Juneja_et

Odovtos-International Journal of Dental Sciences (Odovtos-Int. J. Dent. Sc.), Online First, 2026. ISSN: 2215-3411

https://revistas.ucr.ac.cr/index.php/Odontos

CLINICAL RESEARCH:

Altered Catecholamine Metabolism in Atypical Facial Pain: Quantitative and Predictive Insights from Salivary COMT Expression

Alteración del metabolismo de catecolaminas en el dolor facial atípico: aportes cuantitativos y predictivos a partir de la expresión salival de COMT

Sarthak Juneja1 https://orcid.org/0009-0000-9919-1860

Suganya Panneer Selvam2 https://orcid.org/0000-0002-9272-6444

Ramya Suresh3 https://orcid.org/0009-0007-7968-786X

Ramya Ramadoss4 https://orcid.org/0000-0001-9408-3158

Sandhya Sundar5 https://orcid.org/0000-0001-5085-8337

1Undergraduate Student, Department of Oral Biology, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, Tamil Nadu, India.

2Associate Professor, Department of Oral Biology, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, Tamil Nadu, India.

3Postgraduate Student, Department of Oral Biology, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, Tamil Nadu, India.

4Professor & Head, Department of Oral Biology, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, Tamil Nadu, India.

5Associate Professor, Department of Oral Biology, Saveetha Dental College and Hospitals, Saveetha Institute of Medical and Technical Sciences, Saveetha University, Chennai, Tamil Nadu, India.

Correspondence to: Suganya Panneer Selvam - suganresearch29@gmail.com

Received: 13-XI-2025 Accepted: 14-I-2026

ABSTRACT: Atypical Facial Pain (AFP) is a chronic orofacial pain disorder marked by persistent, idiopathic facial pain significantly affecting the mental health and quality of life. Identifying objective biomarkers remains challenging. Catechol-O-methyltransferase (COMT), a key enzyme in catecholamine metabolism, modulates pain and stress responses; its reduced activity has been linked to chronic pain. This study assessed salivary COMT levels in AFP patients and explored its diagnostic utility through statistical and machine-learning approaches. A comparative cross-sectional study was conducted with 20 AFP patients and 20 age- and sex-matched healthy controls. Unstimulated saliva samples (8:00-10:00 AM) were analyzed using a Human COMT ELISA kit (KRISHGEN Biosystems). Group differences were tested with an independent samples t-test, and effect size, confidence intervals, and power were calculated. Diagnostic performance was evaluated using ROC and logistic regression, with leave-one-out cross-validation (LOOCV) for model validation. AFP patients showed significantly lower salivary COMT (14.20±3.15 ng/mL) than controls (20.08±2.39 ng/mL; t(18)=3.03, p=0.0073). The mean difference (7.03 ng/mL; 95% CI: 2.07-11.49) yielded a large effect size (Cohen’s d=1.35) and power of 0.82. ROC analysis produced an AUC=0.86 with 90% sensitivity and specificity at 17.7 ng/mL. Logistic regression confirmed COMT as a significant AFP predictor (β= –1.19, p<0.05). LOOCV achieved 75% accuracy, 0.86 precision, 0.60 recall, and AUC=0.82. Salivary COMT levels are markedly reduced in AFP, reflecting altered catecholamine metabolism and pain modulation. Both statistical and machine-learning analyses confirmed excellent discriminative accuracy, supporting COMT as a promising non-invasive, accessible biomarker for AFP screening and diagnosis.

KEYWORDS: Pain; Biomarker; Mental health; COMT; Saliva.

RESUMEN: El dolor facial atípico (DFA) es un trastorno de dolor orofacial crónico caracterizado por dolor facial persistente e idiopático que afecta significativamente la salud mental y la calidad de vida. La identificación de biomarcadores continúa siendo un desafío. Este estudio evaluó los niveles salivales de COMT en pacientes con DFA y exploró su utilidad diagnóstica mediante enfoques estadísticos y de aprendizaje automático. Se realizó un estudio comparativo transversal con 20 pacientes con DFA y 20 controles sanos emparejados por edad y sexo. Las muestras de saliva no estimulada (8:00-10:00 a.m.) fueron analizadas utilizando un kit ELISA para COMT humano (KRISHGEN Biosystems). Las diferencias entre grupos se evaluaron mediante la prueba t para muestras independientes, y se calcularon el tamaño del efecto, los intervalos de confianza y la potencia estadística. Los pacientes con DFA presentaron niveles salivales de COMT significativamente más bajos (14,20±3,15 ng/mL) en comparación con los controles (20,08±2,39 ng/mL; t(18)=3,03; p=0,0073). El análisis ROC produjo un área bajo la curva (AUC) de 0,86, con una sensibilidad y especificidad del 90 % en un punto de corte de 17,7 ng/mL. La regresión logística confirmó que COMT es un predictor significativo de DFA (β=–1,19; p<0,05). Los niveles salivales de COMT se encuentran marcadamente reducidos en el DFA, lo que refleja alteraciones en el metabolismo de catecolaminas y en la modulación del dolor. Tanto los análisis estadísticos como los de aprendizaje automático confirmaron una excelente capacidad discriminativa, lo que respalda a COMT como un biomarcador prometedor, no invasivo y accesible para el tamizaje y diagnóstico del dolor facial atípico.

PALABRAS CLAVE: Dolor; Biomarcador; Salud mental; COMT; Saliva.

INTRODUCTION

Atypical Facial Pain (AFP), also referred to as Persistent Idiopathic Facial Pain (PIFP), is a chronic orofacial pain disorder characterized by persistent, poorly localized, and often unilateral pain in the face or oral cavity significantly affects the mental health & quality of life (1). According to the International Classification of Orofacial Pain (ICOP, 1st edition), PIFP is classified under Idiopathic Orofacial Pain, denoting continuous facial pain occurring in the absence of identifiable neurological, dental, or structural pathology (2). Unlike classical trigeminal neuralgia, AFP lacks distinct neurological or dental pathology, making diagnosis challenging and frequently leading to delayed or inappropriate treatment. The pain is typically dull, deep, and continuous, persisting for months or years, and often resistant to conventional analgesic therapy. The complex clinical presentation and overlapping features with other neuropathic or musculoskeletal conditions have contributed to diagnostic uncertainty, highlighting the need for objective and quantifiable biomarkers that can facilitate early and accurate identification of this condition (3).

The precise pathophysiology of AFP remains elusive; however, current evidence suggests that it involves a combination of neurological, somatic, and psychological mechanisms. Altered central pain processing, particularly within the spinal trigeminal nucleus and limbic structures, has been implicated in the development of chronic facial pain syndromes(4). Functional imaging studies reveal abnormal brainstem and cortical activation patterns in AFP patients, consistent with impaired sensory modulation and hyperexcitability of nociceptive pathways. Psychological factors, including anxiety and depression, further exacerbate symptom perception and pain chronicity, suggesting that AFP arises from a multifactorial interplay between neurobiological and psychosocial factors (5). These complexities highlight the limitations of purely clinical or symptomatic diagnosis and highlight the potential of biochemical markers to enhance diagnostic precision.

Catechol-O-methyltransferase (COMT) is a key enzyme responsible for the inactivation of catecholamine neurotransmitters such as dopamine, norepinephrine, and epinephrine, which play essential roles in stress regulation, emotional processing, and pain modulation. Alterations in COMT activity have been linked to variations in pain sensitivity, stress response, and susceptibility to chronic pain disorders. Genetic variants of COMT—particularly the Val158Met and Arg128Ala polymorphisms—can influence enzymatic activity, thereby affecting catecholamine turnover and individual pain thresholds (6). Previous studies have demonstrated that reduced COMT activity is associated with heightened pain perception and diminished analgesic response, suggesting its role as a molecular mediator of pain sensitivity (7,8). These findings provide a compelling rationale for exploring COMT as a potential biomarker in AFP.

Saliva has recently emerged as a valuable non-invasive diagnostic fluid in biomedical and dental research. It contains a diverse range of biological molecules—enzymes, hormones, proteins, and nucleic acids—that reflect both local and systemic physiological states (9). Salivary collection is painless, simple, and repeatable, making it ideal for screening, monitoring, and longitudinal studies. Studies have identified salivary biomarkers for conditions such as oral cancer, periodontal disease, and neuropathic pain, emphasizing its diagnostic potential (10,11). Given the presence of COMT in saliva and its known role in catecholamine metabolism, quantifying salivary COMT concentrations might provide a biochemical overview of the neurochemical alterations underlying AFP, bridging the gap between molecular neuroscience and clinical pain assessment. The present study aims to quantitatively assess salivary COMT enzyme levels in patients with atypical facial pain and compare them with healthy controls using enzyme-linked immunosorbent assay (ELISA).

METHODOLOGY

Study Design

This study was designed to quantitatively analyse salivary catechol-O-methyltransferase (COMT) levels in patients diagnosed with atypical facial pain (AFP) compared to healthy controls using an enzyme-linked immunosorbent assay (ELISA). All patients with AFP included in this study were diagnosed based on the International Headache Society (IHS) criteria (12). AFP was characterized by persistent oral or facial pain lasting more than two hours daily for at least three months, typically described as dull, nagging, or poorly localized, and occurring without any detectable neurological deficits. As AFP is a diagnosis of exclusion, each participant underwent a thorough clinical assessment, including imaging and dental evaluation, to rule out alternative causes such as trigeminal neuralgia, sinusitis, dental infections, temporomandibular disorders, or intracranial lesions. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board (IRB) of Saveetha Dental College and Hospital, Saveetha Institute of Medical and Technical Sciences (IRB Approval Number: IHEC/SDC/UG-2339/23/OBIO/146).

The overall workflow consisted of four phases: (1) participant recruitment and saliva collection, (2) COMT quantification using ELISA, (3) statistical analyses including inferential and diagnostic modeling, and (4) machine learning–based classification analysis for predictive validation.

Participants

Twenty patients clinically diagnosed with atypical facial pain were recruited from the Department of Oral Medicine at Saveetha Dental College and Hospital. Diagnosis was confirmed by an experienced oral medicine specialist based on the International Classification of Headache Disorders (ICHD-3) criteria. An equal number of healthy control participants (n=20), age- and sex-matched, were selected from the general population of the same institution. All participants provided written informed consent before inclusion in the study.

Inclusion and Exclusion Criteria

Participants aged 18-65 years were eligible for inclusion. AFP patients required a confirmed clinical diagnosis (pain persisting daily for at least two hours per day over a minimum duration of three months) as per IHS criteria, while controls were required to have no history of facial pain, neuropathic disorders, or chronic pain conditions. Participants with systemic diseases, neurological disorders, or those taking medications known to influence COMT activity (such as monoamine oxidase or COMT inhibitors) were excluded from the study. Individuals with a recent history of smoking or tobacco use, recent dental procedures, or active oral infections were also not included. Only subjects who met all the defined inclusion and exclusion criteria were finally enrolled for analysis to ensure the reliability and biological validity of salivary COMT measurements.

Sample Collection

Unstimulated whole saliva was collected between 8:00 AM and 10:00 AM to minimize diurnal variation. Participants were instructed to refrain from eating, drinking, or brushing their teeth at least 30 minutes before collection. Using sterile saliva collection devices, participants allowed saliva to pool naturally and expectorated into pre-labeled tubes. Samples were immediately placed on ice and transported to the laboratory, where they were centrifuged at 3,000 rpm for 10 minutes at 4°C to remove debris. The supernatant was aliquoted into sterile microcentrifuge tubes and stored at −80°C until biochemical analysis.

ELISA Procedure

Salivary COMT concentrations were determined using a Human COMT ELISA kit (KRISHGEN Biosystems – GENLISATM Human COMT ELISA). The 96-well plate was prepared according to the manufacturer’s protocol, including all standard and control reagents (standard COMT solutions, assay buffer, substrate, stop solution, and wash buffer). Fifty microliters (50 µL) of each sample or standard were pipetted into duplicate wells. Following incubation for 2 hours at room temperature with gentle shaking (300 rpm), plates were washed four times with wash buffer. Subsequently, 100 µL of detection antibody was added, incubated for 1 hour, and washed again. Finally, 100 µL of substrate was added and incubated in the dark for 30 minutes before adding 50 µL stop solution. Absorbance was measured at 450 nm using a BeneSphera microplate reader (Avantor, Radnor, USA). A standard curve was generated, and COMT concentrations were interpolated from the absorbance readings. All samples were analyzed in duplicate to ensure accuracy and reproducibility.

Data Analysis and Statistical Methods

Descriptive and Inferential Statistics

All statistical analyses were performed using SPSS (version 26.0), GraphPad Prism 9.0, and R (version 4.2.2). Data were first assessed for normality using the Shapiro–Wilk test. Between-group differences in salivary COMT concentrations were analyzed using an independent samples t-test. Effect size was computed as Cohen’s d to quantify the magnitude of the difference, and a 95% confidence interval (CI) for the mean difference was calculated. A post-hoc power analysis was performed (α=0.05) to assess the adequacy of the sample size.

Diagnostic Accuracy Analysis

The diagnostic performance of salivary COMT as a biomarker was evaluated using Receiver Operating Characteristic (ROC) curve analysis. The Area Under the Curve (AUC) was computed to assess discriminatory power, and the optimal diagnostic threshold was determined using Youden’s Index. Corresponding sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated.

Logistic Regression Analysis

A binary logistic regression model was fitted to examine the predictive relationship between salivary COMT levels (independent variable) and AFP diagnosis (dependent variable: 0=healthy, 1=AFP). The model estimated the odds ratio (OR) and regression coefficient (β), quantifying the likelihood of AFP diagnosis per unit change in COMT concentration. Model significance was tested using the Wald statistic, and model fit was evaluated via the Hosmer–Lemeshow test.

Machine Learning Classification

To complement traditional inferential analysis, a machine learning (ML) approach using logistic regression with Leave-One-Out Cross-Validation (LOOCV) was implemented to validate the predictive ability of salivary COMT. Each sample served once as a validation case while the remaining samples were used for training.

Model performance was evaluated using key classification metrics, including accuracy, precision, recall (sensitivity), and F1-score. The predictive discrimination ability of the model was assessed through the Receiver Operating Characteristic Area Under the Curve (ROC AUC), and the optimal decision threshold was determined using Youden’s Index. The final logistic regression coefficient and intercept were reported, indicating that the probability of atypical facial pain (AFP) was inversely related to salivary COMT concentration—lower COMT levels corresponded to a higher likelihood of AFP. All machine learning procedures were implemented in Python (v3.11) using the scikit-learn (v1.5) library to ensure transparency, reproducibility, and robustness of the analysis.

RESULTS

Descriptive Analysis of Salivary

COMT Levels

The quantitative analysis of salivary catechol-O-methyltransferase (COMT) enzyme concentrations revealed a clear difference between patients diagnosed with atypical facial pain (AFP) and healthy control subjects. The mean COMT level in healthy controls was 20.08±2.39 ng/mL, whereas patients with AFP exhibited a markedly lower mean concentration of 14.20±3.15 ng/mL, indicating a reduction of approximately 29% in the AFP group (Table 1).

Table 1. Comparison of salivary COMT enzyme levels between groups.

Group	Mean (ng/mL)	Standard Deviation	Minimum	Maximum
Healthy Controls	20.08	2.39	10.98	24.93
Atypical Facial Pain	14.20	3.15	2.16	22.85

The data exhibited a relatively narrow distribution among healthy controls, while the AFP group displayed greater inter-individual variability, with values ranging from 2.16 to 22.85 ng/mL. This wider spread may reflect differing degrees of biochemical dysregulation among AFP patients.

Inferential Statistics

An independent samples t-test demonstrated that the difference in mean salivary COMT concentration between groups was statistically significant (t(18)=3.03, p=0.0073). The mean difference between the two groups was 7.03 ng/mL, with a 95% confidence interval (CI) ranging from 2.07 to 11.49 ng/mL, confirming the robustness of the observed difference (Table 2). The effect size (Cohen’s d=1.35) indicated a large practical difference between groups, signifying a meaningful biological variation in COMT levels associated with AFP. A post-hoc power analysis confirmed that the study had adequate power (1−β=0.82) to detect this effect at a 5% significance level.

Table 2. Inferential statistics and effect size.

Statistical Parameter	Value	Interpretation
t-statistic	3.03	Significant difference between groups
p-value	0.0073	<0.05, statistically significant
Mean Difference	7.03 ng/mL	Healthy > AFP
95% CI (Mean Diff.)	2.07–11.49 ng/mL	Excludes zero (significant)
Cohen’s d	1.35	Large effect size
Post-hoc Power	0.82	Adequate statistical power

Diagnostic Potential of Salivary COMT Levels

To evaluate the diagnostic accuracy of salivary COMT levels in differentiating AFP patients from healthy controls, a Receiver Operating Characteristic (ROC) curve was constructed. The Area Under the Curve (AUC) was 0.86, indicating excellent discriminative ability (Table 3).

An optimal diagnostic threshold of 17.7 ng/mL (determined via Youden’s Index) provided a sensitivity of 90% and a specificity of 90%, as illustrated in Figure 1.

Table 3. Diagnostic accuracy metrics from ROC analysis.

Metric	Value	Interpretation
AUC (Area Under Curve)	0.86	Excellent diagnostic accuracy
Optimal Cut-off	17.7 ng/mL	Derived from Youden’s Index
Sensitivity	90%	High true positive rate
Specificity	90%	High true negative rate

Figure 1. ROC Curve illustrating the diagnostic performance of salivary COMT levels in distinguishing AFP patients from healthy controls.

Machine Learning–Based Classification

To further evaluate the discriminative capacity of COMT levels, a logistic regression classifier was trained using Leave-One-Out Cross-Validation (LOOCV) to minimize bias given the small sample size. The model achieved a cross-validated accuracy of 75%, with precision=0.86, recall (sensitivity)=0.60, and an F1-score of 0.71. The ROC AUC of the logistic model was 0.82, closely matching the ROC analysis from inferential statistics, confirming the robustness of COMT as a predictor of AFP. Optimization using Youden’s Index yielded an optimal probability threshold of 0.49, corresponding to sensitivity=80% and specificity=90%, improving the detection of AFP cases while minimizing false positives (Figure 2.A). The final logistic regression coefficient was −1.19 (p<0.05), indicating that higher COMT levels are associated with a lower probability of AFP (Figure 2.B) (Table 4).

Figure 2. (A) ROC curve showing classification performance of logistic regression model (AUC = 0.82), (B) Logistic probability curve depicting predicted probability of AFP against salivary COMT concentration with the optimal cut-off (≈17.7 ng/mL) marked.

Table 4. Machine Learning Classification Performance (LOOCV Logistic Regression).

Metric	Value	Interpretation
Accuracy	0.75	75% correct classifications
Precision	0.86	High reliability of AFP predictions
Recall (Sensitivity)	0.60	Correct AFP detection rate
F1-score	0.71	Balanced precision-recall
ROC AUC	0.82	Good discriminative power
Optimal Threshold	0.49	Probability cut-off
Sensitivity (optimal)	0.80	Improved detection rate
Specificity (optimal)	0.90	Strong negative predictive value

Data Visualisation and Distribution Patterns

Both box plots and violin plots (Figure 3.A and Figure 3.B) demonstrate that the AFP group exhibits a downward shift in COMT concentrations relative to healthy controls. The AFP distribution curve appears broader, reflecting greater biological variability, whereas control values cluster more tightly around the mean.

Figure 3. Box plot showing the distribution of COMT concentrations between study groups.

Interpretation of Findings

Collectively, these findings confirm that salivary COMT enzyme levels are significantly reduced in patients with atypical facial pain, and that this reduction is both statistically and clinically meaningful. The large effect size, high AUC values, and consistent logistic regression performance all indicate that COMT serves as a strong independent discriminator between AFP and healthy states. Given its established role in catecholamine metabolism and pain modulation, reduced COMT activity may reflect impaired dopaminergic and adrenergic regulation, which could contribute to the chronic, neuropathic characteristics of AFP. The integration of statistical and machine learning analyses strengthens the biomarker validity of salivary COMT, supporting its use as a non-invasive diagnostic tool. Future studies incorporating multivariate regression models, genetic polymorphism analysis (e.g., COMT Val158Met), and larger sample sizes are warranted to further elucidate the mechanistic and diagnostic significance of COMT in atypical facial pain.

DISCUSSION

This study is the first to evaluate salivary catechol-O-methyltransferase (COMT) levels in patients diagnosed with atypical facial pain (AFP). COMT plays an essential role in the degradation of catecholamine neurotransmitters such as dopamine, norepinephrine, and epinephrine, all of which are deeply involved in pain modulation and stress response (13). The present findings demonstrate that salivary COMT concentrations were significantly lower in AFP patients compared to healthy controls (p=0.0073), with a mean difference of 7.03 ng/mL. This reduction was not only statistically significant but also biologically meaningful, as indicated by a large effect size (Cohen’s d=1.35), signifying a substantial deviation in COMT activity between affected individuals and controls.

The diagnostic potential of salivary COMT was further supported by receiver operating characteristic (ROC) analysis, which yielded an area under the curve (AUC) of 0.86, reflecting excellent discriminative ability. An optimal diagnostic threshold of 17.7 ng/mL provided both high sensitivity (90%) and specificity (90%) in distinguishing AFP patients from healthy subjects. These findings suggest that COMT may serve as a robust, non-invasive biomarker for identifying individuals with atypical facial pain. The post-hoc statistical power of 82% indicates that the present study was adequately powered to detect the observed difference, thereby reinforcing the reliability of these outcomes. To complement the classical statistical analysis, a supervised machine learning approach using logistic regression with leave-one-out cross-validation was employed to assess the predictive performance of COMT levels. The model achieved an overall accuracy of 75%, a precision of 0.86, a recall (sensitivity) of 0.60, and an ROC AUC of 0.82. When optimized using Youden’s index, the sensitivity improved to 80% and specificity remained at 90%, mirroring the statistical findings and confirming the predictive value of salivary COMT. The machine learning model’s negative coefficient indicated that higher COMT levels were associated with a lower probability of AFP, consistent with the biological premise of reduced enzymatic activity contributing to pain hypersensitivity.

COMT has long been recognized as a regulator of pain perception through its influence on adrenergic and dopaminergic pathways (14). Reduced enzymatic activity or expression of COMT can result in elevated catecholamine levels, amplifying nociceptive signaling and stress-induced hyperalgesia (15). Experimental and clinical studies have shown that diminished COMT activity contributes to increased pain sensitivity, consistent with the lower salivary enzyme levels observed in AFP patients in this study (16-18). These biochemical alterations may underlie the aberrant central pain modulation and persistent, diffuse facial pain characteristic of AFP. The convergence between inferential and machine learning analyses strengthens the evidence that COMT is a reliable biomolecular discriminator between AFP and healthy states.

Genetic studies have highlighted the influence of COMT variants, such as Val158Met and Arg128Ala, on pain sensitivity, stress response, and analgesic outcomes (19,20). Previous studies further establish that functional COMT haplotypes significantly modulate catecholamine metabolism and contribute to inter-individual variability in “idiopathic” and functional pain disorders. Individuals carrying low-activity COMT alleles typically exhibit heightened pain perception and reduced tolerance to discomfort (21). The findings of the present study align with this genetic evidence, suggesting that reduced salivary COMT enzyme levels may reflect underlying genotypic variations or stress-related downregulation of COMT expression, which collectively contribute to the heightened pain sensitivity observed in AFP. The integration of machine learning classification models with genotypic and biochemical data in future studies may provide deeper insights into the molecular mechanisms driving AFP and help develop precision-based diagnostic frameworks.

Saliva as a diagnostic medium offers considerable advantages due to its non-invasive collection, ease of handling, and biochemical richness (22-24). Studies reported that COMT polymorphisms are associated with experimental pain thresholds in healthy individuals; however, their work focused exclusively on genotyping and blood-derived biomarkers, without assessing COMT protein levels in saliva or examining patients with clinical pain conditions (25,26). The current findings reinforce the potential of saliva-based biomarkers for pain-related conditions, particularly when evaluating enzymes such as COMT that are central to neurotransmitter metabolism. Beyond COMT, previous studies have demonstrated that salivary proteins and enzymes, including total protein and amylase, correlate with oral and systemic health indicators (27-29). Thus, quantifying salivary COMT could form part of a broader proteomic biomarker panel for early detection and monitoring of chronic orofacial pain disorders. The application of machine learning approaches, as demonstrated in this study, further enhances diagnostic reliability by objectively classifying disease states based on continuous biomarker data.

The results of the present study also align with emerging research exploring COMT and other biomarkers in neuropsychiatric conditions such as major depressive disorder (MDD). While COMT has been implicated in modulating mood and treatment responsiveness, its utility as a predictive biomarker remains under investigation (30,31). The present findings contribute to this growing body of evidence, underscoring that altered COMT activity may represent a shared neurobiological pathway between chronic pain and affective dysregulation. The use of predictive algorithms, such as logistic regression or neural network models, may in the future facilitate early identification of high-risk individuals by integrating biochemical, genetic, and psychological predictors. This biochemical overlap supports a biopsychosocial understanding of AFP, integrating both physiological and psychological dimensions of pain experience.

Nevertheless, it is important to acknowledge that COMT does not act in isolation. Pain perception is the result of a complex interaction among multiple neurotransmitter systems, including serotonin and norepinephrine, which collectively regulate central pain pathways (32). Furthermore, while salivary COMT levels provide valuable peripheral indicators, they may not perfectly reflect central nervous system enzymatic activity. The technological challenges of accurately quantifying COMT, along with individual variability in stress, hormonal status, and medication use, may influence the measured values (33,34). Therefore, while COMT holds strong promise as a biomarker, it should be interpreted within a multidimensional diagnostic framework rather than as a standalone indicator. As chronic pain is a recognised mental illness–linked burden impacting quality of life, productivity, and life expectancy, identifying non-invasive biomarkers such as COMT may reduce diagnostic delay and prevent progression of pain-related disorders.

FUTURE PERSPECTIVE

Based on these findings, future research should expand to multicentric studies with larger, genetically stratified cohorts to confirm COMT’s diagnostic reliability and explore genotype–phenotype correlations. Integrating salivary COMT measurement with additional neurochemical and inflammatory biomarkers, clinical parameters, and advanced machine learning algorithms could yield robust multivariate predictive models for precision pain diagnostics. Such integrative approaches may ultimately transform salivary-based biomarker profiling into a practical, non-invasive tool for early detection, prognosis, and personalized management of chronic orofacial pain disorders, including AFP.

CONCLUSION

The present study shows that salivary catechol-O-methyltransferase (COMT) levels are significantly reduced in patients with atypical facial pain (AFP) compared to healthy individuals, underscoring the enzyme’s potential as a non-invasive diagnostic biomarker. The combination of strong statistical significance (p=0.0073), a large effect size (Cohen’s d=1.35), and high diagnostic accuracy (AUC=0.86; sensitivity and specificity=90%) demonstrates both the reliability and clinical relevance of COMT quantification in differentiating AFP from healthy states. Complementary machine learning analysis further validated these findings, achieving an AUC of 0.82 and confirming that lower salivary COMT concentrations accurately predict AFP, consistent with the enzyme’s established role in catecholamine metabolism and pain modulation pathways. These findings highlight the importance of incorporating salivary biomarkers into personalized health and medicine approaches for complex disease presentations such as AFP.

SOURCE OF FUNDING: Self-funding.

ETHICAL POLICY & INSTITUTIONAL REVIEW BOARD STATEMENT: The study was duly approved by the Saveetha university institutional ethical committee board obtained IHEC/SDC/UG-2339/23/OBIO/146.

PATIENT DECLARATION OF CONSENT: Informed consent was obtained from the patients & the patient’s identity is not revealed.

DATA AVAILABILITY STATEMENT: Not applicable.

Conflict of interest: The authors declare no conflicts of interest.

AUTHOR CONTRIBUTION STATeMENT: Original Drafting & Data collection, S.J.; Conceptualization & Supervision, S.P.S.; Data analysis, R.S. and R.R.; Review, R.S., S.P.S. and S.S.; Methodology, R.R.; Editing, S.P.S. and S.S.

REFERENCES

1. Schweiger V., Nocini R., De Santis D., et al. Persistent Idiopathic Facial Pain (PIFP) in Patients Referred to a Multidisciplinary Centre in Italy: A Retrospective Observational Study. J Clin Med. 2022; 11 (13): 3821. Published 2022 Jul 1. doi:10.3390/jcm11133821

2. International Classification of Orofacial Pain, 1st edition (ICOP). (2020). Cephalalgia : an international journal of headache, 40 (2): 129-221. https://doi.org/10.1177/0333102419893823

3. Jalan, S., & Ramesh, R. (2025). Assessment of Nanosized Plastics Exposure in Children: Evaluating the Impact of Daily Interaction with Plastic Items on Saliva Contamination. Int. J. Clin. Pediatr. Dent. 18 (9): 1087-1096. https://doi.org/10.5005/jp-journals-10005-3243

4. Ziegeler C., Schulte L.H., May A. Altered trigeminal pain processing on brainstem level in persistent idiopathic facial pain. Pain. 2021; 162 (5): 1374-1378. doi:10.1097/j.pain.0000000000002126

5. Tizzoni R., Tizzoni M., Clerici C.A. Atypical odontalgia and trigeminal neuralgia: psychological, behavioral and psychopharmacological approach in a dental clinic - an overview of pathologies related to the challenging differential diagnosis in orofacial pain. F1000Res. 2021; 10: 317. Published 2021 Apr 23. doi:10.12688/f1000research.51845.3

6. Dharaniprasad G., Samantaray A., Srikanth L., Hanumantha Rao M., Chandra A., Sarma PVGK. Novel Arg128Ala variant in Catechol-O-methyltransferase gene influence persistent pain. Clin Genet. 2020 Feb; 97 (2): 378-380. doi: 10.1111/cge.13650. Epub 2019 Oct 24. PMID: 31646620.

7. Firfirey F., Shamley D., September A.V. Polymorphisms in COMT and OPRM1 Collectively Contribute to Chronic Shoulder Pain and Disability in South African Breast Cancer Survivors'. Genes (Basel). 2022 Dec 21; 14 (1): 9. doi: 10.3390/genes14010009. PMID: 36672750; PMCID: PMC9858584.

8. Zhang Z., Gewandter J.S., Geha P. Brain Imaging Biomarkers for Chronic Pain. Front Neurol. 2022 Jan 3; 12: 734821. doi: 10.3389/fneur.2021.734821. PMID: 35046881; PMCID: PMC8763372.

9. Al-Haddad, A., Azmi, N. N., Penukonda, R., Chia, M. A., Lee, X. W., & K Anuar, A. F. (2025). The pattern of antibiotics prescription for endodontic infections in Malaysia: Cross sectional survey study. Australian endodontic journal: the journal of the Australian Society of Endodontology Inc, 51 (1): 55-63. https://doi.org/10.1111/aej.12900

10. Basilicata M., Pieri M., Marrone G., Nicolai E., Di Lauro M., Paolino V., Tomassetti F., Vivarini I., Bollero P., Bernardini S., Noce A. Saliva as Biomarker for Oral and Chronic Degenerative Non-Communicable Diseases. Metabolites. 2023 Jul 27; 13 (8): 889. doi: 10.3390/metabo13080889. PMID: 37623833; PMCID: PMC10456419.

11. Umapathy V.R., Natarajan P.M., Swamikannu B. Review Insights on Salivary Proteomics Biomarkers in Oral Cancer Detection and Diagnosis. Molecules. 2023 Jul 7; 28 (13): 5283. doi: 10.3390/molecules28135283. PMID: 37446943; PMCID: PMC10343386.

12. Pfaffenrath, V., Rath, M., Pöllmann, W., & Keeser, W. (1993). Atypical facial pain--application of the IHS criteria in a clinical sample. Cephalalgia : an international journal of headache, 13 Suppl 12, 84–88. https://doi.org/10.1177/0333102493013S1218

13. Šanjug, J.; Kuna, K.; Goldštajn, M.Š.; Dunkić, L.F.; Carek, A.; Negovetić Vranić, D. Relationship between COMT Gene Polymorphism, Anxiety, and Pain Perception during Labour. J. Clin. Med. 2023, 12, 6298. https://doi.org/10.3390/jcm12196298

14. Zhou, Q.; Verne, G.N. Molecular Mechanisms and Pathways in Visceral Pain. Cells 2025, 14, 1146. https://doi.org/10.3390/cells14151146

15. Zhang X., Kanter K., Chen J., et al. Low catechol-O-methyltransferase and stress potentiate functional pain and depressive behavior, especially in female mice. Pain. 2020 Feb; 161 (2): 446-458. DOI: 10.1097/j.pain.0000000000001734. PMID: 31972854; PMCID: PMC7282074.

16. Meloto C.B., Bortsov A.V., Bair E., et al. Modification of COMT-dependent pain sensitivity by psychological stress and sex. Pain. 2016; 157 (4): 858-867. doi:10.1097/j.pain.0000000000000449

17. Regensburger, M., Ip, C. W., Kohl, Z., Schrader, C., Urban, P. P., Kassubek, J., & Jost, W. H. (2023). Clinical benefit of MAO-B and COMT inhibition in Parkinson's disease: practical considerations. Journal of neural transmission (Vienna, Austria : 1996), 130 (6), 847-861. https://doi.org/10.1007/s00702-023-02623-8

18. Aranäs, C., Edvardsson, C. E., Shevchouk, O. T., Zhang, Q., Witley, S., Blid Sköldheden, S., Zentveld, L., Vallöf, D., Tufvesson-Alm, M., & Jerlhag, E. (2023). Semaglutide reduces alcohol intake and relapse-like drinking in male and female rats. EBioMedicine, 93: 104642. https://doi.org/10.1016/j.ebiom.2023.104642

19. Yang, M., Baser, R. E., Khanin, R., Autuori, I., Li, Q. S., Panageas, K. S., Orlow, I., & Mao, J. J. (2023). COMT Val158Met Affects the Analgesic Response to Acupuncture Among Cancer Survivors With Chronic Pain. The journal of pain, 24 (9): 1721-1730. https://doi.org/10.1016/j.jpain.2023.05.005

20. Dharaniprasad, G., Samantaray, A., Srikanth, L., Hanumantha Rao, M., Chandra, A., & Sarma, P. V. G. K. (2020). Novel Arg128Ala variant in Catechol-O-methyltransferase gene influence persistent pain. Clinical genetics, 97 (2): 378-380. https://doi.org/10.1111/cge.13650

21. Belfer I. (2013). Nature and nurture of human pain. Scientifica, 2013: 415279. https://doi.org/10.1155/2013/415279

22. Panneer Selvam S., Ramadoss R., Shanmugam R., et al. (February 18, 2024) Assessment of Female Hormonal Influence on COVID-19 Vaccine Response: A Prospective Cohort Study. Cureus 16 (2): e54417. doi:10.7759/cureus.54417

23. Abhinav, R. P., Bennett, C. M., Anjana, R. M., Pramodkumar, T. A., Senthilmurugan, M., Livingston, P. M., Pradeepa, R., Mohan, V., & Williams, J. (2025). Expression of Salivary and Serum IGF-1 and IGFBP-3 in Individuals With Diabetes and Oral Cancer. Journal of maxillofacial and oral surgery, 24 (2): 432-440. https://doi.org/10.1007/s12663-024-02212-6

24. Hemavathy, O. R., Marimuthu Ramaswamy, M., Mohana Priya, C. D., & Bhardwaj, S. (2025). Role of Liquid Biopsy in Oral Premalignant and Malignant Lesions: Correlation with CD24 and CD44 Expression in Early Diagnosis of Oral Cancer. Journal of maxillofacial and oral surgery, 24 (1): 207-212. https://doi.org/10.1007/s12663-024-02398-9

25. Xu, F., Yin, J., Xiong, E., Wang, R., Zhai, J., Xie, L., Li, Y., Qin, X., Wang, E., Zhang, Q., Zuo, Y., Fan, S., & Wang, S. (2020). COMT gene variants and β-endorphin levels contribute to ethnic differences in experimental pain sensitivity. Molecular pain, 16: 1744806920908474. https://doi.org/10.1177/1744806920908474

26. Wiegand A., Blickle A., Brückmann C., Weller S., Nieratschker V., Plewnia C. Dynamic DNA Methylation Changes in the COMT Gene Promoter Region in Response to Mental Stress and Its Modulation by Transcranial Direct Current Stimulation. Biomolecules. 2021; 11 (11): 1726. Published 2021 Nov 19. doi:10.3390/biom11111726

27. Kurien, V.T. Vino; Shanker, Rajesh Kashyap; Hegde, Shashikanth; Onisha, V.; Pillai, Siva A S. Evaluation of Salivary Parameters and Oral Health Status in Periodontally Healthy Subjects and Chronic Periodontitis Subjects. Advances in Human Biology 14 (3): p 203-209, Jul-Sep 2024. | DOI: 10.4103/aihb.aihb_3_24

28. Kavitha, L., Kuzhalmozhi, M., Vijayashree Priyadharsini, J., Arun Kumar, A., Umadevi, K. M. R., & Ranganathan, K. (2025). Microbial signatures in head and neck squamous cell carcinoma: an in silico study. Journal of applied oral science : revista FOB, 33, e20240392. https://doi.org/10.1590/1678-7757-2024-0392

29. Laputková, G., Schwartzová, V., Bánovčin, J., Alexovič, M., & Sabo, J. (2018). Salivary Protein Roles in Oral Health and as Predictors of Caries Risk. Open life sciences, 13: 174-200. https://doi.org/10.1515/biol-2018-0023

30. Beer, C., Rae, F., Semmler, A., & Voisey, J. (2024). Biomarkers in the Diagnosis and Prediction of Medication Response in Depression and the Role of Nutraceuticals. International journal of molecular sciences, 25 (14): 7992. https://doi.org/10.3390/ijms25147992

31. Stolz, L. A., Kohn, J. N., Smith, S. E., Benster, L. L., & Appelbaum, L. G. (2023). Predictive Biomarkers of Treatment Response in Major Depressive Disorder. Brain sciences, 13 (11): 1570. https://doi.org/10.3390/brainsci13111570

32. Forssell H., Jääskeläinen S., List T., Svensson P., Baad-Hansen L. An update on pathophysiological mechanisms related to idiopathic oro-facial pain conditions with implications for management. J Oral Rehabil. 2015 Apr; 42 (4): 300-22. doi: 10.1111/joor.12256. Epub 2014 Dec 8. PMID: 25483941.

33. Sharav Y., Katsarava Z., Charles A. Facial presentations of primary headache disorders. Cephalalgia. 2017 Jun; 37 (7): 714-719. doi: 10.1177/0333102417705374. Epub 2017 Apr 11. PMID: 28399652.

34. Abhinaya L.M., Muthukrishnan Arvind, Deepika Rajendran. Neuropathic Pain Management in a Tertiary Care Oral Medicine Unit. Int J Dentistry Oral Sci. 2021; 8 (8): 4010- 4015.