Abstract
Background
Stomatitis is a common oral disease that seriously affects patients’ quality of life. However, the risk of drug-induced stomatitis has not been systematically evaluated based on large-scale real-world data.
Objectives
To identify and validate high-risk drugs associated with drug-induced stomatitis, characterize the baseline features and time-to-onset heterogeneity of drug-induced stomatitis, preliminarily explore underlying mechanisms and provide evidence for clinical risk management.
Methods
This pharmacovigilance study used the FDA Adverse Event Reporting System (FAERS, 2004 Q1–2025 Q3). Disproportionality analysis (ROR, PRR, IC, EBGM), multivariable logistic regression, and time-to-onset (TTO) analysis were applied to identify high-risk drugs.
Results
A total of 55,735 stomatitis reports were screened from 58,209,958 records. The majority were female (62.1%), aged 40–80 years, from the United States (55.5%), and reported by health professionals (58.3%). We identified 366 suspicious drugs and confirmed 222 high-risk drugs. The strongest signals were found in antineoplastic and immunomodulatory agents, followed by alimentary tract, nervous system, and anti-infective drugs. TTO analysis showed obvious heterogeneity in onset time among different drug categories.
Conclusion
This study first systematically identifies high-risk drugs associated with drug-induced stomatitis. The main mechanisms involve mucosal epithelial injury, barrier repair disorder, local irritation, and immune imbalance. These results provide evidence for clinical risk management, rational drug use, and early intervention.
Key words: Stomatitis, Pharmacovigilance, FAERS, Disproportionality analysis, Adverse drug reaction
Graphical abstract
Introduction
Stomatitis, defined as inflammation and ulceration of the oral mucosa, encompasses several clinical forms, including aphthous stomatitis, contact stomatitis, and mucositis. As a frequently encountered oral condition, it significantly impairs patients' quality of life.1, 2, 3 With an estimated lifetime prevalence exceeding 39%, stomatitis represents a highly prevalent oral condition, underscoring the urgent need for effective prevention and therapeutic strategies.1 Clinical presentations range from mild erythema to severe, painful ulcerations that commonly interfere with essential oral functions such as eating, swallowing, and speaking.4 In severe or persistent cases, stomatitis may lead to systemic complications, posing substantial risks to overall patient health.5
The pathogenesis of stomatitis is complex and multifactorial, involving local irritants, infectious agents, immune dysregulation, and adverse drug reactions.6, 7, 8 With the increasing use of pharmacotherapy in the management of systemic diseases, drug-related adverse events have garnered growing attention in modern clinical practice.9 Although certain drug associations have been identified,10, 11 risk of drug-induced stomatitis across a broad spectrum of drugs has not yet been systematically evaluated, and evidence derived from large-scale, real-world data remains scarce.
As the largest post-marketing drug safety surveillance database worldwide, FDA Adverse Event Reporting System (FAERS) provides a unique resource for detecting rare and unexpected adverse drug reactions.12, 13, 14 By enabling the analysis of millions of spontaneous reports, FAERS facilitates disproportionality analyses that generate potential signals of drug–adverse event associations.15
Recent real-world pharmacovigilance studies, such as Wang et al.,16 have focused specifically on drug-induced aphthous ulcers, a well-defined clinical entity characterized by recurrent, localized ulcerative lesions of the oral mucosa.1,7 While such work provides valuable insights into this specific oral phenotype, it primarily centres on ulcerative manifestations alone and does not encompass the full spectrum of drug-related oral mucosal reactions, including inflammation, erythema, erosion, and non-recurrent or ill-defined mucosal damage.17 As a standardized MedDRA terminology, stomatitis acts as an umbrella concept that covers the complete spectrum of oral mucosal adverse events, and is widely recognized in oncology and pharmacovigilance for the assessment of drug-associated oral toxicity.15,18 Adopting this broader definition of stomatitis enables a more holistic capture of oral mucosal adverse reactions and facilitates the identification of diversified drug safety signals, offering a complementary perspective to previous investigations that focused narrowly on aphthous ulcers.16
To address this knowledge gap, we utilized the FAERS to conduct a comprehensive assessment of drug-related stomatitis risk. Using data mining, disproportionality analysis and multivariable logistic regression, we sought to identify high-risk drugs, verify the stability of drug-stomatitis associations, characterize demographic and clinical features and time-to-onset patterns, preliminarily explore the underlying mechanisms,17 and generate evidence for clinical risk prevention and rational medication.
Methods
Study design
The study design and workflow are detailed in Figure 1. For the population-level data mining, we initially queried the FAERS database (https://www.fda.gov/drugs) to identify drugs potentially associated with stomatitis. Signal detection was performed using four algorithms: the reporting odds ratio (ROR), proportional reporting ratio (PRR), information component (IC), empirical Bayes geometric mean (EBGM), and statistical significance determined by Fisher's exact test after false discovery rate (FDR) correction.19, 20, 21 These algorithms are routinely employed for pharmacovigilance signal detection by major regulatory authorities, including the FDA and the WHO Uppsala Monitoring Centre (UMC).22,23 Detailed mathematical formulas, threshold criteria for positive signals, and corresponding references are provided in Supplementary Tables S1 and S2. Drugs demonstrating positive signals were further evaluated with multivariable logistic regression, adjusting for potential confounders including age, gender, weight, reporter occupation, country, and report year. Only drugs that met the positivity criteria across all disproportionality metrics and maintained statistical significance (FDR-adjusted P < .05) in both Fisher's exact test and the multivariable logistic regression model were advanced to the time-to-onset (TTO) analysis. This integrated approach has been previously employed to enhance the robustness and reliability of pharmacovigilance signal detection.20,24
Fig. 1.
Flowchart of the study design. EBGM, empirical Bayes geometric mean; IC, information component; PRR, proportional reporting ratio; ROR, reporting odds ratio.
Signal detection of stomatitis-associated drugs based on FAERS
This study utilized adverse event reports from the FDA Adverse Event Reporting System (FAERS) database, spanning from the first quarter (Q1) of 2004 to the third quarter (Q3) of 2025. Data were extracted from the Drug Information (DRUG) and Reaction Information (REAC) tables for signal detection analysis. In the REAC table, adverse events (AEs) were coded using the Medical Dictionary for Regulatory Activities (MedDRA). Specifically, reports associated with stomatitis were identified by screening the 'Preferred Term' (PT) field for the specific descriptor 'stomatitis'. For each Individual Case Safety Report (ICSR), a binary indicator was generated to denote the presence or absence of stomatitis, utilizing the unique primary identifier for each case. Subsequently, drugs designated with the role code 'Primary Suspect' (PS) were filtered from the DRUG table. In addition, we assigned an anatomical therapeutic chemical (ATC) code to each drug to group and analyze all stomatitis cases related to each drug. These suspect agents were then integrated with the adverse reaction labels based on unique report IDs to consolidate the final dataset for analysis.
Next, based on a 2 × 2 contingency table, a joint distribution matrix was constructed for each drug and stomatitis (a: both present, b: drug only, c: stomatitis only, d: neither present), and multiple metrics for adverse reaction signals were calculated, including the ROR, PRR, IC, EBGM, and the P-value from Fisher's exact test, and perform multiple comparison correction using the FDR method. The criteria for signal detection were provided in Supplementary Tables S1 and S2.25,26 To control for potential confounding factors, multivariate logistic regression analysis was also applied.
Logistic regression analysis
To further validate the potential independent associations between drug candidates and stomatitis reports, we constructed a multivariate logistic regression model based on the FAERS database for validation analysis.
First, a binary exposure variable was constructed for each drug. Reports containing the specific drug were assigned a value of 1, while those without the drug were coded as 0. Patient baseline information tables (containing whether or not the stomatitis event was reported, age, gender, body weight, occupation of the reporter, country, and report year) collated based on the exposure information for each drug were merged to form a structured data framework required for the analysis to control for the potential confounding effects of factors. The form of the model is as follows:
Logit [P (stomatitis report = 1)] = β0 + β1 × Drug Exposure + β2 × Age + β3 × Gender + β4 × Weight + β5 × Reporter Occupation + β6 × Country + β7 × Report Year. Regression coefficients for each drug exposure were extracted from the regression model, the OR and their 95% CIs were calculated, and the corresponding P-values were obtained. Statistical significance was determined by P < .05 after correction for FDR.
In addition to standard conventional logistic regression, we applied Firth logistic regression to improve the robustness of the OR estimates, particularly in scenarios with sparse events or rare exposures.
Ultimately, the results from the multivariable logistic regression were integrated with the initial disproportionality signals. Drugs were prioritized as high-confidence candidates only if they met the following stringent criteria: (1) demonstrated positive signals across all metrics (ROR, PRR, IC, EBGM, and significant after correction for FDR); and (2) achieved an FDR-adjusted P-value < .05 in the multivariable regression analysis. This dual-validation strategy was designed to identify high-risk drugs with robust clinical associations, providing a solid foundation for subsequent mechanistic investigations and the development of targeted intervention strategies.
Time-to-onset (TTO) analysis
Utilizing the FAERS database, the time-to-onset (TTO) for stomatitis-related adverse event reports was determined, defined as the interval between the initiation of the suspect drug and the onset of the adverse event. Building upon the preliminary signal detection and multivariable logistic regression analysis, we quantified the total number of reports and calculated the mean, median, and interquartile range (IQR) of the TTO. These metrics were employed to characterize the temporal distribution and kinetic profiles of stomatitis events associated with the prioritized drugs.
Statistics analysis
All data analyses were conducted in R (version 4.3.1), with statistical significance established at a P < .05.
Results
Baseline characteristics of the study population
A total of 55,735 stomatitis reports were identified from 58,209,958 records in the FAERS database (Table 1). The distributions of age and sex showed that both male and female patients were predominantly aged between 40 and 80 years, with a notably higher proportion of females aged >40 years compared with males. Regarding body weight, 24.9% of patients (13,876 cases) were within the 50–100 kg range, whereas a substantial proportion of data was missing (69.1%, 38,511 cases). During the 21 years of the study period, the annual number of adverse event reports related to stomatitis increased fluctuatly, with females showing a steeper and more variable trajectory, while males exhibited a more gradual and stable rise. Notably, reports were more concentrated in the recent period (2022-2025), accounting for 30.1% (16,775 cases). Geographically, the majority of reports originated from the United States (55.5%, 30,931 cases), followed by Canada (10.8%, 6,042 cases). Regarding reporter occupation, healthcare professionals accounted for a slightly higher proportion of reports (58.3%, 32,467 cases) compared with consumers (36.9%, 20,552 cases) (Table 1 and Figure 2).
Table 1.
Baseline data of patients with reported stomatitis in the FAERS database.
| Characteristic | N (%) |
|---|---|
| Total | N = 55,735 |
| Gender | |
| Male | 15,793 (28.3%) |
| Female | 34,631 (62.1%) |
| Missing | 5311 (9.5%) |
| Age | |
| < 18 | 1865 (3.3%) |
| 18–64 | 20,974 (37.6%) |
| 65–85 | 14,217 (25.5%) |
| > 85 | 663 (1.2%) |
| Missing | 18,016 (32.3%) |
| Weight | |
| < 50 | 2029 (3.6%) |
| 50–100 | 13,876 (24.9%) |
| > 100 | 13,19 (2.4%) |
| Missing | 38,511 (69.1%) |
| Country | |
| United States | 30,931 (55.5%) |
| Canada | 6042 (10.8%) |
| Japan | 5 273 (9.5%) |
| Germany | 1774 (3.2%) |
| Italy | 1165 (2.1%) |
| United Kingdom | 1056 (1.9%) |
| France | 994 (1.8%) |
| Others | 5504 (9.9%) |
| Missing | 2996 (5.4%) |
| Report Year | |
| 2004–2012 | 8311 (14.9%) |
| 2013–2015 | 7629 (13.7%) |
| 2016–2018 | 11,044 (19.8%) |
| 2019–2021 | 11,976 (21.5%) |
| 2022–2024 | 16,775 (30.1%) |
| Reporter Occupation | |
| Consumer | 20,552 (36.9%) |
| Health professional | 32,467 (58.3%) |
| Others | 96 (0.2%) |
| Missing | 2620 (4.7%) |
Fig. 2.
Baseline characteristics of patients with stomatitis-related adverse event reports in the FAERS database. (A) Distribution of patient age and gender. (B) Annual temporal trends of reported cases. (C) Geographic distribution of reports by country. (D) Occupation of reporters.
Signal detection of stomatitis-associated drugs
As illustrated by the Venn diagram (Figure 3A), 366 statistically significant signals associated with adverse drug events were detected through the use of four disproportionality analysis algorithms and Fisher's exact test. All included drugs demonstrated consistent positive signals across the four primary algorithms, with FDR-corrected P-values from Fisher's exact test remaining below 0.05, indicating robust signal stability. Subsequently, the top 50 drugs with the most significant signals were further identified (Table 2). Additionally, an Anatomical Therapeutic Chemical (ATC) code was assigned to each drug for classification purposes. The word cloud visualization (Figure 3B) revealed that antineoplastic agents and immunomodulatory drugs predominated among the detected signals. Representative drugs included Ibrance (ROR = 11.42), Afinitor (ROR = 26.49), Cabometyx (ROR = 19.31), and Sutent (ROR = 10.81). In addition, neurological agents such as Nicotrol (ROR = 16) were also identified. Other notable drugs included Hydrea (ROR = 16), as well as systemic anti-infective agents such as Imeth (ROR = 9). Furthermore, Periogard (ROR = 17.85), classified under alimentary tract and metabolism drugs, exhibited a relatively strong association signal, although the number of reports in this category was limited. Detailed results for all identified drugs are provided in Table S3.
Fig. 3.
Identification and classification of drugs associated with stomatitis. (A) Venn diagram illustrating the intersection of positive signals identified by the four disproportionality algorithms and Fisher’s exact test. (B) Word cloud displaying the distribution of Anatomical Therapeutic Chemical (ATC) classification codes for the identified positive drugs.
Table 2.
The top 50 signal detection results of stomatitis-associated drugs using ROR, PRR, IC, EBGM and Fisher’s test from the FAERS database based on fisher test.
| Drug name | Classification | Number | ROR (95% Cl) | PRR (95% Cl) | EBGM(EBGM05) | IC(IC025) | P-value | P-value (FDR) |
|---|---|---|---|---|---|---|---|---|
| Ibrance | Antineoplastic and immunomodulating agents | 2484 | 11.42 (10.97 - 11.89) | 11.31(10.87 - 11.78) | 10.86 (10.43) | 3.44 (3.38) | <.001 | <.001 |
| Afinitor | Antineoplastic and immunomodulating agents | 2309 | 26.49 (25.4 - 27.63) | 25.89(24.84 - 26.97) | 24.86 (23.83) | 4.64 (4.56) | <.001 | <.001 |
| Cabometyx | Antineoplastic and immunomodulating agents | 2152 | 19.31 (18.49 - 20.16) | 18.99(18.19 - 19.82) | 18.29 (17.51) | 4.19 (4.12) | <.001 | <.001 |
| Sutent | Antineoplastic and immunomodulating agents | 1288 | 10.81 (10.22 - 11.42) | 10.71(10.14 - 11.31) | 10.48 (9.92) | 3.39 (3.3) | <.001 | <.001 |
| Methotrexate | Antineoplastic and immunomodulating agents | 2035 | 4.07 (3.9 - 4.26) | 4.06(3.89 - 4.24) | 3.95 (3.78) | 1.98 (1.91) | <.001 | <.001 |
| Imbruvica | Antineoplastic and immunomodulating agents | 768 | 4.14 (3.85 - 4.44) | 4.12(3.84 - 4.43) | 4.08 (3.8) | 2.03 (1.92) | <.001 | <.001 |
| Orencia | Antineoplastic and immunomodulating agents | 714 | 3.88 (3.6 - 4.18) | 3.87(3.6 - 4.17) | 3.83 (3.56) | 1.94 (1.82) | <.001 | <.001 |
| Capecitabine | Antineoplastic and immunomodulating agents | 1416 | 9.2 (8.72 - 9.7) | 9.13(8.66 - 9.62) | 8.92 (8.46) | 3.16 (3.07) | <.001 | <.001 |
| Xeloda | Antineoplastic and immunomodulating agents | 571 | 9.34 (8.6 - 10.15) | 9.27(8.54 - 10.06) | 9.18 (8.45) | 3.2 (3.06) | <.001 | <.001 |
| Lenvima | Antineoplastic and immunomodulating agents | 555 | 9.66 (8.89 - 10.51) | 9.59(8.82 - 10.42) | 9.5 (8.74) | 3.25 (3.1) | <.001 | <.001 |
| Methotrexate sodium | Antineoplastic and immunomodulating agents | 655 | 4.21 (3.9 - 4.55) | 4.2(3.89 - 4.54) | 4.16 (3.85) | 2.06 (1.94) | <.001 | <.001 |
| Inlyta | Antineoplastic and immunomodulating agents | 492 | 11.96 (10.94 - 13.08) | 11.84(10.84 - 12.93) | 11.74 (10.74) | 3.55 (3.39) | <.001 | <.001 |
| Gilotrif | Antineoplastic and immunomodulating agents | 441 | 24.43 (22.22 - 26.86) | 23.9(21.78 - 26.22) | 23.72 (21.57) | 4.57 (4.36) | <.001 | <.001 |
| Tarceva | Antineoplastic and immunomodulating agents | 366 | 6.38 (5.75 - 7.07) | 6.35(5.73 - 7.03) | 6.31 (5.69) | 2.66 (2.49) | <.001 | <.001 |
| Fluorouracil | Antineoplastic and immunomodulating agents | 444 | 6.99 (6.37 - 7.68) | 6.95(6.34 - 7.63) | 6.91 (6.29) | 2.79 (2.63) | <.001 | <.001 |
| Piqray | Antineoplastic and immunomodulating agents | 289 | 15.5 (13.8 - 17.42) | 15.29(13.63 - 17.15) | 15.22 (13.55) | 3.93 (3.69) | <.001 | <.001 |
| Erbitux | Antineoplastic and immunomodulating agents | 277 | 8.18 (7.26 -9.21) | 8.12(7.22 - 9.14) | 8.09 (7.18) | 3.02 (2.8) | <.001 | <.001 |
| Panitumumab | Antineoplastic and immunomodulating agents | 427 | 13.17 (11.97 - 14.49) | 13.02(11.84 - 14.31) | 12.93 (11.75) | 3.69 (3.51) | <.001 | <.001 |
| Everolimus | Antineoplastic and immunomodulating agents | 2751 | 22.67 (21.8 - 23.56) | 22.23(21.4 - 23.09) | 21.18 (20.37) | 4.4 (4.34) | <.001 | <.001 |
| Tykerb | Antineoplastic and immunomodulating agents | 220 | 8.93 (7.82 - 10.2) | 8.86(7.77 - 10.11) | 8.83 (7.73) | 3.14 (2.9) | <.001 | <.001 |
| Cometriq | Antineoplastic and immunomodulating agents | 200 | 18.91 (16.44 - 21.75) | 18.59(16.2 - 21.34) | 18.53 (16.11) | 4.21 (3.89) | <.001 | <.001 |
| Busulfex | Antineoplastic and immunomodulating agents | 172 | 22.93 (19.71 - 26.68) | 22.46(19.37 - 26.05) | 22.4 (19.25) | 4.49 (4.09) | <.001 | <.001 |
| Doxorubicin | Antineoplastic and immunomodulating agents | 487 | 4.95 (4.53 - 5.42) | 4.94(4.52 - 5.39) | 4.9 (4.48) | 2.29 (2.15) | <.001 | <.001 |
| Cetuximab | Antineoplastic and immunomodulating agents | 421 | 7.97 (7.24 - 8.78) | 7.92(7.2 - 8.72) | 7.87 (7.15) | 2.98 (2.8) | <.001 | <.001 |
| Periogard | Alimentary tract and metabolism | 133 | 17.85 (15.03 - 21.19) | 17.56(14.84 - 20.79) | 17.52 (14.76) | 4.13 (3.71) | <.001 | <.001 |
| Alkeran | Antineoplastic and immunomodulating agents | 102 | 17.52 (14.4 - 21.31) | 17.25(14.22 - 20.91) | 17.22 (14.15) | 4.11 (3.61) | <.001 | <.001 |
| Erlotinib | Antineoplastic and immunomodulating agents | 540 | 5.12 (4.7 - 5.57) | 5.1(4.68 - 5.55) | 5.06 (4.64) | 2.34 (2.2) | <.001 | <.001 |
| Cabozantinib | Antineoplastic and immunomodulating agents | 2474 | 19.1 (18.34 - 19.89) | 18.79(18.05 - 19.55) | 18 (17.28) | 4.17 (4.1) | <.001 | <.001 |
| Lenvatinib | Antineoplastic and immunomodulating agents | 634 | 9.19 (8.5 - 9.94) | 9.12(8.44 - 9.86) | 9.03 (8.35) | 3.17 (3.04) | <.001 | <.001 |
| Sunitinib malate | Antineoplastic and immunomodulating agents | 955 | 10.72 (10.05 - 11.43) | 10.62(9.97 - 11.32) | 10.46 (9.81) | 3.39 (3.28) | <.001 | <.001 |
| Busulfan | Antineoplastic and immunomodulating agents | 279 | 13.07 (11.61 - 14.72) | 12.93(11.5 - 14.53) | 12.87 (11.43) | 3.69 (3.45) | <.001 | <.001 |
| Ibrutinib | Antineoplastic and immunomodulating agents | 812 | 4.04 (3.77 - 4.33) | 4.03(3.76 - 4.32) | 3.98 (3.72) | 1.99 (1.89) | <.001 | <.001 |
| Lapatinib | Antineoplastic and immunomodulating agents | 263 | 7.8 (6.9 - 8.8) | 7.75(6.86 - 8.74) | 7.71 (6.83) | 2.95 (2.73) | <.001 | <.001 |
| Erlotinib hydrochloride | Antineoplastic and immunomodulating agents | 313 | 7.97 (7.13 - 8.91) | 7.92(7.09 - 8.84) | 7.88 (7.05) | 2.98 (2.78) | <.001 | <.001 |
| Palbociclib | Antineoplastic and immunomodulating agents | 2525 | 10.62 (10.21 - 11.06) | 10.53(10.12 - 10.96) | 10.1 (9.7) | 3.34 (3.27) | <.001 | <.001 |
| Afatinib | Antineoplastic and immunomodulating agents | 474 | 22.85 (20.85 - 25.03) | 22.38(20.47 - 24.48) | 22.2 (20.26) | 4.47 (4.28) | <.001 | <.001 |
| Sunitinib | Antineoplastic and immunomodulating agents | 1362 | 10.22 (9.68 - 10.78) | 10.13(9.6 - 10.68) | 9.91 (9.38) | 3.31 (3.22) | <.001 | <.001 |
| Chlorhexidine gluconate | Alimentary tract and metabolism | 206 | 15.58 (13.57 - 17.88) | 15.36(13.41 - 17.6) | 15.31 (13.34) | 3.94 (3.64) | <.001 | <.001 |
| Regorafenib | Antineoplastic and immunomodulating agents | 305 | 6.85 (6.12 - 7.67) | 6.81(6.09 - 7.62) | 6.78 (6.06) | 2.76 (2.57) | <.001 | <.001 |
| Chlorhexidine | Alimentary tract and metabolism | 253 | 10.81 (9.55 - 12.24) | 10.71(9.47 - 12.11) | 10.67 (9.42) | 3.42 (3.18) | <.001 | <.001 |
| Axitinib | Antineoplastic and immunomodulating agents | 501 | 11.26 (10.31 - 12.3) | 11.15(10.22 - 12.17) | 11.06 (10.12) | 3.47 (3.31) | <.001 | <.001 |
| Alpelisib | Antineoplastic and immunomodulating agents | 319 | 14.14 (12.65 - 15.79) | 13.96(12.52 - 15.58) | 13.89 (12.43) | 3.8 (3.58) | <.001 | <.001 |
| Vincristina sulfato (809su) | Antineoplastic and immunomodulating agents | 3 | 66.6 (20.73 - 213.98) | 62.66(20.92 - 187.7) | 62.66 (19.5) | 5.97 (0.43) | <.001 | <.001 |
| Imeth | Antiinfectives for systemic use | 9 | 6.89 (3.57 - 13.26) | 6.85(3.57 - 13.13) | 6.85 (3.55) | 2.78 (1.19) | <.001 | <.001 |
| Purixan | Antineoplastic and immunomodulating agents | 11 | 5.61 (3.1 - 10.15) | 5.59(3.1 - 10.07) | 5.58 (3.09) | 2.48 (1.18) | <.001 | <.001 |
| Hydrea | Other drugs | 16 | 3.93 (2.41 - 6.43) | 3.92(2.4 - 6.4) | 3.92 (2.4) | 1.97 (1.04) | <.001 | <.001 |
| Nicotrol | Nervous system | 16 | 3.62 (2.22 - 5.92) | 3.62(2.22 - 5.9) | 3.61 (2.21) | 1.85 (0.95) | <.001 | <.001 |
| Vepesid | Antineoplastic and immunomodulating agents | 7 | 9.04 (4.3 - 19.02) | 8.97(4.29 - 18.76) | 8.97 (4.26) | 3.17 (1.1) | <.001 | <.001 |
| Afinitor 5mg novartis | Antineoplastic and immunomodulating agents | 3 | 89.43 (27.51 - 290.8) | 82.45(27.83 - 244.27) | 82.45 (25.36) | 6.37 (0.43) | <.001 | <.001 |
| Pegylated liposomal irinotecan | Antineoplastic and immunomodulating agents | 6 | 12.6 (5.63 - 28.18) | 12.46(5.62 - 27.6) | 12.46 (5.57) | 3.64 (1.14) | <.001 | <.001 |
Logistic regression analysis
We applied both conventional and Firth-corrected logistic regression models to further validate these associations. Our results revealed that numerous drugs were significantly associated with an elevated risk of stomatitis (Table S4, Table S5). We generated a forest plot for the 50 most significant drugs (Figure 4). Among them, periogard (OR = 145.02, 95% CI: 59.22–358.02) exhibited the highest risk estimates. Strong associations were also observed for antineoplastic and immunomodulatory agents such as dactinomycin (OR = 78.09, 95% CI: 49.61-119.16) and cometriq (cabozantinib) capsule (OR = 69.68, 95% CI: 30.54-142.06),alimentary tract and metabolism drugs such as tums (OR = 72.97, 95% CI: 30.82-157.22), nervous system drugs such as nicotine polacrilex (OR = 7.55, 95% CI: 5.28-10.43). Notably, the lower bounds of the 95% confidence intervals for all identified signal drugs were greater than 1, and FDR-adjusted p-values were all <.05, confirming the statistical robustness of the findings.
Fig. 4.
The forest plot of identified top 50 significant stomatitis-related drugs. CI, confidence interval; OR, odds ratio; LogOR, log odds ratio.
Subsequently, 20 representative drugs of both statistical significance and clinical importance were prioritized based on FDR-adjusted P-values (Table 3). Their corresponding adverse reaction profiles, as documented in official prescribing information, are summarized in Table 4 to provide a clinical benchmark for interpreting our findings
Table 3.
Representative drugs associated with stomatitis: signal detection and statistical results.
| Drug name | ROR (95% Cl) | PRR (95% Cl) | EBGM (EBGM05) | IC (IC025) | P-value (FDR) | Signal result |
|---|---|---|---|---|---|---|
| Ibrance | 11.42 (10.97–11.89) | 11.31(10.87–11.78) | 10.86 (10.43) | 3.44 (3.38) | <.001 | Positive |
| Afinitor | 26.49 (25.4–27.63) | 25.89(24.84–26.97) | 24.86 (23.83) | 4.64 (4.56) | <.001 | Positive |
| Sutent | 10.81 (10.22–11.42) | 10.71(10.14–11.31) | 10.48 (9.92) | 3.39 (3.3) | <.001 | Positive |
| Methotrexate | 4.07 (3.9–4.26) | 4.06(3.89–4.24) | 3.95 (3.78) | 1.98 (1.91) | <.001 | Positive |
| Orencia | 3.88 (3.6–4.18) | 3.87(3.6–4.17) | 3.83 (3.56) | 1.94 (1.82) | <.001 | Positive |
| Inlyta | 11.96 (10.94–13.08) | 11.84(10.84–12.93) | 11.74 (10.74) | 3.55 (3.39) | <.001 | Positive |
| Gilotrif | 24.43 (22.22–26.86) | 23.9(21.78–26.22) | 23.72 (21.57) | 4.57 (4.36) | <.001 | Positive |
| Panitumumab | 13.17 (11.97–14.49) | 13.02(11.84–14.31) | 12.93 (11.75) | 3.69 (3.51) | <.001 | Positive |
| Everolimus | 22.67 (21.8–23.56) | 22.23(21.4–23.09) | 21.18 (20.37) | 4.4 (4.34) | <.001 | Positive |
| Cometriq | 18.91 (16.44–21.75) | 18.59(16.2–21.34) | 18.53 (16.11) | 4.21 (3.89) | <.001 | Positive |
| Periogard | 17.85 (15.03–21.19) | 17.56(14.84–20.79) | 17.52 (14.76) | 4.13 (3.71) | <.001 | Positive |
| Suboxone | 4.2 (3.68–4.8) | 4.19(3.67–4.78) | 4.17 (3.65) | 2.06 (1.85) | <.001 | Positive |
| Dactinomycin | 17.22 (12.94–22.9) | 16.95(12.8–22.45) | 16.94 (12.74) | 4.08 (3.26) | <.001 | Positive |
| Leflunomide | 3.84 (3.48–4.24) | 3.83(3.47–4.23) | 3.81 (3.45) | 1.93 (1.77) | <.001 | Positive |
| Sirolimus | 6.39 (5.48–7.45) | 6.36(5.46–7.41) | 6.34 (5.44) | 2.66 (2.39) | <.001 | Positive |
| Nicotine polacrilex | 13.6 (11.42–16.21) | 13.44(11.31–15.98) | 13.41 (11.26) | 3.75 (3.35) | <.001 | Positive |
| Folotyn | 62.41 (50.2–77.59) | 58.95(48–72.39) | 58.86 (47.34) | 5.88 (4.82) | <.001 | Positive |
| Incivek | 3.69 (3.18–4.3) | 3.68(3.17–4.28) | 3.68 (3.16) | 1.88 (1.63) | <.001 | Positive |
| Diclofenac potassium | 4.11 (3.2–5.29) | 4.1(3.19–5.27) | 4.1 (3.18) | 2.03 (1.6) | <.001 | Positive |
| Tums ultra | 8.05 (6.32–10.26) | 8(6.29–10.17) | 7.99 (6.27) | 3 (2.5) | <.001 | Positive |
Table 4.
Classes, mechanisms and adverse effects of representative drugs associated with stomatitis.
| Drug name | Drug class | Mechanism of action | Common adverse reactions | Primary indications |
|---|---|---|---|---|
| Ibrance | CDK4/6 inhibitor | blocks G1→S phase of cell cycle | Neutropenia, infections, fatigue, nausea | Treatment of HR-positive, HER2-negative advanced or metastatic breast cancer. |
| Afinitor | mTOR inhibitor | Inhibits mTOR pathway, blocks cell cycle progression and protein synthesis | Stomatitis, fatigue, nausea, abnormal liver function, pneumonitis | Postmenopausal women with advanced HR+/HER2- breast cancer after failure of letrozole or anastrozole; adults with progressive, unresectable, locally advanced or metastatic pancreatic neuroendocrine tumors (PNET); renal angiomyolipoma with tuberous sclerosis complex (TSC) not requiring immediate surgery. |
| Everolimus | mTOR inhibitor | Inhibits mTOR pathway, blocks cell cycle progression and protein synthesis | Stomatitis, fatigue, nausea, abnormal liver function, pneumonitis | As generic: same as Afinitor above; also prophylaxis of organ rejection in adult kidney transplant recipients at low to moderate immunologic risk. |
| Methotrexate | Antimetabolite | Inhibits dihydrofolate reductase, blocks DNA synthesis and cell proliferation | Nausea, vomiting, oral ulcers, hepatotoxicity, myelosuppression | Malignant diseases (e.g., acute lymphoblastic leukemia); severe, active rheumatoid arthritis; polyarticular juvenile idiopathic arthritis; severe, recalcitrant psoriasis. |
| Orencia | T-cell costimulation modulator | Binds CD80/CD86, inhibits T-cell activation | Headache, upper respiratory tract infection, nausea, infusion reaction | Moderately to severely active rheumatoid arthritis; moderately to severely active polyarticular juvenile idiopathic arthritis in patients ≥2 years; prophylaxis of acute graft-versus-host disease. |
| Suboxone | Opioid partial agonist/antagonist | Buprenorphine activates μ-opioid receptor; naloxone antagonizes opioid effects to deter misuse | Nausea, headache, constipation, sweating, withdrawal symptoms | Treatment of opioid dependence. |
| Incivek | HCV NS3/4A protease inhibitor | Inhibits hepatitis C virus replication | Rash, fatigue, anaemia, anorectal discomfort | In combination with peginterferon alfa and ribavirin for genotype 1 chronic hepatitis C. |
| Periogard | Oral antiseptic | Disrupts bacterial cell membrane, inhibits plaque formation | Tooth discoloration, taste alteration, oral mucosal irritation | Treatment of gingivitis characterized by gingival redness, swelling, and bleeding on probing, as part of a professional treatment program. |
| Nicotine polacrilex | Nicotine replacement therapy | Activates nicotine receptors, relieves withdrawal symptoms | Oral irritation, headache, nausea, hiccups | Aid to smoking cessation for relief of nicotine withdrawal symptoms, including cravings. |
| Sirolimus | Immunosuppressant | Inhibits mTOR, blocks T-cell activation and cell cycle progression | Hyperlipidemia, stomatitis, anaemia, increased infection risk | Prophylaxis of organ rejection in kidney transplant patients ≥13 years; facial angiofibroma associated with TSC; lymphangioleiomyomatosis (LAM). |
| Tums ultra | Antacid | Neutralizes gastric acid | Constipation, bloating, hypercalcemia | Relief of heartburn, acid indigestion, sour stomach, and upset stomach. |
| Diclofenac potassium | Nonsteroidal anti-inflammatory drug (NSAID) | Inhibits cyclooxygenase, reduces prostaglandin synthesis | GI discomfort, headache, dizziness, elevated liver enzymes | Depending on formulation: primary dysmenorrhea; mild to moderate pain; signs and symptoms of osteoarthritis; acute treatment of migraine with or without aura. |
| Cometriq | Multikinase inhibitor | Inhibits MET, VEGFR2, RET, etc.; suppresses tumor growth and angiogenesis | Diarrhoea, hand-foot syndrome, hypertension, fatigue | Progressive, metastatic medullary thyroid cancer (MTC). |
| Sutent | Multitarget tyrosine kinase inhibitor | Inhibits VEGFR, PDGFR, etc.; anti-angiogenic and antitumor | Fatigue, diarrhoea, hypertension, hand-foot syndrome | GIST after progression on/intolerance to imatinib; advanced renal cell carcinoma (RCC); adjuvant treatment of high-risk recurrent RCC post-nephrectomy. |
| Panitumumab | Anti-EGFR monoclonal antibody | Blocks EGFR signalling pathway, inhibits tumour cell proliferation | Rash, hypomagnesemia, fatigue, infusion reaction | RAS wild-type metastatic colorectal cancer (mCRC). |
| Leflunomide | Immunosuppressant | Inhibits pyrimidine synthesis, blocks T-cell proliferation | Diarrhoea, hepatotoxicity, hypertension, alopecia | Active rheumatoid arthritis in adults. |
| Gilotrif | EGFR/HER2 inhibitor | Irreversibly inhibits EGFR/HER2 tyrosine kinase | Diarrhoea, rash, stomatitis, paronychia | First-line metastatic NSCLC with non-resistant EGFR mutations; metastatic squamous NSCLC progressing after platinum-based chemotherapy. |
| Inlyta | VEGFR inhibitor | Inhibits VEGFR1-3, suppresses angiogenesis | Hypertension, fatigue, diarrhoea, hypothyroidism | In combination with avelumab for first-line advanced RCC; monotherapy for advanced RCC after failure of prior systemic therapy. |
| Folotyn | Antifolate antineoplastic agent | Inhibits dihydrofolate reductase, interferes with DNA synthesis | Hepatotoxicity, myelosuppression, mucositis, fatigue | Relapsed or refractory peripheral T-cell lymphoma (PTCL). |
| Dactinomycin | Antitumor antibiotic | Intercalates into DNA, inhibits RNA synthesis | Myelosuppression, nausea, vomiting, local tissue necrosis | Various solid tumours: Wilms tumour, rhabdomyosarcoma, Ewing sarcoma, metastatic nonseminomatous testicular cancer; regional perfusion for locally advanced solid tumours. |
Time-to-onset analysis
The time-to-onset of stomatitis associated with high-risk drugs was analysed using descriptive statistics, including mean, median, and interquartile range. Table 5 presents the onset time of the most significant drugs identified through logistic regression analysis. The distribution and variability of onset time were illustrated using scatter plots (Figure 5A), reflecting substantial heterogeneity among drugs. Dotplot analysis based on the Anatomical Therapeutic Chemical (ATC) classification system (Figure 5B) was performed to compare drug-related risks across categories. A Kruskal-Wallis test revealed a statistically significant difference in the onset time of stomatitis across different drug categories (P < .001). Specifically, drugs classified under the cardiovascular system exhibited a shortest onset time of stomatitis (4.86 days), followed by Various (11.5 days), antiparasitic products (15.6 days), Genito urinary system and sex hormones (27.1 days), antiinfectives for systemic use (45.4 days), antineoplastic and immunomodulating agents (85.6 days), alimentary tract and metabolism (104.7 days), nervous system (123.5 days), other drugs (370.64 days), musculo - skeletal system (567.4 days) and systemic hormonal preparations (747 days). The induction time of specific drugs are shown in Table S6.
Table 5.
Time-to-onset (TTO) summary for the selected top significant stomatitis-associated drugs from Logistic Regression Analysis.
| Drug name | Count | Mean | Median | Q1 | Q3 |
|---|---|---|---|---|---|
| Ibrance | 333 | 219.2222 | 71 | 21 | 252 |
| Afinitor | 784 | 128.523 | 20 | 10 | 70 |
| Cabometyx | 460 | 59.72826 | 25 | 12 | 53.25 |
| Sutent | 265 | 120.6151 | 26 | 13 | 77 |
| Methotrexate | 259 | 256.5251 | 12 | 5 | 121 |
| Imbruvica | 93 | 222.7312 | 44 | 13 | 178 |
| Orencia | 27 | 182.1481 | 113 | 23 | 288 |
| Actemra | 50 | 120.3 | 32.5 | 10.25 | 83.25 |
| Capecitabine | 401 | 48.53616 | 15 | 8 | 41 |
| Xeloda | 189 | 35.80952 | 12 | 7 | 33 |
| Lenvima | 107 | 42.60748 | 23 | 10.5 | 50 |
| Methotrexate sodium | 70 | 253.9143 | 8.5 | 4 | 30 |
| Inlyta | 77 | 133 | 44 | 14 | 161 |
| Gilotrif | 232 | 29.96983 | 6 | 3 | 13.25 |
| Suboxone | 35 | 191.3429 | 70 | 12 | 205.5 |
| Tarceva | 106 | 78.74528 | 14.5 | 7 | 50 |
| Nexavar | 163 | 24.17178 | 8 | 4 | 14.5 |
| Fluorouracil | 166 | 45.90964 | 15 | 7 | 36 |
| Docetaxel | 223 | 28.1435 | 7 | 5 | 14 |
| Cisplatin | 133 | 20.58647 | 13 | 8 | 24 |
| Stivarga | 78 | 21.5641 | 7.5 | 3 | 16.5 |
| Erbitux | 110 | 26.08182 | 16 | 7 | 28.75 |
| Leflunomide | 29 | 94.27586 | 28 | 13 | 83 |
| Panitumumab | 224 | 62.83929 | 14.5 | 8 | 58 |
| Everolimus | 904 | 124.2622 | 21 | 10 | 75.25 |
| Tykerb | 96 | 44.85417 | 13 | 4 | 31.5 |
| Cometriq | 38 | 60.57895 | 27.5 | 7.25 | 79.5 |
| Incivek | 57 | 33.38596 | 22 | 7 | 43 |
| Vectibix | 51 | 100.8235 | 20 | 10 | 77 |
| Cetuximab | 214 | 32.28037 | 13 | 6 | 29 |
| Nicotine polacrilex | 19 | 11.42105 | 3 | 1 | 9.5 |
| Arava | 23 | 84.43478 | 25 | 12.5 | 74.5 |
| Sirolimus | 50 | 43 | 16.5 | 7 | 48.75 |
| Alkeran | 40 | 12.725 | 6.5 | 3.75 | 10.5 |
| Folotyn | 43 | 33.06977 | 8 | 5 | 18 |
| Cabozantinib | 571 | 55.59194 | 24 | 11 | 51.5 |
| Mobocertinib | 32 | 41.09375 | 4 | 2 | 16.25 |
| Dactinomycin | 20 | 55.45 | 26.5 | 10.5 | 55.75 |
| Campto | 17 | 15.64706 | 14 | 12 | 18 |
| Zejula | 69 | 89.18841 | 22 | 13 | 104 |
| Irinotecan hcl | 25 | 95.76 | 20 | 7 | 56 |
| Irinotecan | 117 | 62.96581 | 25 | 12 | 62 |
| Fluorouracile teva | 12 | 35.16667 | 30 | 4 | 41.75 |
| Cometriq (cabozantinib) capsule | 5 | 5.8 | 5 | 3 | 6 |
| Vincristine | 58 | 32.10345 | 13.5 | 6 | 30.25 |
| Nirogacestat | 23 | 13.04348 | 7 | 4.5 | 13 |
Fig. 5.
Time-to-onset (TTO) distribution of stomatitis events. (A) Dot plot illustrating the TTO profiles for individual identified positive drugs. (B) Dot plot displaying the TTO distributions categorized by the Anatomical Therapeutic Chemical (ATC) classifications of the identified positive drugs.
Discussion
This study represents the first systematic pharmacovigilance analysis of drug-related stomatitis. Utilizing 55,737 reports from the FAERS database, a multiple signal detection strategy combining five disproportionality methods (ROR, PRR, IC, EBGM, and Fisher’s test) with multivariate logistic regression was employed, leading to the identification of multiple drugs exhibiting consistently positive associations. Through detailed profiling of representative agents, these drugs were mainly categorized into several major pharmacological classes including antineoplastic and immunomodulating agents, antimetabolites, alimentary tract and metabolism drugs, nervous system drugs, systemic anti-infective agents, and other drugs. The findings provide critical evidence-based insights for risk stratification and clinical management of drug-induced stomatitis.
Antineoplastic and immunomodulating agents exhibited the highest number of stomatitis-related reports and the strongest signal intensities, encompassing dozens of drugs with diverse mechanisms of action—underscoring the broad impact of antineoplastic therapies on the oral mucosa. mTOR inhibitors were among the agents with the strongest signals. Everolimus is indicated for advanced renal cell carcinoma, pancreatic neuroendocrine tumours, and subependymal giant cell astrocytoma associated with tuberous sclerosis complex.27 mTOR inhibitors exert antitumor effects by blocking the PI3K/AKT/mTOR signalling pathway, thereby inhibiting cell cycle progression and protein synthesis.28, 29, 30 However, this mechanism may concurrently compromise the self-renewal capacity of oral mucosal epithelial cells and interfere with mucosal barrier repair. Prior studies have confirmed that mTOR inhibitor–induced stomatitis is dose-dependent18 and closely associated with inhibition of keratinocyte proliferation.31 In addition, mTOR inhibitors may induce local inflammation and promote the release of pro-inflammatory cytokines,32 further exacerbating mucosal injury.33,34 These findings align with clinical observations of high stomatitis incidence among mTOR inhibitor users,35 which often represents a dose-limiting toxicity.36,37
CDK4/6 inhibitors represent another targeted therapy class significantly associated with stomatitis. Ibrance is approved for hormone receptor–positive, human epidermal growth factor receptor 2–negative advanced breast cancer.38 CDK4/6 inhibitors selectively inhibit cyclin-dependent kinases 4 and 6, blocking retinoblastoma protein phosphorylation and arresting the cell cycle at the G1 phase.39,40 While this mechanism effectively suppresses tumour cell proliferation, it also impairs the turnover of highly proliferative oral mucosal epithelial cells, leading to mucosal thinning and delayed repair, thereby predisposing to stomatitis.41
Multi-target tyrosine kinase inhibitors also demonstrated significant stomatitis risk in this study. Cabometyx is primarily indicated for renal cell carcinoma and hepatocellular carcinoma; Sutent is used for renal cell carcinoma, gastrointestinal stromal tumours, and pancreatic neuroendocrine tumours; Cometriq has similar indications.42 These agents exert anti-angiogenic and antitumor effects through the inhibition of multiple targets, including vascular endothelial growth factor receptors (VEGFRs) and platelet-derived growth factor receptors (PDGFRs).43 Inhibition of VEGFR can disrupt the integrity and regenerative capacity of oral mucosal microvasculature, compromising blood supply and nutrient delivery to mucosal tissues, thereby impairing mucosal barrier repair mechanisms.44 Furthermore, clinical observations and mechanistic studies suggest that these drugs may also contribute to stomatitis development through direct epithelial cell damage and induction of local inflammatory responses.17
Antimetabolites are conventional chemotherapeutic agents closely associated with stomatitis. Methotrexate is used as both an anti-rheumatic and antineoplastic agent for rheumatoid arthritis and various malignancies; capecitabine and fluorouracil are fluoropyrimidine antineoplastic drugs primarily indicated for gastrointestinal cancers such as colorectal and gastric cancer.45 These agents interfere with DNA and RNA synthesis by inhibiting dihydrofolate reductase or thymidylate synthase, thereby blocking cell division.17 Due to their rapid turnover rate, oral mucosal epithelial cells are particularly susceptible to antimetabolites, frequently developing mucositis and ulceration early in the course of treatment.46
Alimentary tract and metabolism drugs primarily comprise topical oral agents and drugs affecting the oral environment, among which chlorhexidine series drugs showed the most prominent association. The stomatitis risk associated with chlorhexidine, a widely used antiseptic mouthwash, was robustly validated in this study. Chlorhexidine exerts antibacterial effects by disrupting bacterial cell membranes and inhibiting plaque formation, but it also directly irritates oral mucosal epithelial cells.47 Prolonged or high-concentration use may lead to oral mucosal erosion and ulceration, manifesting as stomatitis.48 This finding suggests that the concentration and duration of chlorhexidine use should be strictly controlled in clinical practice to avoid excessive mucosal irritation.
Antacids such as tums also exhibited an association with stomatitis. These agents alleviate gastric discomfort through gastric acid neutralization; however, their topical use in the oral cavity or during the swallowing process may exert physical or chemical irritation on the oral mucosa. Furthermore, calcium ion deposition and alterations in oral pH may also contribute to the development of stomatitis, although the underlying mechanisms remain elusive and warrant further investigation.49
Nervous system drugs mainly include nicotine replacement therapies and opioids, with stomatitis risk primarily linked to drug formulation and local irritation. Nicotine replacement therapies such as nicotine polacrilex and nicorette, commit showed clear stomatitis signals. Nicotine lozenges or chewing gums are absorbed through the oral mucosa, and high local nicotine concentrations can directly irritate the mucosal epithelium, triggering inflammatory responses.50 Furthermore, excipients and flavouring agents may also cause chemical irritation to the oral mucosa.51 In clinical practice, appropriate usage instructions should be provided to patients to avoid prolonged retention or excessive frequency when using nicotine replacement therapies.
Opioid/buprenorphine preparations were also significantly associated with stomatitis. Suboxone is primarily used for opioid dependence. Local irritation from the drug components and excipients to the oral mucosa is likely the main cause of stomatitis associated with these sublingual formulations.52 Additionally, buprenorphine may indirectly exacerbate mucosal injury by affecting salivary secretion and altering the oral microenvironment.52 When using such drugs, attention should be paid to patients’ oral health status, and alternative routes of administration should be considered when necessary.
Systemic anti-infective agents mainly include anti-hepatitis C virus (HCV) agents, for which stomatitis risk was identified in this study. HCV protease inhibitors were significantly associated with stomatitis. Incivek and Victrelis, previously used as direct-acting antivirals for hepatitis C, are known in clinical practice to cause rash, anaemia, and oral mucosal reactions.53 The underlying mechanisms may involve direct mucosal irritation by drug metabolites, immune-related responses, or effects on the oral microbiome.54 Patients with HCV infection often present with immune dysregulation, which may increase susceptibility to oral complications when treated with these agents.
Other drug categories encompass various over-the-counter oral care products and other medications, with stomatitis risk primarily related to local irritation or immunomodulation. Mouthwashes and toothpastes exhibited significant stomatitis risk signals. Ingredients such as alcohol, flavouring agents, preservatives, and surfactants may cause chemical irritation to the oral mucosa, inducing stomatitis.55 Long-term use of alcohol-containing mouthwashes may be particularly detrimental to the oral mucosal barrier, increasing the risk of mucosal injury.56 Clinically, patients should be advised to choose alcohol-free formulations and avoid excessive frequency of use.
This study systematically investigated drug–stomatitis associations using the FAERS database and identified multiple high-risk drugs, providing important insights for clinical medication safety. However, inherent limitations of spontaneous reporting systems, such as passive surveillance, must be acknowledged. Issues including selective reporting, missing data, duplicate entries, and lack of verification may introduce bias and affect the accuracy of results. Additionally, despite adjustments made through multivariable logistic regression, the potential influence of remaining confounding factors cannot be entirely ruled out. Consequently, the statistical signals identified should be interpreted as preliminary alerts. This study focused on overall analyses of drug groups sharing the common feature of stomatitis and subtype-specific analyses for different stomatitis phenotypes were not performed, representing an important direction for future research. We emphasize that FAERS data are better suited for generating risk hypotheses than for establishing causal relationships. Future studies—including mechanistic investigations, temporal assessments, and epidemiological studies such as cohort and case–control designs—are needed to validate the positive signals. Moreover, clinically actionable guidelines should be developed within a multidimensional framework such as pharmacovigilance triad.
In summary, this study systematically identified multiple drugs strongly associated with stomatitis through multiple disproportionality analyses and logistic regression validation. Antineoplastic and immunomodulating agents represented the drug class with the most prominent stomatitis risk, particularly mTOR inhibitors, CDK4/6 inhibitors, multi-target kinase inhibitors, and antimetabolites, with mechanisms involving interference with cell proliferation, impaired mucosal repair, or local irritation. Additionally, topical oral agents (e.g., chlorhexidine mouthwash), nicotine replacement therapies, and certain antiviral drugs demonstrated clear stomatitis risks. These findings have multi-level implications. For clinicians, they facilitate precise identification of drug triggers, optimization of prescribing decisions, and early diagnosis of drug-induced stomatitis. For the field of pharmacovigilance, they contribute to building a dedicated framework for oral pharmacovigilance and supplement stomatitis risk data across the full drug spectrum. For patients, they may help reduce the incidence of stomatitis, decrease systemic complications, and improve quality of life.
Conclusion
This study first systematically identifies 222 high-risk drugs associated with drug-induced stomatitis including antineoplastic, immunomodulating agents, Antiparasitic products, cardiovascular system drugs and so on. These associations were detected through a combination of multiple disproportionality analyses and multivariate logistic regression analysis. While these findings reveal important safety signals, they reflect statistical associations rather than definitive causal relationships.
These results provide evidence for clinical risk management, rational drug use, and early intervention. However, further prospective studies and experimental validation are still needed to confirm and elucidate the underlying mechanisms of these associations.
Ethics statement
Ethical approval and informed consent were not required for this study, as all analyses were conducted using anonymized, publicly accessible data involving no direct human or animals participation.
Data availability
Data sharing is not applicable to this article as no new data were created or analyzed.
Acknowledgments
This research was funded by the National Natural Science Foundation of China (No.82100962) and Zhejiang Provincial Natural Science Foundation (No.LQ21H140003).
Author contributions
Zhichao Liu, Lianjie Peng, and Shihan Wang contributed equally to this work. They were jointly responsible for the study design, data analysis, visualization, interpretation of findings, and drafting of the manuscript. Zhenyi Liu and Leyi Hu assisted with data preprocessing, visualization, data curation, and the literature review. Huiming Wang, Mengfei Yu, and Ying Zhou provided supervision, participated in the conception and design of the study, interpreted the results, and critically revised the manuscript for important intellectual content. All authors have read and approved the final manuscript.
Conflict of interest
None disclosed.
Footnotes
Supplementary material associated with this article can be found in the online version at doi:10.1016/j.identj.2026.109662.
Contributor Information
Huiming Wang, Email: whmwhm@zju.edu.cn.
Mengfei Yu, Email: yumengfei@zju.edu.cn.
Ying Zhou, Email: zhouying071@zju.edu.cn.
Appendix. Supplementary materials
References
- 1.Stoopler E.T., Villa A., Bindakhil M., Díaz D.L.O., Sollecito T.P. Common oral conditions: a review. JAMA. 2024;331(12):1045–1054. doi: 10.1001/jama.2024.0953. [DOI] [PubMed] [Google Scholar]
- 2.Gupta A., West H.J. Mucositis (or stomatitis) JAMA Oncol. 2016;2(10):1379. doi: 10.1001/jamaoncol.2016.2103. [DOI] [PubMed] [Google Scholar]
- 3.Elad S., Yarom N., Zadik Y., Kuten-Shorrer M., Sonis S.T. The broadening scope of oral mucositis and oral ulcerative mucosal toxicities of anticancer therapies. CA: A Canc J Clin. 2022;72(1):57–77. doi: 10.3322/caac.21704. [DOI] [PubMed] [Google Scholar]
- 4.Brown T.J., Gupta A. Management of cancer therapy-associated oral mucositis. JCO Oncol Pract. 2020;16(3):103–109. doi: 10.1200/JOP.19.00652. [DOI] [PubMed] [Google Scholar]
- 5.Anderson C., Saunders D. Oral mucositis in head and neck cancer patients. Semin Radiat Oncol. 2025;35(2):271–277. doi: 10.1016/j.semradonc.2025.02.011. [DOI] [PubMed] [Google Scholar]
- 6.Jin Y., Wang J., Wang Y. Unraveling the complexity of radiotherapy- and chemotherapy-induced oral mucositis: insights into pathogenesis and intervention strategies. Support Care Cancer. 2025;33(3):195. doi: 10.1007/s00520-025-09239-6. [DOI] [PubMed] [Google Scholar]
- 7.Teoh L., Moses G., McCullough M.J. A review and guide to drug-associated oral adverse effects-Oral mucosal and lichenoid reactions. Part 2. J Oral Pathol Med. 2019;48(7):637–646. doi: 10.1111/jop.12910. [DOI] [PubMed] [Google Scholar]
- 8.Peng L., Nie S., Sun Y., Wang S., Wang Y., Liu Z. Unraveling the causal association of epigenetic age acceleration with common oral diseases and its underlying mechanisms: findings from Mendelian randomization and integrative genetic analysis. Clin Epigenet. 2025;17(1):189. doi: 10.1186/s13148-025-02005-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Angamo M.T., Chalmers L., Curtain C.M., Bereznicki L.R. Adverse-drug-reaction-related hospitalisations in developed and developing countries: a review of prevalence and contributing factors. Drug Saf. 2016;39(9):847–857. doi: 10.1007/s40264-016-0444-7. [DOI] [PubMed] [Google Scholar]
- 10.Hosonaka K., Yamaoka K., Ikeda N., et al. Disproportionality analysis of stomatitis associated with anticancer drugs using the Japanese Adverse Drug Event Report Database. Oncology. 2024;102(9):810–818. doi: 10.1159/000535331. [DOI] [PubMed] [Google Scholar]
- 11.Riad A. National Pharmacovigilance Assessment of Oral Adverse Events Following COVID-19 Vaccination in Germany (2020-2023) International Dental Journal. 2025;75(5) doi: 10.1016/j.identj.2025.100906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yu RJ, Krantz MS, Phillips EJ, Stone CA, Jr. Emerging causes of drug-induced anaphylaxis: a review of anaphylaxis-associated reports in the FDA adverse event reporting system (FAERS). J Allergy Clin Immunol Pract 2021 9(2):819–829. e2. [DOI] [PMC free article] [PubMed]
- 13.Chen X.D., Xiao K.H., Zhou C.B. Drug-induced retinal vein occlusion: a disproportionality analysis from the FDA adverse event reporting system (2004-2023) Front Pharmacol. 2024;15 doi: 10.3389/fphar.2024.1480269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wang J., Zou D., Li Y., Liu P., Guo C. Drug-induced tooth discoloration: an analysis of the US food and drug administration adverse event reporting system. Front Pharmacol. 2023;14:1161728. doi: 10.3389/fphar.2023.1161728. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Marni M., Simoens D., Romero N., Jones W.K., Kaja S. Disproportionality analysis of oral toxicities associated with PI3K/AKT/mTOR pathway inhibitors using the FAERS database. Pharmaceuticals (Basel) 2025;18(10):1580. doi: 10.3390/ph18101580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wang X., Huang Y., Zhang Y., Li X., Yang L., Wang P. Unveiling the risk: A comprehensive analysis of drug-induced aphthous ulcers using real-world data from FAERS. Medicine (Baltimore) 2026;105(10) doi: 10.1097/MD.0000000000047844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sonis S.T. The pathobiology of mucositis. Nat Rev Cancer. 2004;4(4):277–284. doi: 10.1038/nrc1318. [DOI] [PubMed] [Google Scholar]
- 18.Martins F., de Oliveira M.A., Wang Q., et al. A review of oral toxicity associated with mTOR inhibitor therapy in cancer patients. Oral Oncol. 2013;49(4):293–298. doi: 10.1016/j.oraloncology.2012.11.008. [DOI] [PubMed] [Google Scholar]
- 19.Glickman M.E., Rao S.R., Schultz M.R. False discovery rate control is a recommended alternative to Bonferroni-type adjustments in health studies. J Clin Epidemiol. 2014;67(8):850–857. doi: 10.1016/j.jclinepi.2014.03.012. [DOI] [PubMed] [Google Scholar]
- 20.Huoshen W., Xiong J., Ma X., et al. Pharmacovigilance-Based identification and mechanistic exploration of periodontitis-related drugs. J Clin Periodontol. 2026;53(1):117–126. doi: 10.1111/jcpe.70040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zhang X., Li X., Huoshen W., Li S. Identification of drugs associated with oral ulcers based on pharmacovigilance. Int Dental Journal. 2025;75(6):103957. doi: 10.1016/j.identj.2025.103957. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bate A., Evans S.J.W. Quantitative signal detection using spontaneous ADR reporting. Pharmacoepidemiol Drug Safety. 2009;18(6):427–436. doi: 10.1002/pds.1742. [DOI] [PubMed] [Google Scholar]
- 23.Szarfman A., Machado S.G., O'Neill R.T. Use of screening algorithms and computer systems to efficiently signal higher-than-expected combinations of drugs and events in the US FDA's spontaneous reports database. Drug Saf. 2002;25(6):381–392. doi: 10.2165/00002018-200225060-00001. [DOI] [PubMed] [Google Scholar]
- 24.Harpaz R., DuMouchel W., Shah N.H., Madigan D., Ryan P., Friedman C. Novel data-mining methodologies for adverse drug event discovery and analysis. Clini Pharmacol Therap. 2012;91(6):1010–1021. doi: 10.1038/clpt.2012.50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mao S., Ma F., Zeng C., et al. Drug-induced gingival diseases: insights from the FDA Adverse Event Reporting System (FAERS) database analysis. Eur J Pharmacol. 2025;1005:178094. doi: 10.1016/j.ejphar.2025.178094. [DOI] [PubMed] [Google Scholar]
- 26.Ning W., Huang X., Li S., et al. Real-world sodium fluoride safety signals: dual-database pharmacovigilance. Int Dent J. 2026;76(2):109423. doi: 10.1016/j.identj.2026.109423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hasskarl J. Everolimus. Recent Results Cancer Res. 2018;211:101–123. doi: 10.1007/978-3-319-91442-8_8. [DOI] [PubMed] [Google Scholar]
- 28.Laplante M., Sabatini D.M. mTOR signaling in growth control and disease. Cell. 2012;149(2):274–293. doi: 10.1016/j.cell.2012.03.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Peterson T.R., Sengupta S.S., Harris T.E., et al. mTOR complex 1 regulates lipin 1 localization to control the SREBP pathway. Cell. 2011;146(3):408–420. doi: 10.1016/j.cell.2011.06.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zhang H., Li G., Li J., et al. BFAR facilitates gastric cancer progression via SPRR3 activating PI3K/AKT/mTOR signal pathway. TransMed. 2026;1(1):100007. [Google Scholar]
- 31.Lambros M., Moreno J., Fei Q., Parsa C., Orlando R., Van Haute L. Transcriptome sequencing reveals the mechanism behind chemically induced oral mucositis in a 3D cell culture model. Int J Mol Sci. 2023;24(5) doi: 10.3390/ijms24055058. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Weichhart T., Säemann M.D. The PI3K/Akt/mTOR pathway in innate immune cells: emerging therapeutic applications. Ann Rheum Dis. 2008;67(Suppl 3):iii70–iii74. doi: 10.1136/ard.2008.098459. [DOI] [PubMed] [Google Scholar]
- 33.Sonis S.T., Tracey C., Shklar G., Jenson J., Florine D. An animal model for mucositis induced by cancer chemotherapy. Oral Surg Oral Med Oral Pathol. 1990;69(4):437–443. doi: 10.1016/0030-4220(90)90376-4. [DOI] [PubMed] [Google Scholar]
- 34.Wang X., Wang Y., Pang Z., et al. Role of necroptosis in pulpitis: integration of bacterial infection, immune imbalance, and oxidative stress. PeerJ. 2025;13 doi: 10.7717/peerj.20209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.de Oliveira M.A., Martins E.M.F., Wang Q., et al. Clinical presentation and management of mTOR inhibitor-associated stomatitis. Oral Oncol. 2011;47(10):998–1003. doi: 10.1016/j.oraloncology.2011.08.009. [DOI] [PubMed] [Google Scholar]
- 36.Boers-Doets C.B., Raber-Durlacher J.E., Treister N.S., et al. Mammalian target of rapamycin inhibitor-associated stomatitis. Future Oncol. 2013;9(12):1883–1892. doi: 10.2217/fon.13.141. [DOI] [PubMed] [Google Scholar]
- 37.Paplomata E., Zelnak A., O'Regan R. Everolimus: side effect profile and management of toxicities in breast cancer. Breast Cancer Res Treat. 2013;140(3):453–462. doi: 10.1007/s10549-013-2630-y. [DOI] [PubMed] [Google Scholar]
- 38.Wekking D., Lambertini M., Dessì M., et al. CDK4/6 inhibitors in the treatment of metastatic breast cancer: Focus on toxicity and safety. Semin Oncol. 2023;50(6):131–139. doi: 10.1053/j.seminoncol.2024.01.002. [DOI] [PubMed] [Google Scholar]
- 39.Shrivastava N., Chavez C.G., Li D., et al. CDK4/6 inhibition induces senescence and enhances radiation response by disabling DNA damage repair in oral cavity squamous cell carcinoma. Cancers (Basel) 2023;15(7) doi: 10.3390/cancers15072005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Wang T.H., Chen C.C., Leu Y.L., et al. Palbociclib induces DNA damage and inhibits DNA repair to induce cellular senescence and apoptosis in oral squamous cell carcinoma. J Formos Med Assoc. 2021;120(9):1695–1705. doi: 10.1016/j.jfma.2020.12.009. [DOI] [PubMed] [Google Scholar]
- 41.Long Q., Li X., Wu G., Zhang J., Li H. Oral adverse effects of CDK4/6 inhibitors among breast cancer patients: a systematic review and meta-analysis. Ann Palliat Med. 2021;10(6):6556–6563. doi: 10.21037/apm-21-1156. [DOI] [PubMed] [Google Scholar]
- 42.Tutunaru C.V., Alexandru D.O., Dracea S.A., Ungureanu L., Popa L.G., Beiu C. Cabozantinib cutaneous toxicity-comprehensive review. Life (Basel) 2025;15(1):72. doi: 10.3390/life15010072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Gotink K.J., Verheul H.M. Anti-angiogenic tyrosine kinase inhibitors: what is their mechanism of action? Angiogenesis. 2010;13(1):1–14. doi: 10.1007/s10456-009-9160-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Yuan A., Kurtz S.L., Barysauskas C.M., Pilotte A.P., Wagner A.J., Treister N.S. Oral adverse events in cancer patients treated with VEGFR-directed multitargeted tyrosine kinase inhibitors. Oral Oncol. 2015;51(11):1026–1033. doi: 10.1016/j.oraloncology.2015.09.003. [DOI] [PubMed] [Google Scholar]
- 45.Walko C.M., Lindley C. Capecitabine: a review. Clin Ther. 2005;27(1):23–44. doi: 10.1016/j.clinthera.2005.01.005. [DOI] [PubMed] [Google Scholar]
- 46.Cardoso L.M., Pansani T.N., Hebling J., de Souza Costa C.A., Basso F.G. Chemotherapy drugs and inflammatory cytokines enhance matrix metalloproteinases expression by oral mucosa cells. Arch Oral Biol. 2021;127:105159. doi: 10.1016/j.archoralbio.2021.105159. [DOI] [PubMed] [Google Scholar]
- 47.Ribeiro D.A., Bazo A.P., da Silva Franchi C.A., Marques M.E., Salvadori D.M. Chlorhexidine induces DNA damage in rat peripheral leukocytes and oral mucosal cells. J Periodontal Res. 2004;39(5):358–361. doi: 10.1111/j.1600-0765.2004.00759.x. [DOI] [PubMed] [Google Scholar]
- 48.Bellissimo-Rodrigues W.T., Menegueti M.G., de Macedo L.D., Basile-Filho A., Martinez R., Bellissimo-Rodrigues F. Oral mucositis as a pathway for fatal outcome among critically ill patients exposed to chlorhexidine: post hoc analysis of a randomized clinical trial. Crit Care. 2019;23(1):382. doi: 10.1186/s13054-019-2664-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Zhu J., Sun C., Li M, Hu G., Zhao X.M., Chen W.H. Compared to histamine-2 receptor antagonist, proton pump inhibitor induces stronger oral-to-gut microbial transmission and gut microbiome alterations: a randomised controlled trial. Gut. 2024;73(7):1087–1097. doi: 10.1136/gutjnl-2023-330168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.An N., Holl J., Wang X., Rausch M.A., Andrukhov O., Rausch-Fan X. Potential suppressive effect of nicotine on the inflammatory response in oral epithelial cells: an in vitro study. Int J Environ Res Public Health. 2021;18(2) doi: 10.3390/ijerph18020483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Hansen E., Arendt-Nielsen L., Boudreau S.A. A comparison of oral sensory effects of three TRPA1 agonists in young adult smokers and non-smokers. Front Physiol. 2017;8:663. doi: 10.3389/fphys.2017.00663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Zheng Z., Chen J., Srinual S., et al. Buprenorphine Salivary gland accumulation sustaining high oral fluid exposure and increasing the risk of streptococcus mutans biofilm formation. J Addict Med. 2025;19(5):520–528. doi: 10.1097/ADM.0000000000001401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Hézode C. Boceprevir and telaprevir for the treatment of chronic hepatitis C: safety management in clinical practice. Liver Int. 2012;32(Suppl 1):32–38. doi: 10.1111/j.1478-3231.2011.02707.x. [DOI] [PubMed] [Google Scholar]
- 54.Di Stasio D., Guida A., Romano A., et al. Hepatitis C virus (HCV) infection: pathogenesis, oral manifestations, and the role of direct-acting antiviral therapy: a narrative review. J Clin Med. 2024;13(14):4012. doi: 10.3390/jcm13144012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Çoğulu D., Aşık A., Yılmaz Süslüer S., et al. In vitro analysis of various mouthwashes: cytotoxic, apoptotic, genotoxic and antibacterial effects. Clin Oral Investig. 2025;29(4):183. doi: 10.1007/s00784-025-06261-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Moharamzadeh K., Franklin K.L., Brook I.M., van Noort R. Biologic assessment of antiseptic mouthwashes using a three-dimensional human oral mucosal model. J Periodontol. 2009;80(5):769–775. doi: 10.1902/jop.2009.080610. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data sharing is not applicable to this article as no new data were created or analyzed.