epigenetic mechanisms. The authors never make that claim, and the data simply do not support it.
In my practice, I see patients with both periodontal disease and metabolic dysfunction frequently. The pragmatic approach right now is to ensure aggressive periodontal treatment is part of any comprehensive diabetes management plan, because the evidence for that intervention actually exists. When patients ask about CBD for inflammation, I discuss what we do and do not know honestly. CBD has demonstrated anti-inflammatory properties in certain contexts, but prescribing it to modulate DNMT3b methylation patterns in periodontal-diabetic comorbidity is not something any clinician should be doing based on this paper.
This review sits at the very earliest stage of the translational research arc: hypothesis generation based on mechanistic synthesis. The epigenetic pathways it describes, including SIRT6 deficiency impairing neutrophil efferocytosis and LPS-driven histone acetylation via NF-kappaB, are individually supported by preclinical experiments but have not been validated as druggable targets in human periodontitis-diabetes comorbidity. For clinicians, the immediate clinical utility is limited. The review is most valuable as a conceptual map of where future research might yield actionable targets, rather than as a guide for current practice.
From a pharmacological standpoint, the pairing of CBD with metformin raises important practical considerations that the review does not address. Metformin has well-characterized effects on hepatic glucose output and AMPK activation, and any epigenetic modulator used alongside it would need rigorous drug-interaction profiling. CBD is a potent inhibitor of cytochrome P450 enzymes, particularly CYP3A
