Clinical biomarkers identification for type 2 Diabetes Mellitus and Periodontitis: Biological and Artificial Intelligence-based investigation

• Grant number: 24/15353-0
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: July 01, 2025
• End date: February 29, 2028
• Field of knowledge: Health Sciences - Dentistry - Periodontology
• Principal Investigator: Raquel Mantuaneli Scarel Caminaga
• Grantee: Francois Isnaldo Dias Caldeira
• Host Institution: Faculdade de Odontologia (FOAr). Universidade Estadual Paulista (UNESP). Campus de Araraquara. Araraquara , SP, Brazil

Abstract

It is well established that Type 2 Diabetes Mellitus (T2DM) influences Periodontitis (P). However, in the presence of T2DM, the longitudinal effect of periodontal treatment on key immune system molecules remains unclear or contradictory, as does its potential modulation by systemic glycemic control. Supported by FAPESP, one of our recent findings was a gradual reduction in the gene (systemic mRNA) and protein expression (gingival crevicular fluid - GCF) of TGF-¿1 following periodontal treatment at 90 and 180 days. Since TGF-¿1 is functionally associated with other key immune molecules, we hypothesize that periodontal treatment may influence not only TGF-¿1 but also other molecules, as suggested by GeneMANIA, which may serve as clinical biomarkers. Given the significant impact of T2DM complications and the need to better investigate systemic-oral interactions in health and disease, Medicine has utilized the Systemic Inflammatory Parameters, but in Dentistry it is rare. Additionally, with advancements in Artificial Intelligence (AI), numerous medical studies have aimed to identify hidden relationships in complex clinical datasets and improve early disease diagnosis. This longitudinal clinical study aims to apply AI to identify biomarkers for the clinical monitoring of T2DM and Periodontitis and to detect causal relationships, advancing our understanding of disease interactions. Moreover, we will investigate TGF-¿1 functional related molecules: IL-17A and IL-23A (Th17 profile), IL-12B, IFN-¿, and IL-18 (Th1 profile), as well as IL-1B, IL-6, and IL-37, primarily questioning whether patients' metabolic compensation (HbA1c levels) and periodontal treatment act as modifying factors of the transcriptional and translational levels of these molecules in both periodontal and systemic contexts. By executing this proposal, we aim to expand and deepen our understanding of potential causal relationships while generating information with the potential for patent development. The methodological stages, including the storage of biological samples, have been completed. A total of 156 patients (minimum 30 per group) were allocated into the following groups: (1) Uncontrolled T2DM + P, (2) Controlled T2DM + P, (3) T2DM without P, (4) Periodontitis, and (5) Healthy Controls. All participants underwent glycemic, lipid, and periodontal evaluations before and after periodontal treatment (90 and 180 days). Patients in the P and T2DM+P groups received non-surgical and/or surgical periodontal treatment (when indicated), while individuals in the Control and T2DM without P groups underwent supragingival scaling, prophylaxis, and oral hygiene reinforcement at the same time points. Leukocyte RNA from each patient (baseline, 90, and 180 days) was stored in an ultrafreezer and will be used to generate cDNA and investigate the transcriptional levels (gene expression) of IL17A, IL23A, IL12B, IFNG, IL18, IL1B, IL6, and IL37 via RT-qPCR. Protein expression (translational levels) of these molecules will be analyzed in GCF and plasma using Multiplex immunoassays. Comparative statistical analyses will be conducted between groups and longitudinally, alongside regressions and correlations to examine variables associations. AI algorithms will be applied for advanced pattern recognition in biomarker identification, as well as methods to determine causal relationships. These results will be integrated with findings from additional molecular analyses conducted on the same patients in a recently completed project. This study aims to provide insights that can help personalize clinical practices and improve monitoring of interconnected chronic diseases. The synergy between AI (ARM) and the depth of our new dataset will allow this project to contribute significantly to the understanding and management of periodontitis and T2DM. (AU)