The association between skin allergy testing and oral squamous cell carcinoma in oral lichen planus: A retrospective cohort study

doi:10.21203/rs.3.rs-4670209/v1

There is a reported association between oral contact allergy and oral lichen planus (OLP). Likewise oral squamous cell carcinoma (oSCC) is associated with OLP. However, risk factors for malignant transformation are unclear and it is not known if oral contact allergy impacts OLP disease severity or subsequent risk of oral squamous cell carcinoma (oSCC). In this study, we aimed to assess the association between allergy testing positivity and oSCC in OLP. We also assessed oSCC prevalence in OLP, allergen prevalence in OLP, and associations between allergy testing and OLP phenotype. To do this, we performed a retrospective cohort evaluation investigating the association between allergy testing positivity, clinical and demographic factors, and the odds of oSCC. Patients were seen at the University of Utah from 2015–2022. The rates of oSCC and allergy testing (patch + scratch testing) results were summarized descriptively and associations between oSCC, disease phenotypes, and allergy testing results were assessed using univariable and multivariable logistic regression. Patients were selected if they had a confirmed diagnosis of oral lichen planus as determined by a dermatologist or a dentist specializing in inflammatory oral disease. Allergy testing was performed by two specialized dermatologists. oSCC diagnosis was identified via chart review and were supported by histologic data. All data was collected via retrospective chart review. Among 587 OLP patients identified, 133 were allergy tested and 77.4% were positive. Of those with a positive allergy test, 10 (25.6%) developed oSCC compared with 2 (5.1%) of patients with a negative allergy test. There was insufficient evidence to support an association between allergy positivity and oSCC (OR = 1.55, 95% CI 0.76–3.15). Of the OLP disease characteristics tracked, only OLP involving the gingiva was associated with positive allergy testing (OR = 2.54,1.11–5.81). Likewise, a higher percent of symptomatic visits was the only clinical factor associated with oSCC (median 87.5% vs 66.7%, p = 0.03). These data suggest that allergy test positivity may not be associated with oSCC. However, it was associated with more pervasive symptomatology, suggesting a possible association with more severe or recalcitrant disease.

Autoimmune

medical dermatology

lichen planus

oral lichen planus

patch testing

oral squamous cell carcinoma

epidemiology

dermatology

Oral lichen planus (OLP) is a chronic inflammatory condition that is more common in women and tends to occur in patients aged 30–60 years old. Approximately 70% of cutaneous LP cases have OLP involvement, but cutaneous involvement is only seen in 20% of primary OLP.^1,2 OLP subtypes include reticular, plaque-like, papular, erosive, ulcerative, atrophic, and bullous.^3,4 Reticular OLP is often asymptomatic and is the most common subtype. Erosive OLP is characterized by atrophic or erythematous ulcerations and can portend to a more severe disease course and malignancy.^1,3 Though diagnosis can often be made clinically, biopsy is important to rule out other conditions or malignancy.¹

OLP is a T-cell driven disease characterized by an unknown trigger leading to CD8 + T-cell mediated epithelial damage.⁵ Associations or potential triggers include viruses, allergy, and stress.^2,3,6–8 Multiple case-series have emerged documenting cutaneous allergy in OLP.^9–15 One study has proposed that contact allergy, at least to metal, could initiate pathogenesis.¹¹ However, it is not generally known to what extent contact allergy drives pathogenesis or if avoidance improves disease. The most common allergens reported are dental metals, but other allergens include dental resins, and fragrance mixes.^9–15

The association between OLP and oral squamous cell carcinoma (oSCC) is well documented.^8,16–23 Two large, multi-center retrospective cohort studies found oSCC to occur in 3.1% of OLP cases 20 years after diagnosis and 2.58% at 33 years after OLP diagnosis. Patients with OLP were 4.8 times more likely to have oSCC than matched population referents and this risk was notably highest in patients with the erosive subtype.^16,21 Patients with OLP who develop oSCC often lack classic risk factors like tobacco use.^8,18 Previous studies of SCC have reported limited but intriguing evidence to suggested a connection with cutaneous allergy.^24,25 In this study, our primary aim was to investigate if positive allergy testing is associated with oSCC in patients with OLP. Our secondary objectives included characterizing allergy test positivity results, associations between allergy test positivity and disease characteristics, oSCC prevalence, and the association between OLP clinical disease characteristics and the likelihood of developing oSCC in our cohort.

Study design

We performed a retrospective cohort evaluation of medical record data.

Patient identification

We queried the Epic electronic medical record at the University of Utah from 01/01/2015 to 09/01/2022 for International Classification of Diseases, Tenth Revision (ICD-10) codes for lichen planus, lichenoid dermatitis, mucositis, and mouth pain. Since a unique oral lichen planus ICD-10 code for oral lichen planus does not exist, we used these codes to identify possible cases, and identified records were manually screened for confirmed diagnosis of OLP made by a dermatologist or dentist specializing in oral inflammatory conditions. We excluded patients with no documented dermatologic or dental follow up, and cases where histologic exams were not supportive or clinician-asserted diagnosis. For analysis of our primary outcome, only patients who received allergy testing (patch + scratch testing) were included.

Outcomes

The primary outcome was the association between oSCC and allergy testing results. Secondary outcomes included allergy testing positivity, chemicals identified, association between OLP disease characteristics and allergy testing positivity, oSCC prevalence, and associations between OLP disease characteristics and oSCC occurrence in our cohort of OLP patients.

Patient and disease characteristics

Patient characteristics included: age, gender, race/ethnicity, and potential medical comorbidities that could contribute to oSCC including tobacco history (smoking, chewed), HPV vaccination, radiation exposure, HIV, hepatitis, or diabetes. OLP disease characteristics included: average patient-reported severity (scaled 1–10, 1 being clear 10 being worst possible), percent of symptomatic visits, extensive disease (affecting ≥ 50% of oral cavity in ≥ 50% of visits), erosive disease during any visit, erosive disease in ≥ 50% of visits, presence of erythema, and area(s) of involvement. Percent of symptomatic visits = ((# of symptomatic visits/total # of visits)*100) and average patient-reported severity = (sum of scores/# of visits).

Patch testing methodology

Our department performs both patch testing and scratch testing, collectively called “allergy testing.” All tests were performed by two dermatologists specialized in these techniques (DP and RD). Patch testing was performed in the standard manner, with an initial reading performed 48 hours after allergen application and a second reading at 96 hours. For each patient, allergens were applied using Chemotechnique IQ chambers with standard, dental metal, and oral/preservative series (Supplement). A positive response was considered: 1+ (papules, erythema), 2+ (erythema, papules, infiltration, discrete vesicles), or 3+ (coalescing vesicles, bullae, erosion) reaction. For patients with 1+, 2+, or 3 + reactions, only those who had continued response on both 1st (48 hrs) and 2nd (96 hrs) reads were considered positive. Patients with borderline “+/-“ reactions were included only if this was present on the 2nd read and/or if the patch testing clinician felt that this was clinically relevant. Scratch testing was performed to assess type I urticaria reactions as this has been reported to occur in the setting of negative patch testing in clinically significant allergy.^26,27 This was performed on the ventral forearms using a Greer Pick, with histamine and glycerin controls. Wheals ≥ 3 mm larger than the negative control were considered positive.

Statistical approach

Patient characteristics, OLP clinical characteristics, allergy testing status (positive to any relevant chemical versus negative), and oSCC data were summarized descriptively using counts/frequencies for categorical variables and median, interquartile range, and range for continuous variables.

Comparisons between patients with oSCC and those without were assessed using univariable logistic regression for categorical variables and quantile regression for continuous predictors (median values compared across oSCC groups). The primary association between oSCC occurrence and allergy testing status (positive to any allergen(s) versus negative testing) was assessed using univariable and multivariable logistic regression.

Model 1: oSCC ~ Allergy testing status

Model 2: oSCC ~ Age + allergy testing status

Model 3: oSCC ~ Age + sex + allergy testing status

Univariable logistic regressions were used to assess unadjusted associations between disease severity measures and allergy testing results and well as oSCC occurance. All tests were two-sided, p ≤ 0.05 was considered significant, and all analyses were performed using Stata v14.2 (StataCorp LLC, College Station, TX).

After removing non-OLP diagnoses, we identified 599 records initially consistent with OLP. After removing patients without follow-up or a confirmed different diagnosis, 587 patients were included (Figure 1). Patient and disease characteristics are shown in Table 1. Patients with OLP had a median age of 60 (interquartile range [IQR] = 51.0-68.0), 73.2% were female, and 92.2% were white. The median average patient-reported severity score was 4 (out of 10) (IQR=3.0-5.0), the median percent of symptomatic visits was 66.7% (IQR=40.0-100%), 25.3% of patients had erosive disease in the majority (≥50%) of visits, and 12.0% of patients had extensive disease (≥50% of mouth bilaterally involved). OLP involved the gingiva (315/587, 53.9%), tongue (248/587, 42.5%), buccal mucosa (473/587, 81.0%), palate (99/587, 17.0%), and mucosal lips (176/587, 30.2%).

We identified 39/587 patients with oSCC (6.64%). oSCC involvement sites included: tongue (n=14, 35.9%), lip (n=12, 30.8%), buccal mucosa (n=9, 23.1%), and palate (n=4, 10.3%). One of these was in-situ (5.0%), 8 were T1 (40.0%), 6 T2 (30.0%), 1 T3 (5.0%), and 4 T4a (20.0%). No cases were metastatic, most were N0 (18, 78.3%), but 2 (8.7%) were N1, 1 (4.3%) N2b, and 2 N3b (8.7%). Of the 39 identified patients, 12.8% (5/39) had died at the time of chart review. Patients developed oSCC a mean of 9.68±10.7 years after OLP diagnosis. The mean duration of OLP follow-up was 7.07±7.67 years.

Results of allergy testing, including reaction severity, are presented in Table 2. We identified 134 patch-tested OLP patients and 133 had available results. Of these 133, 22.6 % (30/133) had positive scratch testing and 70.7% (94/133) positive patch testing, with some patients testing positive in both modalities. When combined, 77.4% (103/133) had a positive allergy test result to one or more allergens. The most common allergens identified were metals (54/103, 52.4%), non-metal dental materials (11/103, 10.7%), flavorings/fragrances (57/103, 55.3%), and preservatives (49/103, 47.6%). Notable individual allergens included: nickel (n=26), palladium (n=13), gold (n=10), cobalt (n=6), acrylates (n=6), mercury/mercuric chloride (n=5), amalgam (n=6), Balsam of Peru (n=30), benzoic acid (n=23), cinnamic aldehyde (n=17), fragrance mix (n=17), dodecyl gallate (n=9), and carvone (n=4) (Table 2). Of patient-supplied products tested, toothpastes, coffee, cigarettes, and lip balm were among allergens.

Associations with oSCC

Among the 39 patients with oSCC, 30.8% (12/39) had allergy testing and 10 had positive results. Of those, 7/10 had a positive reaction to a metal, 2/10 to non-metal dental materials, 4/10 to fragrance/flavorings, and 5/10 to preservatives (patients could have allergies to more than one allergen). We found insufficient evidence to support an association between allergy testing positivity and oSCC (OR=1.55, 95% CI 0.76-3.15). This was unchanged after adjusting for age (model 1, aOR= 1.52, 95% CI 0.31-7.35) or age + sex (model 2, aOR=1.66, 0.34-8.17) though caution should be applied to adjusted results as our number of events was low. In the overall cohort, patients with oSCC were more likely to be male (OR 2.52, 95% CI 1.30-4.86) and had a higher percent of symptomatic OLP visits (median 87.5% vs 66.7%, p=0.03). We did not find evidence for differences in age, traditional oSCC risk factors (tobacco use, alcohol consumption, immunosuppression, radiation history, diabetes, HIV or hepatitis), average patient-reported severity, erosive disease, or extensive disease (Table 1).

Associations between allergy testing and OLP phenotype

Of all the OLP disease characteristics tracked, only OLP involving the gingiva was significantly associated with positive allergy testing (OR=2.54, 95% CI 1.11 - 5.81) (Supplementary Table 1). In regards to factors associated with allergy testing in this cohort, patients with symptoms during their initial visit (OR=4.59, 1.95-10.78), extensive disease (OR=1.8, 1.05-3.09), erosive disease during any visit (OR=2.00, 1.31-3.06), erythematous disease (OR=3.54, 2.15-5.88), involvement of the gingiva (OR=1.70, 1.14-2.54), palatal involvement (OR=1.82, 1.13-2.92) and patients receiving immunosuppressants (OR=1.49, 1.00-2.21) were more likely to get allergy testing.

Treatments

OLP treatments are presented in Supplementary Table 2. In our cohort, 96.9% (569/587) of our patients received topical therapy for OLP. Use of systemic therapies was common with 42.9% (252/587) of patients currently receiving a systemic medication at their last visit. These mediations included: methotrexate 46/252 (18.3%), hydroxychloroquine 112/252 (44.4%), azathioprine 22/252 (8.73%), mycophenolate 40/252 (15.9%), dapsone 1/252 (0.39%), cyclosporine 5/252 (1.98%), apremilast 14/252 (5.56%), and prednisone 12/587 (4.76%). While immunosuppression has been proposed as a potential risk factor for oSCC in this population,²⁸ we found no evidence for association between past or current use of immunosuppressants and oSCC (OR=1.12, 0.57-2.20).

We found that oSCC prevalence was 7% in our population. The prevalence of malignant transformation has been previously reported between 0.5-3%.^19,23 This discrepancy may stem from previously reported cohorts being population-based whereas ours was from an tertiary-care referral-based autoimmune dermatology specialist clinic which likely services a more severe patient phenotype. This is evidenced by the high rates of erosive and wide spread OLP as well as high rates of systemic therapy usage seen in our cohort. In regards to our primary endpoint, we found insufficient data to support an association between positive allergy testing results and development of oSCC. However, we only identified 12 OLP patients with oSCC who were allergy tested, limiting power.

Like prior studies, in our overall OLP population who developed oSCC we saw a low burden of traditional oSCC risk factors,^18,21,29 and no demographic or clinical risk factors were associated with oSCC except male gender and higher proportion of symptomatic visits. The association with males in our cohort could be the product of unmeasured or unreported lifestyle differences and healthcare disparities. Previous research on gender differences have been mixed with some showing no difference²¹ and others showing higher rates of oSCC in women.¹⁸ Notably, among our OLP patients with oSCC, none had previously used chewing tobacco, none were currently using tobacco, and none had hepatitis C or HIV. The association between oSCC and higher number of symptomatic OLP visits may support the role of chronic inflammation in OLP pathogenesis and emphasizing the necessity of timely diagnosis and oSCC screening. Further research is needed to further elucidate this pathway inform oSCC risk factors in the setting of OLP in order to improve oSCC screening in this population.

The high allergy test positivity rate seen in our cohort (77%) was comparable, but slightly higher, than rates seen in other OLP cohorts.^9,13–15 The most common allergens identified were dental filling materials (metals, acrylates, resins), fragrance mix, flavorings, and preservatives. In one retrospective review of 24 patients with OLP, 16/24 (67%) had positive patch test readings, with 8/16 (50%) of those being deemed “clinically relevant.” Arguably, patients who have positive allergy testing may have allergy-induced oral lichenoid reaction rather than idiopathic OLP. However, in our experience the clinical histopathologic appearance of these entities is often indistinguishable. Thus, it is possible that patients who are allergy-test positive represent a separate diagnosis (allergic mucositis). However, our subjective clinical experience is that avoidance of these allergens often improves disease but does not eliminate it. This would seemingly support an exacerbation of underlying lichenoid disease, rather than a de-facto allergic mucositis. We are currently working to formally assess these observations.

In general, past studies have focused on dental metals and some smaller series have reported associations with fragrance, flavoring, and preservative allergens.^{10,11,13,30–32} While metals and dental filling products can often be traced to a defined source, the high burden of allergy to fragrance, flavorings, and preservatives seen in our cohort is important and highlights the clinical difficulty contact allergy may present in OLP since these allergens are often less apparent, more difficult to trace, and harder to avoid. The connection between these allergy findings and clinical OLP phenotype or progression is unclear. For example, it is unclear if contact allergy can trigger OLP pathophysiology or if OLP damages the mucosal surface and creates an environment conducive to allergic disease. At least one study suggests that a predominance of CD8 + cells and the activation of pathways such as TSLP-TSLPR signaling by oral metals has been shown to increase the infiltration of T cells in epithelium – but it is unclear if this contributes to disease propagation or initiation.¹¹ In limited case reports and series, avoidance of allergens has improved OLP clinical course, but again it is unclear if this is helping the underlying disease or an overlapping allergic mucositis.^14,30,32 Further assessment evaluating the effect of elimination diet and allergen avoidance is needed to help improve understanding of allergen relevance and impact on OLP and guide clinical recommendations.

Clinically, we believe these results support the practice of allergy testing patients with OLP not responsive to initial topical therapy. The high prevalence of allergens outside of dental metals and materials suggests that broad testing may provide important insight into potential contributory allergens. Importantly, we found additional allergies to products that patients used but which may not be revealed with standard panels. This underscores the importance of a thorough allergen history and the performance of patch testing, ideally, by a contact dermatitis/patch testing expert as customization may improve test sensitivity.

This study has important limitations. Since our population is taken from a single tertiary-care referral center, patch-testing and oSCC prevalence may not be generalizable to the larger OLP population. Additionally, it is possible that some identified allergens are not relevant to OLP disease. Some allergens did not have clear relevance, but given the unclear role of allergy in OLP and the sometimes non-intuitive role of some allergens, we erred on the side of complete reporting rather than attempting to retrospectively decide relevance. Further research is needed, including investigating allergen avoidance, to improve understanding of the allergens most likely to be of import outside of dental metals and materials. Moreover, we were unable to assess whether disease characteristics – such as a history of food sensitivities or treatment-recalcitrant disease – impacted the decision to pursue allergy testing, limiting generalizability to broader OLP cohorts. Lastly, though our cohort was relatively large, oSCC is still relatively rare, and our power for assessing the association between allergy and oSCC was low. It is possible that larger, multicenter, investigations would reveal a relationship.

In this large, tertiary center-based cohort, we did not find evidence for an association between allergy test positivity and oSCC occurrence. oSCC prevalence was 7%, highlighting the importance of early diagnosis and diligent cancer screening in patients with OLP. Like other studies, we did not find evidence for associations between oSCC occurrence and traditional risk factors for oSCC such as tobacco use (smoked and chewed), hepatitis, immune suppression, radiation, diabetes, or immune suppression – highlighting the need for better understanding of oncogenesis and its risk factors in this population. Allergy testing positivity was common (77%), though association with disease severity was unclear. Further investigation of oSCC risk factors in OLP are needed, as well as data better elucidating the role of contact allergy in OLP.

OLP= Oral lichen planus; oSCC = Oral Squamous Cell Carcinoma; ICD-10 = International Classification of Diseases, Tenth Revision; IQR = Interquartile Range; OR = Odds ratio

Funding:

None

Conflicts of interest:

Authors report no conflicts of interest.

Author Contribution

VS, RS, CT, MC, performed manual chart review with direct oversight and quality control assessments by ZH. DP and RD, in addition to ZH, CH, JZ, JR, CBH, assisted with expert input when there were questions during chart review. Data analysis was performed by ZH. Initial manuscript drafts were written by VS, RS, and CT with oversight from ZH. Subsequent input and review was performed by all authors. All Authors reviewed the manuscript.

Data Availability

Data availability statement: Studies extracted, data extracted from articles, analytic code are all available by request from the authors.

Didona D, Caposiena Caro RD, Sequeira Santos AM, Solimani F, Hertl M. Therapeutic strategies for oral lichen planus: State of the art and new insights. Front Med (Lausanne). 2022;9:997190. doi:10.3389/fmed.2022.997190
Alrashdan MS, Cirillo N, McCullough M. Oral lichen planus: a literature review and update. Arch Dermatol Res. 2016;308(8):539-551. doi:10.1007/s00403-016-1667-2
Krupaa Rj, Sankari Sl, Masthan KMK, Rajesh E. Oral lichen planus: An overview. J Pharm Bioall Sci. 2015;7(5):158. doi:10.4103/0975-7406.155873
Olson MA, Rogers RS, Bruce AJ. Oral lichen planus. Clinics in Dermatology. 2016;34(4):495-504. doi:10.1016/j.clindermatol.2016.02.023
Solimani F, Pollmann R, Schmidt T, et al. Therapeutic Targeting of Th17/Tc17 Cells Leads to Clinical Improvement of Lichen Planus. Front Immunol. 2019;10:1808. doi:10.3389/fimmu.2019.01808
Kurago ZB. Etiology and pathogenesis of oral lichen planus: an overview. Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology. 2016;122(1):72-80. doi:10.1016/j.oooo.2016.03.011
Mohammadi M, Abbaszadeh H, Mohtasham N, Salehiniya H, Shafaie E. The association between high‐risk human papillomavirus and oral lichen planus. Clinical & Exp Dental Res. 2023;9(1):93-99. doi:10.1002/cre2.707
Fitzpatrick SG, Hirsch SA, Gordon SC. The malignant transformation of oral lichen planus and oral lichenoid lesions: a systematic review. J Am Dent Assoc. 2014;145(1):45-56. doi:10.14219/jada.2013.10
Tiwari SM, Gebauer K, Frydrych AM, Burrows S. Dental patch testing in patients with undifferentiated oral lichen planus. Aust J Dermatology. 2018;59(3):188-193. doi:10.1111/ajd.12692
Finne K, Göransson K, Winckler L. Oral lichen planus and contact allergy to mercury. Int J Oral Surg. 1982;11(4):236-239. doi:10.1016/s0300-9785(82)80073-2
Yunizar MF, Watanabe M, Liu L, Minami N, Ichikawa T. Metal Allergy Mediates the Development of Oral Lichen Planus via TSLP-TSLPR Signaling. J Clin Med. 2022;11(3):519. doi:10.3390/jcm11030519
Agha-Hosseini F, Gholamrezayi E, Moosavi MS. Patch test of dental materials in Oral Lichen Planus with considering the role of saliva. Sci Rep. 2021;11(1):8264. doi:10.1038/s41598-021-87778-8
Kim TW, Kim WI, Mun JH, et al. Patch Testing with Dental Screening Series in Oral Disease. Ann Dermatol. 2015;27(4):389-393. doi:10.5021/ad.2015.27.4.389
Lomaga MA, Polak S, Grushka M, Walsh S. Results of patch testing in patients diagnosed with oral lichen planus. J Cutan Med Surg. 2009;13(2):88-95. doi:10.2310/7750.2008.08017
Olejnik M, Jenerowicz D, Adamski Z, Czarnecka-Operacz M, Dorocka-Bobkowska B. The prevalence of contact hypersensitivity in patients with oral lichen planus. Postepy Dermatol Alergol. 2022;39(4):668-674. doi:10.5114/ada.2021.107549
Laniosz V, Torgerson RR, Ramos-Rodriguez AJ, et al. Incidence of squamous cell carcinoma in oral lichen planus: a 25-year population-based study. Int J Dermatol. 2019;58(3):296-301. doi:10.1111/ijd.14215
Markopoulos AK, Antoniades D, Papanayotou P, Trigonidis G. Malignant potential of oral lichen planus; a follow-up study of 326 patients. Oral Oncol. 1997;33(4):263-269. doi:10.1016/s0964-1955(97)00005-5
Rajentheran R, McLean NR, Kelly CG, Reed MF, Nolan A. Malignant transformation of oral lichen planus. Eur J Surg Oncol. 1999;25(5):520-523. doi:10.1053/ejso.1999.0689
Aghbari SMH, Abushouk AI, Attia A, et al. Malignant transformation of oral lichen planus and oral lichenoid lesions: A meta-analysis of 20095 patient data. Oral Oncol. 2017;68:92-102. doi:10.1016/j.oraloncology.2017.03.012
Tsushima F, Sakurai J, Uesugi A, et al. Malignant transformation of oral lichen planus: a retrospective study of 565 Japanese patients. BMC Oral Health. 2021;21(1):298. doi:10.1186/s12903-021-01652-7
Arduino PG, Magliano A, Gambino A, et al. Risk of Malignant Transformation in 3173 Subjects with Histopathologically Confirmed Oral Lichen Planus: A 33-Year Cohort Study in Northern Italy. Cancers (Basel). 2021;13(22):5740. doi:10.3390/cancers13225740
Eisen D. The clinical features, malignant potential, and systemic associations of oral lichen planus: a study of 723 patients. J Am Acad Dermatol. 2002;46(2):207-214. doi:10.1067/mjd.2002.120452
van der Meij EH, Mast H, van der Waal I. The possible premalignant character of oral lichen planus and oral lichenoid lesions: a prospective five-year follow-up study of 192 patients. Oral Oncol. 2007;43(8):742-748. doi:10.1016/j.oraloncology.2006.09.006
Demehri S, Cunningham TJ, Hurst EA, Schaffer A, Sheinbein DM, Yokoyama WM. Chronic allergic contact dermatitis promotes skin cancer. J Clin Invest. 2014;124(11):5037-5041. doi:10.1172/JCI77843
Sigurgeirsson B. Skin disease and malignancy. An epidemiological study. Acta Derm Venereol Suppl (Stockh). 1992;178:1-110.
Orb Q, Millsop JW, Harris K, Powell D. Prevalence and interest in the practice of scratch testing for contact urticaria: a survey of the American contact dermatitis society members. Dermatitis. 2014;25(6):366-369. doi:10.1097/DER.0000000000000086
Saluja SS, Davis CL, Chong TA, Powell DL. Contact Urticaria to Nickel: A Series of 11 Patients Who Were Prick Test Positive and Patch Test Negative to Nickel Sulfate 2.5% and 5.0. Dermatitis. 2016;27(5):282-287. doi:10.1097/DER.0000000000000211
Agha‐Hosseini F, Moosavi M, Ghaffarpour M. Investigating the factors proposed in oral lichen planus malignant transformation: A literature review. Health Science Reports. 2023;6(5):e1267. doi:10.1002/hsr2.1267
Silverman S, Gorsky M, Lozada-Nur F, Giannotti K. A prospective study of findings and management in 214 patients with oral lichen planus. Oral Surg Oral Med Oral Pathol. 1991;72(6):665-670. doi:10.1016/0030-4220(91)90007-y
Yiannias J, Elazhary R, Hand J, Pakzad S, Rogersiii R. Relevant contact sensitivities in patients with the diagnosis of oral lichen planus☆, ☆☆. Journal of the American Academy of Dermatology. 2000;42(2):177-182. doi:10.1016/S0190-9622(00)90123-3
Lomaga MA, Polak S, Grushka M, Walsh S. Results of Patch Testing in Patients Diagnosed with Oral Lichen Planus. J Cutan Med Surg. 2009;13(2):88-95. doi:10.2310/7750.2008.08017
Larsen KR, Johansen JD, Reibel J, Zachariae C, Pedersen AML. Symptomatic oral lesions may be associated with contact allergy to substances in oral hygiene products. Clin Oral Invest. 2017;21(8):2543-2551. doi:10.1007/s00784-017-2053-y

Table 1. Clinical and demographic factors and their association with oral squamous cell carcinoma. All comparisons were made using univariable logistic regression.

	Oral Squamous Cell carcinoma
Demographics/clinical factors	No (n=548)	Yes (n=39)	Odds ratio (95% CI)	P-value
Age, median (interquartile range)	60.0 (50.0-68.0)	64.0 (53.0-70.0)	4 (-1.57-9.57)	0.16
Gender
Female	408 (74.6%)	21 (53.8%)	Ref
Male	139 (25.4%)	18 (46.2%)	2.52 (1.3 - 4.86)	0.01
Race
White	498 (92.2%)	34 (91.9%)	Ref
Black	1 (0.2%)	0 (0.0%)	N/a
Asian	15 (2.8%)	0 (0.0%)
Other	26 (4.8%)	3 (8.1%)
Ethnicity
Non-Hispanic	518 (96.3%)	35 (94.6%)	Ref
Hispanic	20 (3.7%)	2 (5.4%)	1.48 (0.33 - 6.59)	0.61
Ever smoked
No	394 (72.2%)	33 (84.6%)	Ref
Yes	152 (27.8%)	6 (15.4%)	0.47 (0.19 - 1.15)	0.1
Current tobacco smoking
No	518 (94.9%)	38 (100.0%)	Ref
Yes	28 (5.1%)	0 (0.0%)	N/A
Pack years	19.5 (5.0-30.0)	35.0 (9.0-40.0)	1.00 (0.98-1.01)	0.86
Chewing tobacco use
No	536 (98.0%)	39 (100.0%)	Ref
Yes	11 (2.0%)	0 (0.0%)	N/a
Alcohol use
No	337 (62.6%)	27 (69.2%)	Ref
Yes	201 (37.4%)	12 (30.8%)	0.75 (0.37-1.50)	0.41
Vaccinated against HPV
Yes	6 (1.1%)	1 (2.6%)	Ref
No	542 (98.9%)	38 (97.4%)	N/A
History of head or neck radiation
No	538 (98.2%)	38 (97.4%)	Ref
Yes	10 (1.8%)	1 (2.6%)	1.42 (0.18 - 11.35)	0.74
History of diabetes
No	458 (83.6%)	32 (82.1%)	Ref
Yes	90 (16.4%)	7 (17.9%)	1.11 (0.48 - 2.6)	0.8
History of hepatitis C
No	536 (97.8%)	39 (100.0%)	Ref
Yes	12 (2.2%)	0 (0.0%)	N/A
History of HIV
No	546 (99.6%)	39 (100.0%)	Ref
Yes	2 (0.4%)	0 (0.0%)	N/A
Allergy testing status
Allergy Testing Negative	28 (5.1%)	2 (5.1%)	Ref
Allergy Testing Positive	93 (17.0%)	10 (25.6%)	1.51 (0.31 - 7.28)	0.61
Allergy Testing Not Performed	427 (77.9%)	27 (69.2%)	0.89 (0.2 - 3.91)	0.87
Time from estimated diagnosis to last visit	58.7 (24.0-107.1)	80.9 (40.8-129.4)	21.8 (-3.21-46.9)	0.09
Average patient-reported severity (0-10 scale), median (IQR)	4.0 (2.9-5.0)	4.5 (3.5-6.1)	1.00 (-00.11-2.11)	0.08
Symptoms during initial visit
No	85 (15.7%)	1 (2.7%)	Ref
Yes	456 (84.3%)	36 (97.3%)	6.71 (0.91 – 49.61)	0.06
Symptoms during last visit
No	241 (44.5%)	13 (34.2%)	Ref
Yes	301 (55.5%)	25 (65.8%)	1.54 (0.77 – 3.07)	0.22
Percent of visits that were symptomatic, median (IQR)	66.7 (37.5-100.0)	87.5 (50.0-100.0)	20.8 (2.27-39.3)	0.03
Erosive disease at any time
No	215 (39.4%)	13 (34.2%)	Ref
Yes	331 (60.6%)	25 (65.8%)	1.25 (0.63 - 2.49)	0.53
Erosive disease during majority of visits
No	407 (75.4%)	25 (65.8%)	Ref
Yes	133 (24.6%)	13 (34.2%)	1.59 (0.79 - 3.2)	0.19
Erythematous disease
No	185 (33.9%)	16 (42.1%)	Ref
Yes	360 (66.1%)	22 (57.9%)	0.71 (0.36 - 1.38)	0.31
Gingival involvement
No	252 (46.2%)	17 (44.7%)	Ref
Yes	294 (53.8%)	21 (55.3%)	1.06 (0.55 - 2.05)	0.87
Tongue involvement
No	316 (58.0%)	19 (50.0%)	Ref
Yes	229 (42.0%)	19 (50.0%)	1.38 (0.71 - 2.67)	0.34
Buccal involvement
No	97 (17.8%)	14 (36.8%)	Ref
Yes	449 (82.2%)	24 (63.2%)	0.37 (0.18 - 0.74)	0.01
Palate involved
No	458 (83.9%)	27 (71.1%)	Ref
Yes	88 (16.1%)	11 (28.9%)	1.07 (0.53 - 2.18)	0.85
Mucosal lips involved
No	381 (69.9%)	26 (68.4%)	Ref
Yes	164 (30.1%)	12 (31.6%)	2.12 (1.01 - 4.43)	0.05
Extensive disease (affecting >50% of mouth bilaterally)
No	482 (88.3%)	32 (84.2%)	Ref
Yes	64 (11.7%)	6 (15.8%)	1.41 (0.57 - 3.51)	0.46
On immunosuppressive medications
No	365 (66.6%)	25 (64.1%)	Ref
Yes	183 (33.4%)	14 (35.9%)	1.12 (0.57 - 2.2)	0.75

Table 2. Details of allergy testing. We collected results from 133 tests. Allergens are organized by overarching category and individual allergen. Individual allergens are presented by their corresponding clinical reaction severity. +/- reactions were included on if present at final reading and/or if clinical relevance was noted by the patch tester.

Allergen	N (%)	Allergen	N (%)	Allergen	N (%)
Metals		Preservatives		Medications
Positive	54 (64.3)	Positive	49 (62.0)	Positive	18 (37.5)
Nickle	26 (48.1)	Octyl-gallate	6 (12.2)	Benzoyl peroxide	3 (16.7)
+/-	1 (3.8)	+/-	1 (16.7)	+/-	1 (33.3)
1+	13 (50.0)	1+	5 (83.3)	1+	1 (33.3)
1+ and ST+	1 (3.8)	Propyl-gallate	1 (2.04)	2+	1 (33.3)
2+	4 (15.4)	2+	1 (100.0)	Neomycin	8 (44.4)
2+ and ST+	1 (3.8)	Dodecyl-gallate	8 (16.3)	+/-	2 (25.0)
ST+	6 (23.1)	+/-	2 (25.0)	1+	5 (62.5)
Aluminum	3 (5.56)	1+	6 (75.0)	3+	1 (12.5)
1+	1 (33.3)	Ammonium Persulfate	10 (20.4)	Sulphur	1 (5.56)
ST+	2 (66.7)	+/-	5 (50.0)	1+	1 (100.0)
Amalgam	6 (11.1)	1+	3 (30.0)	Polymyxin	1 (5.56)
+/-	1 (16.7)	2+	2 (20.0)	1+	1 (100.0)
1+	3 (50.0)	Methylisothiazolinone	5 (10.2)	Benzocaine	1 (5.56)
2+	1 (16.7)	1+	2 (40.0)	1+	1 (100.0)
3+	1 (16.7)	2+	2 (40.0)	Bactroban	1 (5.56)
Mercury	5 (9.26)	3+	1 (20.0)	+/-	1 (100.0)
+/-	1 (20.0)	Paraphenylenediamine	1 (2.04)	Estradiol	1 (5.56)
1+	3 (60.0)	1+	1 (100.0)	+/-	1 (100.0)
3+	1 (20.0)	Thimerosol	1 (2.04)	Bacitracin	2 (11.1)
Palladium	13 (24.1)	2+	1 (100.0)	1+	2 (100.0)
+/-	1 (7.7)	Sorbitan Sesquioleate	1 (2.04)	Triple antibiotic	2 (11.1)
1+	7 (53.8)	+/-	1 (100.0)	1+	2 (100.0)
2+	2 (15.4)	Formaldehyde	1 (2.04)	Triclosan	1 (5.56)
3+	2 (15.4)	1+	1 (100.0)	1+	1 (100.0)
ST+	1 (7.7)	Ethylenediamine dihydrochloride	1 (2.04)	Glutaraldehyde	1 (5.56)
Cobalt	6 (11.1)	+/-	1 (100.0)	+/-	1 (100.0)
1+	2 (33.3)	Quaternium	1 (2.04)	Zilatinib	1 (5.56)
ST+	4 (66.7)	1+	1 (100.0)	1+	1 (100.0)
Titanium	2 (3.70)	DMD Hydantoin	1 (2.04)	Aquaphor	1 (5.56)
1+	2 (100.0)	1+	1 (100.0)	1+	1 (100.0)
Indium	5 (9.26)	Imidazolidinyl urea	1 (2.04)	Hydroquinone	2 (11.1)
+/-	2 (40.0)	1+	1 (100.0)	+/-	2 (100.0)
1+	2 (40.0)	Bronopol	1 (2.04)
ST+	1 (20.0)	+/-	1 (100.0)	Patient-provided items
Zinc	1 (1.85)	Diazolidinyl urea (Germall II)	1 (2.04)	Positive	6 (16.7)
1+	1 (100.0)	1+	1 (100.0)	Shellac	2 (33.3)
Lead	1 (1.85)	Mixed dialkyl thioureas	1 (2.04)	1+	1 (50.0)
1+	1 (100.0)	1+	1 (100.0)	2+	1 (50.0)
Chromates	3 (5.56)	Sodium metabisulfite	1 (2.04)	Aquafresh toothpaste	1 (16.7)
+/-	1 (33.3)	2+	1 (100.0)	1+	1 (100.0)
1+	1 (33.3)	Propelene glycol	2 (4.08)	Arm and Hammer toothpaste	1 (16.7)
ST+	1 (33.3)	1+	2 (100.0)	1+	1 (100.0)
Copper	1 (1.85)	Wool alcohol	1 (2.04)	Home sunscreen	1 (16.7)
1+	1 (100.0)	+/-	1 (100.0)	1+	1 (100.0)
Ammonium Hexachloropatinate	1 (1.85)	Grotan BK	1 (2.04)
1+	1 (100.0)	+/-	1 (100.0)	Other
		Benzoic acid	23 (46.9)	Positive	15 (33.3)
Dental (non-metal)		1+	4 (17.4)	Compositae	2 (13.3)
Positive	11 (26.8)	ST+	18 (78.3)	+/-	1 (50.0)
Acrylates	6 (54.5)	+/- and ST+	1 (4.3)	1+	1 (50.0)
1+	3 (50.0)			Textile dye	1 (6.67)
2+	3 (50.0)	Flavorings and fragrances		1+	1 (100.0)
Tregma	1 (9.09)	Positive	57 (65.5)	Propolis	3 (20.0)
1+	1 (100.0)	Balsam of Peru	29 (50.9)	+/-	1 (33.3)
Clove Oil	2 (18.2)	+/-	4 (13.8)	1+	2 (66.7)
1+	2 (100.0)	1+	14 (48.3)	Carmine	3 (20.0)
Epoxy Resins	2 (18.2)	2+	1 (3.4)	1+	3 (100.0)
1+	1 (50.0)	ST+	10 (34.5)	Toluenesulfonamide formaldehyde resin	2 (13.3)
2+	1 (50.0)	Carvone	4 (7.02)	1+	2 (100.0)
		+/-	1 (25.0)	Carba mix	1 (6.67)
		1+	2 (50.0)	+/-	1 (100.0)
		2+	1 (25.0)	p-tert-Butylphenol formaldehyde resin	1 (6.67)
		Fragrance mix	17 (29.8)	+/-	1 (100.0)
		+/-	5 (29.4)	Ethyleneurea, Melamine formaldehyde mix	1 (6.67)
		1+	12 (70.6)	1+	1 (100.0)
		Vanillin	1 (1.75)	Ammonium molybdate tetrahydrate	1 (6.67)
		+/-	1 (100.0)	1+	1 (100.0)
		Americhol	3 (5.26)
		1+	3 (100.0)
		Peppermint	1 (1.75)
		1+	1 (100.0)
		Coffee	1 (1.75)
		1+	1 (100.0)
		Anethole	2 (3.51)
		1+	2 (100.0)
		Eugenol	2 (3.51)
		+/-	2 (100.0)
		Orange oil	1 (1.75)
		2+	1 (100.0)
		Honey	1 (1.75)
		+/-	1 (100.0)
		Limonene	2 (3.51)
		+/-	1 (50.0)
		1+	1 (50.0)
		Linalool	1 (1.75)
		+/-	1 (100.0)
		Benzophenone	4 (7.02)
		+/-	2 (50.0)
		1+	2 (50.0)
		Cinnamic aldehyde	17 (29.8)
		1+	2 (11.8)
		ST+	14 (82.4)
		Unknown	1 (5.88)

No competing interests reported.

The association between skin allergy testing and oral squamous cell carcinoma in oral lichen planus: A retrospective cohort study

Status:

Version 1

Abstract

Figures

Introduction

Methods

Study design

Patient identification

Outcomes

Patient and disease characteristics

Patch testing methodology

Statistical approach

Results

Discussion

Conclusion

Abbreviations

Declarations

Funding:

Author Contribution

Data Availability

References

Tables

Additional Declarations

Status:

Version 1