© The authors, 2026

Introduction

Oral cavity cancer (OCC) is the most common head and neck malignancy worldwide [1,2]. Traditionally, this disease affects older males with a significant history of tobacco and alcohol use [1,2]. Since the 1960s, however, smoking rates have steadily declined as increasing awareness of its harmful effects has led to healthcare policy changes [3]. An overall decline in alcohol consumption since the 1980s has also been observed [4]. Despite this trend, the incidence of OCC is rising, particularly in young adults under the age of 45 years [1,2,5]. This growing subset of patients lacks typical risk factors and continues to be a topic of intense research in head and neck oncology, as the underlying causative agent remains unidentified.

Chronic oral inflammation has been recognized as a key player in carcinogenesis. Tumor-associated macrophages (TAMs) are a major component of inflammatory infiltrates released at sites of injury, irritation, or infection and promote the inflammatory process [6]. Through a series of complex interactions and signaling, TAMs establish and maintain an inflammatory environment that leads to oxidative stress, irreversible DNA damage, and ultimately tumor formation [6].

Several drivers of chronic oral inflammation are implicated in the development of OCC. Genetic testing has been largely unrevealing. A comparison of oral cavity tumors in young versus older patients demonstrated similar mutational burden profiles [5]. The oral microbiome has also been investigated. Tumor bacterial load was found to be negatively correlated with the expression of p53, a tumor suppressor protein, potentially contributing to cancer development [7]. Smoking and drinking lead to a pro-inflammatory environment associated with increased pathogenic bacteria and a depletion of commensal bacteria [6]. Furthermore, smokers with poor oral hygiene are at a greater than 3-fold increased risk of OCC than those who maintain good oral hygiene [8]. Chronic dental trauma has also been evaluated. Compared with smokers, non-smoking females and females under age 50 years had significantly higher rates of lateral tongue cancer, a frequent site of accidental oral trauma [9]. Non-smokers who wear dentures were also more likely to develop carcinoma in the gingiva or floor of the mouth, areas in which dentures chronically rub and irritate [9]. Intra-oral contact allergies are also thought to contribute, as longstanding exposure can lead to stomatitis or inflammation of the oral mucosa [10].

Chronic oral inflammation appears to be critical to the development of OCC; however, its role in young, risk factor-negative patients remains poorly characterized. While individual studies have examined periodontal disease, the oral microbiome, and dental trauma, comprehensive population-level data evaluating the collective burden of oral diseases, inflammatory conditions, and allergies as OCC risk factors are lacking, particularly with stratification by age and substance use status. This gap is significant given the rising incidence of OCC in young adults without traditional behavioral risk factors. At the molecular level, microRNAs (miRNAs) have been reported to play important roles in regulating periodontal soft and hard tissue homeostasis [11], and miRNA dysregulation has been implicated in the pathogenesis of oral lichen planus (OLP), an oral potentially malignant disorder (OPMD) [12].

The objectives of this population-based case-control study are (1) to evaluate associations between pre-existing oral, allergic, and inflammatory conditions and OCC risk; (2) to determine whether these associations differ by age (<50 versus ≥50 years) and tobacco/alcohol use history; and (3) to identify potentially modifiable risk factors that may inform targeted screening and prevention strategies for high-risk populations.

Methods

Data were obtained from the statewide population database (SPD), a longitudinal resource housed at an NCI-designated cancer institute that integrates the State Cancer Registry (SCR) with statewide hospital discharge files and electronic medical records from two university health systems.

Inclusion criteria for cases were age ≥18 years at diagnosis, first primary histologically confirmed OCC (ICD-O-3 topography codes C00–C06), diagnosis between 1996 and 2016, and available pre-diagnosis health records in the SPD. Exclusion criteria were in situ disease only, prior cancer diagnosis at any site, diagnosis before 1996, and insufficient pre-diagnosis health records (<1 year). For the comparison cohort, inclusion criteria were age ≥18 years, no cancer diagnosis during the study period, and frequency-matched to cases on birth year (±2 years), sex, and birthplace (state versus other).

Exposure histories were assembled from ICD‑9/10 and Current Procedural Terminology (CPT) codes recorded at least one year before the index date (diagnosis for cases and pseudo‑diagnosis for comparators). A total of 300 codes were grouped into three, non‑overlapping categories: (1) oral diseases, including odontogenic infections, oral lesions, and dental procedures; (2) allergies, encompassing systemic or local allergic disorders and related testing or treatment; and (3) inflammatory conditions, defined as systemic or local inflammatory disorders with oral manifestations, oral injuries, or oral trauma. A participant was classified as exposed if any qualifying code was present before the index date. Smoking status (ever/never) was captured from ICD‑9/10 and CPT codes before cancer diagnosis. Body mass index (BMI) was abstracted from driver‑license records obtained at least one year before diagnosis. Missing BMI values were multiply imputed with predictive mean matching using age, sex, race/ethnicity, Charlson Comorbidity Index (CCI), smoking, and cancer status; estimates were combined with Rubin’s rules. Analyses were additionally stratified by age (< 50 versus ≥ 50 years) and by tobacco and/or alcohol history to explore effect modification.

All statistical analyses were conducted in SAS v9.4 (SAS Institute, Cary, NC). Unconditional logistic‑regression models estimated odds ratios (ORs) and 95% confidence intervals (CIs) for the association between each exposure category and OCC. Models adjusted for the frequency‑matching factors (age at index, sex, and calendar year) and the following a priori confounders: race/ethnicity (White, Asian, Pacific Islander, Other/Unknown), BMI (continuous), CCI (0, 1, ≥ 2), and smoking status (ever versus never). Population‑attributable fractions (PAFs) and 95% CIs were calculated from the fully adjusted ORs using the Greenland–Thomas method. Sensitivity analyses compared complete‑case to imputed‑BMI results, excluded exposures recorded within two years of the index date to reduce reverse‑causality bias, and refit models stratified by smoking/alcohol status and by sex. Two‑sided P‑values < 0.05 were considered statistically significant; ORs are reported with their corresponding 95% CIs.

Results

Baseline characteristics of OCC patients and general population

The study included 754 OCC patients and 3758 population controls. Sex distribution and family history of head and neck cancer were similar between groups. Twenty percent of OCC cases were diagnosed before age 50 years. Compared with controls, OCC patients were more likely to smoke or drink, have lower BMI, and lack any family history of cancer (Tab. 1). Race distributions also differed overall, although both cohorts were predominantly White. Finally, OCC cases had lower CCI scores. No other notable differences were observed.

OCC risk by conditions

Oral and dental conditions were associated with higher OCC odds (Fig. 1). Inflammatory disorders also showed an association. No association was observed for allergies.

Fig. 1 OCC risk by conditions. OCC, oral cavity cancer; Gen. Pop., general population; OR, odds ratio; and CI, confidence interval. Totals: OCC 754; Gen. Pop. 3758. †Models adjust for sex, race/ethnicity, residence, baseline BMI, Charlson Comorbidity Index, smoking, and alcohol. *P from multivariable logistic regression.

Age-stratified OCC risk by conditions

Oral and dental conditions were associated with OCC regardless of age; a stronger association was present in the younger cohort (<50 years) (Fig. 2). Inflammatory disorders were associated with OCC only in the ≥50 years stratum. Allergies were not associated with OCC in either age group.

Fig. 2 Age-stratified OCC risk by conditions. OCC, oral cavity cancer; Gen. Pop., general population; OR, odds ratio; CI, confidence interval; and n, participants Totals: <50 years (OCC 151; Gen. Pop. 743); ≥50 years (OCC 603; Gen. Pop. 3015). † Unadjusted counts; cell-suppression applied. * Adjusted logistic regression.

Tobacco/alcohol-stratified OCC risk by conditions

In non-smokers/non-drinkers, oral and dental conditions showed a nearly 4-fold greater OCC risk and inflammatory disorders revealed a modestly increased risk (Fig. 3). No associations were detected in the smoking/drinking stratum, nor for allergies in either stratum.

Fig. 3 Tobacco/alcohol-stratified OCC risk by conditions. OCC, oral cavity cancer; Gen. Pop., general population; NS/ND, non-smoking/non-drinking; +S/D, smoking & drinking; OR, odds ratio; CI, confidence interval; and n, participants. Totals: NS/ND 730/3729; +S/D 24/29. †Unadjusted counts; IRB cell-suppression applied. *P from adjusted logistic regression.

OCC risk by condition in NS/ND adults by age

Within non-smoking/non-drinking adults <50 years, oral and dental conditions were associated with an almost 9-fold greater OCC risk, while adults ≥50 years demonstrated a moderately elevated risk (Fig. 4). Inflammatory disorders among non-smoking/non-drinking adults ≥50 years revealed a smaller increased OCC risk. No other associations reached statistical significance.

Fig. 4 OCC risk by condition in NS/ND adults by age. OCC, oral cavity cancer; Gen. Pop., general population; NS/ND, non-smoking/non-drinking; OR, odds ratio; CI, confidence interval; and n, participants. Totals: <50 years (OCC 151; Gen. Pop. 743); ≥50 years (OCC 603; Gen. Pop. 3015). †Unadjusted counts; IRB cell-suppression applied. *Adjusted logistic regression.

Discussion

In this population-based case-control study of 754 OCC patients and 3758 matched controls, we found that pre-existing oral diseases were associated with a 3.2-fold increased OCC risk overall, rising to an 8.8-fold excess in young non-smoking/non-drinking adults. Inflammatory disorders conferred a more modest but significant 1.7-fold increased risk, while allergies showed no association.

Since the latter half of the 20th century, tobacco use and alcohol consumption have seen an overall decline in the United States [3,4]. Despite this downward trend, there has been a progressive increase in OCC in adults under 45 years of age without behavioral risk factors [1,2,5]. A number of etiologies have been implicated, including the oral microbiome, dental trauma, poor oral hygiene, periodontal disease, and stomatitis [8–10]. Importantly, these factors all share the potential to establish a chronic inflammatory environment within the oral cavity, ultimately serving as a nidus for carcinogenesis. There is a paucity of data evaluating the link between oral diseases, allergies, or inflammatory conditions and OCC risk, particularly in young adults and those without traditional risk factors. Thus, we sought to determine whether individuals with these pre-existing conditions were at an elevated risk for OCC and specifically evaluated this relationship based on age and among non-smoking/non-drinking individuals.

Our baseline characteristics demonstrated a lower family history of any cancer in the OCC cohort compared with the general population. The literature is conflicted on family history of cancer and personal OCC risk. Fantozzi et al. conducted a retrospective study evaluating this relationship and found that a history of cancer in one or more siblings was associated with a 1.6-times greater risk of oral squamous cell carcinoma (OSCC) [13]. This contrasts with a large case-control study in France by Radoi et al., which did not identify an increased rate of OCC in individuals with a family history of cancer [14]. Yet, a higher risk of OCC among subjects with a first-degree relative with head and neck cancer was observed [14].

In our study, we found similar rates of family history of head and neck cancer in OCC patients and the general population. In our study, we found similar unadjusted rates of family history of head and neck cancer in OCC patients and the general population. In contrast, prior studies that accounted for tobacco and alcohol use have consistently reported increased OCC risk associated with family history, regardless of tobacco and alcohol exposure [15]. The differences observed in our study may be attributed to the unique characteristics of the SPD and its underlying population. The state population has distinctive demographic and genetic features due to founder effects and migration patterns that may influence cancer susceptibility patterns differently than in other studied populations. The comprehensive genealogical data and extensive family history information available in the SPD represent both a strength and a potential source of methodological variation when comparing our results to previous studies.

Oral and dental conditions were statistically more prevalent in the OCC cohort compared with the general population. This difference was even more pronounced, nearly 9 times higher, when specifically evaluating the young, risk factor-negative OCC group. These findings are congruent with previous studies in OCC patients without behavioral factors. A case-control study by Zhou et al. reported that periodontal disease and eating disorders were associated with OSCC among never-smokers in the National Institutes of Health All of Us database [16]. Both conditions contribute to poor oral health, which has been established as a risk factor for OCC even in the absence of tobacco or alcohol exposure [8].

Several mechanistic clues help explain the striking, 9‑fold OCC excess we observed in young never‑smokers/never‑drinkers with antecedent oral disease. Severe periodontitis has been shown to double oral cancer risk in never‑smokers in both a nationwide Taiwanese cohort of > 700,000 adults and a U.S. case–control study [17,18]. Tumor tissues from non‑smoking OCC patients are enriched for Fusobacterium nucleatum and Porphyromonas gingivalis—periodontal pathogens able to promote inflammation, cell proliferation, and cellular invasion via decreased expression of the tumor suppressor protein, p53 [19,20]. Tobacco and alcohol exacerbate this by shifting the microbiome toward a “pathogenic‑rich” state [7]. In their absence, microbiome‑mediated mechanisms may play an even larger etiologic role, magnifying the effect size we observed. Finally, early‑onset periodontitis has been linked to host genetic and immunologic susceptibilities (e.g., toll‑like‑receptor variants) that could predispose young adults to OCC independently of behavioral exposures [18]. Collectively, these data support a microbiome‑ and inflammation‑mediated pathway underlying the pronounced OCC risk in young, risk‑factor‑negative individuals.

A prior diagnosis of inflammatory disorders was independently associated with OCC in our cohort, reinforcing the concept that chronic inflammation creates a pro-carcinogenic environment in the oral cavity. This finding aligns with prior reports linking systemic or inflammatory states to OSCC. Zhou et al. observed higher OCC risk among individuals with hyperlipidemia, HIV infection, and oral-related autoimmune diseases—all proxies for sustained inflammatory activation [16]. Likewise, chronic oral inflammation characterizes several OPMDs, such as OLP, oral lichenoid lesions, oral graft-versus-host disease, and lupus erythematosus [21], that the World Health Organization recognizes as harboring a markedly elevated transformation risk [21,22]. Taken together, our data and the existing literature underscore inflammation not merely as a correlate but as a plausible causal driver of OCC development.

Malignant transformation in OPMDs occurs through several chronic inflammation-driven mechanisms. These include oxidative/nitrosative stress that causes DNA damage and mutations in key regulatory genes, activation of pro-inflammatory signaling pathways (particularly NF-κB) that drive cytokine production, dysregulated cell proliferation, and resistance to apoptosis; and microenvironment remodeling involving epithelial-mesenchymal transition, angiogenesis, and immunosuppression [23–25]. While these processes vary somewhat across specific disorders, they collectively create a persistent inflammatory milieu that promotes genomic instability and malignant transformation of oral epithelial cells [25]. Our findings may differ from these reported associations due to methodological differences, the specific patient population studied, or inadequate power to detect these relationships in our cohort.

At the molecular level, miRNAs have been reported to play important roles in regulating periodontal soft and hard tissue homeostasis during inflammatory and mechanical stress responses. In a randomized clinical trial of 21 patients, Polizzi et al. examined the expression of miRNA-7, -21, -100, -125b, and -200b families in gingival crevicular fluid and demonstrated that periodontal tissue remodeling significantly impacted miRNA expression profiles, with miRNA-21-5p showing significant upregulation on the compression side [11]. Furthermore, miRNA dysregulation has been implicated in the pathogenesis of OLP, a chronic inflammatory mucocutaneous disorder affecting the oral mucosa with an estimated global prevalence of 0.5–2.2% [12]. The etiology of OLP is unclear, but its pathogenesis involves immune-mediated mechanisms through autoimmune apoptosis by activated cytotoxic CD8⁺ T-lymphocytes and miRNA dysregulation [12].

We did not identify an association between allergies and OCC. Prior studies have identified an inverse relationship. A case-control study by Stott-Miller et al. observed a decreased risk of OSCC in individuals with a history of allergies to dust, pollen, or mold [26]. It is important to note, however, that our study only evaluated for the presence of general allergies. This is relevant given the growing body of evidence suggesting intraoral exposure to dental amalgam, specifically, may cause contact allergy and induce carcinogenesis. In a prospective study of 65 OSCC patients, 34% were found to be allergic to at least one metal adjacent to their tumor [27]. Importantly, the authors did not exclude tobacco or alcohol use. Further research that distinctly evaluates intraoral metal allergies, while controlling for confounding factors such as tobacco and alcohol use, is warranted to clarify their potential role in OCC risk.

Inflammation is a major risk factor for OCC. Many cases are related to complications of an oral ailment that are potentially preventable. Modifiable risk factors include tobacco use, alcohol consumption, poor oral hygiene, and chronic oral irritation or inflammation. While there is a lack of high-certainty evidence to support the use of screening tests for oral cancer and conditions that may lead to OCC in the general population, maintaining good oral hygiene with annual dental exams and cleanings and avoiding tobacco and alcohol are relatively simple preventative measures [28]. Additionally, early detection can improve treatment and survival outcomes. Dental professionals should be watchful for signs of OPMD and malignancies during routine oral exams.

The recognition of OPMDs as precursor lesions emphasizes the importance of early intervention. A wide range of topical pharmacological strategies have been reported to prevent OLP malignant progression [12]. In a randomized controlled trial of 47 OLP patients, Polizzi et al. demonstrated that topical fluocinonide 0.05% gel significantly reduced OLP symptoms, signs, and disease extension score at 6-month follow-up compared with placebo [12]. These findings suggest that timely identification and management of oral preneoplastic conditions may represent a viable strategy for reducing OCC risk and inflammatory disease progression.

There were several limitations to this study, including its retrospective design. In addition, the rarity of young, non-smoking and non-drinking patients resulted in a small sample size for this subcategory and underpowered our study. The study's limited statistical power may have prevented detection of significant differences in risk factors, particularly the potential roles of inflammation and allergies in OCC development, between younger and older OCC patients compared with the general population. Furthermore, CPT and ICD codes were used to query data. This approach favored comprehensiveness over specificity and may have led to inadvertent inclusion of patients with other predisposing conditions not explicitly delineated within the three categories of risk factors in the study. Conversely, since dental records are typically excluded from medical records, this approach may have excluded pertinent data regarding periodontal health and nuanced dental procedures that would only be captured from patients’ dental records. Finally, a significant limitation of this study is its demographic homogeneity, as over 90% of participants were White individuals from the SPD, potentially limiting the generalizability of our findings to populations with different racial backgrounds, geographic locations, and environmental exposure profiles.

Conclusion

Our statewide cohort reveals that antecedent diseases of the mouth and teeth represent the strongest modifiable risk factor for OCC, with particularly pronounced effects in young, never-smoking, never-drinking adults, a population traditionally considered low-risk. While chronic inflammatory conditions conferred modest but significantly elevated risk, allergies showed no association. These findings challenge current risk assessment paradigms and support incorporating systematic oral health screening, including periodontal assessment, evaluation of chronic irritation sources, and documentation of recurrent oral lesions into routine dental and primary care encounters for all adults, regardless of age or behavioral risk factors. Furthermore, evidence supporting early identification and treatment of OPMDs such as OLP suggests that intervention at the preneoplastic stage may represent a viable strategy for reducing OCC risk.

Acknowledgments

Research was supported by the National Center for Research Resources grant “Sharing Statewide Health Data for Genetic Research” (R01 RR021746, G. Mineau, PI), with additional support from the Utah Department of Health and Human Services and the University of Utah. Partial support for all UPDB datasets was provided by the University of Utah Huntsman Cancer Institute and its Cancer Center Support Grant (P30 CA042014). The Utah Cancer Registry is funded by the National Cancer Institute’s SEER Program (Contract HHSN261201800016I) and the CDC’s National Program of Cancer Registries (Cooperative Agreement NU58DP007131), with additional support from the University of Utah and the Huntsman Cancer Foundation. We thank University of Utah Health Data Science Services and Intermountain Health for data and analytics support; the University of Utah Pedigree and Population Resource; and the University of Utah Health Enterprise Data Warehouse for establishing the Master Subject Index linking the UPDB and UUHC.

Funding

This work was supported by the Huntsman Cancer Institute donors.

Conflicts of interest

The authors declare that they have no conflict of interest in relation to this article.

Data availability statement

This study used data from a statewide, governed population database that links vital records, a state cancer registry, and health system records. Due to legal/ethical restrictions and data-use agreements, the underlying individual-level data are not publicly available. De-identified, aggregate data dictionaries and analysis outputs may be shared on reasonable request to the corresponding author, subject to approval by the data custodians and Institutional Review Board and completion of a data-use agreement. Code used for data management and analysis is available from the corresponding author upon request.

Author contribution statement

1) Vanessa Husmann: Conceptualization, Methodology, Formal Analysis, Investigation, Data Curation, Writing—Original Draft, Visualization, Project Administration.

2) Krista Ocier: Formal Analysis, Investigation, Data Curation.

3) Esther Chun-Pin Chang: Writing—Review & Editing.

4) Ankita Date: Writing—Review & Editing.

5) David Gill: Writing—Review & Editing.

6) Vikrant Deshmukh: Writing—Review & Editing.

7) Mia Hashibe: Conceptualization, Methodology, Writing—Review & Editing, Supervision.

8) Jason P. Hunt: Conceptualization, Methodology, Writing—Review & Editing, Supervision, Funding Acquisition.

Ethics approval

This study received ethical approval from the Institutional Review Board at the University of Utah on 14 November 2010 under IRB: 00045048.

Informed consent

Not applicable because only de-identified data were analyzed; no identifiable private information was used.

References

All Tables

All Figures

	Fig. 1 OCC risk by conditions. OCC, oral cavity cancer; Gen. Pop., general population; OR, odds ratio; and CI, confidence interval. Totals: OCC 754; Gen. Pop. 3758. †Models adjust for sex, race/ethnicity, residence, baseline BMI, Charlson Comorbidity Index, smoking, and alcohol. *P from multivariable logistic regression.
In the text

	Fig. 2 Age-stratified OCC risk by conditions. OCC, oral cavity cancer; Gen. Pop., general population; OR, odds ratio; CI, confidence interval; and n, participants Totals: <50 years (OCC 151; Gen. Pop. 743); ≥50 years (OCC 603; Gen. Pop. 3015). † Unadjusted counts; cell-suppression applied. * Adjusted logistic regression.
In the text

	Fig. 3 Tobacco/alcohol-stratified OCC risk by conditions. OCC, oral cavity cancer; Gen. Pop., general population; NS/ND, non-smoking/non-drinking; +S/D, smoking & drinking; OR, odds ratio; CI, confidence interval; and n, participants. Totals: NS/ND 730/3729; +S/D 24/29. †Unadjusted counts; IRB cell-suppression applied. *P from adjusted logistic regression.
In the text

	Fig. 4 OCC risk by condition in NS/ND adults by age. OCC, oral cavity cancer; Gen. Pop., general population; NS/ND, non-smoking/non-drinking; OR, odds ratio; CI, confidence interval; and n, participants. Totals: <50 years (OCC 151; Gen. Pop. 743); ≥50 years (OCC 603; Gen. Pop. 3015). †Unadjusted counts; IRB cell-suppression applied. *Adjusted logistic regression.
In the text