1. IntroductionOral squamous cell carcinoma (OSCC) is the most frequent malignant neoplasia of the head and neck region [1]. The risk of OSCC both within the oral cavity and the larynx is augmented by smoking and/or alcohol consumption, despite OSCCs located in the pharynx being associated with infection with human papillomavirus (HPV), especially HPV-16 and HPV-18 [2]. The underlying mechanisms of cancer development are relatively well understood based on knowledge about fundamental cellular key alterations and evidence of histopathological progression from epithelial atypia through increasing stages of dysplasia [2,3]. Accordingly, several entities providing an augmented risk of OSCC have been identified and classified as oral potential malignant disorders by the World Health Organization (WHO) [4,5].Recurrent aphthous stomatitis (RAS; synonyms: recurrent aphthous/oral ulceration and canker sore) is characterized by the repeated formation of painful ulcers in the oral cavity. Depending on the specific clinical presentation, minor aphthae, major aphthae, and herpetiform aphthae are differentiated. RAS is found among the most frequent diseases of the oral mucosa. Females are more often affected than males, and the onset is typically found during the second or third decade [6]. Both the frequency of recurrence and the duration of disease show vast interindividual differences. To date, the etiology remains widely unknown. Due to its association with Behcet’s disease, stimulated T cells and monocytes are suspected to be involved in an underlying immunopathogenic process [7,8]. Despite the fact that aphthae heal spontaneously, symptomatic therapies are available, for example local administration of corticosteroids [9]. Even though it has to be emphasized that the recent literature does not provide any evidence for an augmented risk of oral cancer in patients suffering from recurrent aphthae [4], the question arises if RAS might potentially establish tumor-promoting inflammation and trigger sustaining proliferative signaling. Both mechanisms are known to be involved in neoplasia genesis [3]. Accordingly, the purpose of the present study was to analyze if the risk of developing oral cancer differed between patients with and without recurrent aphthae. In the recent literature, this question has not yet been specifically addressed. Due to the fact that RAS is not classified as an oral potentially malignant disorder (OPMD) [4], it was hypothesized that the risk of developing OSCC was not significantly higher in patients suffering from RAS compared to individuals without recurrent aphthae.The TriNetX Global Health Research Network (TriNetX, Cambridge, MA, USA) was chosen to retrieve related data, as it provides access to a significant number of medical records from more than 120 healthcare organizations (HCOs) from 19 countries. TriNetX is a real-world database intending to enable HCOs, contract research institutes, and biopharmaceutical companies to access and exchange longitudinal clinical data and to provide state-of-the-art statistical analytics. By December 2021, TriNetX had collected electronic medical records from more than 250 million individuals. The network has previously been used to research medical topics of worldwide interest, including the coronavirus disease 2019 (COVID-19) pandemic [10,11]. 4. DiscussionThe present work aimed at assessing the risk of developing oral cancer among patients suffering from RAS compared to subjects without recurrent aphthae. This study was the first one to address this topic by retrospectively analyzing real-world data from multiple centers to investigate larger cohorts. As RAS is not classified as a premalignant condition, it was assumed that the risk of oral cancer did not differ between both cohorts. Different from the expressed hypothesis, it was found that the risk of OSCC was significantly higher in individuals with RAS in relation to subjects with no recurrent aphthae. The authors are well aware that the obtained results are in contrast with the recent literature in terms of RAS being not classified as an OPMD [4], and at least there is very limited evidence that supports the findings from the present work. A population-based frequency-matched case–control study from Taiwan found RAS in combination with dry eye syndrome to be a risk factor for oral cancer (HR = 3.41 (95% CI lower: 1.69 and upper: 6.86)), especially among females aged 50–69 years (HR = 5.56 (95% CI lower: 1.70 and upper: 18.25)) [18]. Sadraro et al. at least discussed the involvement of oxidative stress in both RAS and OSCC, which possibly links both entities regarding their pathogenesis [19]. Furthermore, there is evidence indicating that RAS could be correlated with genetic alterations [20]. Accordingly, a study from Lu et al. revealed an altered expression of IncRNA Cancer Susceptibility Gene 2 (CASC2) in patients with RAS [21]. CASC2 is known to play an oncogenic role in OSCC as well as in other malignant tumors [22,23]. Together with the results from the present study, a new hypothesis may be formulated in terms of carefully assuming RAS to be a risk factor for OSCC.However, the results cannot be interpreted critically enough, and conclusions should be made with the greatest caution. Along with the retrospective nature of the study come specific limitations which need to be addressed. First of all, neoplastic ulcers can mimic benign ulcerative lesions, including aphthae [24]. Theoretically, there is a potential risk that OSCCs or precursor lesions were wrongly diagnosed as RAS, and the initial diagnoses were not revised after the (delayed) diagnosis of OSCC.Furthermore, RAS is diagnosed based on the clinical presentation of aphthae and a positive anamnesis for their repeated recurrence. However, other entities can mimic RAS or at least present as aphthous lesions. Again, wrong diagnoses can be the result. As an example, secondary syphilis is known to potentially manifest as aphthous enanthema [25]. Even though syphilis is relatively rare in industrialized countries today, it was frequently described as a re-emerging global public health problem, particularly among men having sex with men [26,27]. Nonetheless, it is actually not classified as an OPMD [4]. Oral lichen planus (OLP), which is also classified as a premalignant condition, can come along with erosions and ulcers as well [4]. Even though those entities can be distinguished from RAS via specific diagnostic measures (serological testing in the case of syphilis and based on histopathological examination after biopsy taking in the case of OLP), it cannot be retrospectively assessed if the relevant differential diagnoses were safely excluded. For that reason, it cannot be safely excluded that cases of RAS found in the database were in fact other premalignant entities causing an augmented incidence of OSCCs among the patients in cohort I.Another potential point of criticism arises from the very limited availability of details on the smoking behavior and alcohol consumption of the included patients. Thus far, TriNetX allows for screening of subjects with nicotine and alcohol dependence or a history of such. However, the respective terms might not have been consistently defined, leading to inhomogeneities in the retrieved data. Furthermore, neither the number of cigarettes smoked per day nor the pack-years were available for smokers. From the individuals classified as alcohol dependent, it was not reported how many units of alcohol were consumed, or since which period of time. All these features do have a well investigated impact on the risk of oral cancer [28]. Accordingly, the applied approach does imply a certain risk of bias. However, it can be carefully assumed that matching for age, gender, and (history of) smoking tobacco and drinking alcohol might have leveled out distribution differences in risk factors, at least to a limited extent. Regarding data on HIV, the applied retrospective approach implies analogous limitations. Assumedly, testing for HIV has not been carried out routinely. As a result, a higher number of patients with undetected HIV infections might eventually have been allocated to cohort I compared to cohort II. This might at least theoretically have contributed to the obtained results. No sufficient data on HPV status were available due to inhomogeneities regarding the methods applied to diagnose HPV. Despite the fact that this is a limitation of the study, as HPV plays a role in the development of oral cancer, it needs to be emphasized that HPV is not associated with RAS. The use of steroids among cohort I was not consistently documented, as a proportion of the patients underwent symptomatic treatment of RAS at their dentists, and therefore outside of the HCO. Furthermore, no geographical data were available, whereby the vast majority of the HCOs within the TriNetX network are located in the United States of America and in China. Assumedly, the majority of the subjects involved origin from the respective countries. Nevertheless, inhomogeneous healthcare standards and/or patient behavior cannot be safely excluded.With the utmost caution, it can be discussed that RAS may provide proinflammatory signaling, which is known to be a potential trigger of cancer cell formation [3]. Based on the consecutive concept of cellular key alterations leading to epithelial atypia followed by dysplasia and tumor development, there is at least a theoretical possibility that recurrent aphthae might contribute to oral cancer. Basically, this was the idea behind the present work, even though the retrieved results were not expected. However, certain aspects which were in the focus of research regarding the etiology of RAS should be discussed concerning the presented findings. Again, the authors highlight that the retrieved results do not allow for specific conclusions, but only for careful assumptions to be made. Recurrent aphthae were partially found to be associated with deficiency syndromes, specifically celiac disease and B12 deficiency [29]. In the recent literature limited evidence is available, indicating an augmented risk of gastrointestinal cancer related to celiac disease [30,31]. Nevertheless, a correlation between B12 deficiency and cancer remains a matter of strong controversy [32]. Despite deficiency diseases, an association of RAS with periodic fever, aphthous stomatitis, pharyngitis, and adenitis syndrome (PFAPA) was discussed [29]. However, no evidence for an augmented risk of cancer was found for PFAPA in the recent literature. Other studies on the etiology of recurrent aphthae investigated a potential genetic background. In particular, the interleukin (IL)-1 beta polymorphism was described to be potentially related to RAS [33]. Accordingly, an association of different polymorphisms with different forms of cancer has been discussed [34,35]. Nevertheless, recent studies indicate a correlation of the IL-4 gene but not the IL-1 beta gene polymorphism with oral cancer [36].Despite the problems with putting the findings into the context of the available literature, there is one conclusion which appears to be legitimate. Delayed diagnosis of OSCC or its precursors leads to the necessity of more aggressive therapies and decreases the survival rate of the affected patients [37,38]. Hence, ulcers in the oral cavity should undergo biopsy, at least in suspicious cases, which means persistence despite 10 to 14 days of treatment, absence of pain, and the presence of specific risk factors for OSCC. This approach appears to be appropriate for the management of oral ulcerous lesions including RAS.

In order to overcome the aforementioned limitations, future research might consider using a prospective approach applying standardized diagnostic measures, to evaluate if the presented results can be confirmed thus far.