Oral cancer and oral potentially malignant disorders.

Abstract

Oral potentially malignant disorders (OPMDs) include a variety of lesions and conditions characterized by an increased risk for malignant transformation (MT) to oral squamous cell carcinoma (OSCC) [1]. Leukoplakia and erythroplakia are the most common OPMDs, while special emphasis has been placed on the premalignant nature of oral lichen planus (OLP) [2]. It is generally accepted that the histopathological features of a given lesion, especially the presence and degree of epithelial dysplasia, are currently the most useful indicators of MT risk [3]. However, histopathological assessment alone does not provide an accurate assessment of MT risk, and other features, such as clinical and molecular parameters, must be taken into account. In this regard, the clinical characteristics of OPMDs can show considerable variation within the same histopathologically defined entity that may be critical to the likelihood of progression towards malignancy, thus, serving as prognostic factors of MT and facilitating clinical decisions for further intervention and followup. Currently, leukoplakia is defined by the World Health Organization as “a white plaque of questionable risk having excluded other known diseases or disorders that carry no risk” [4]. It is a clinical term only and histopathologically may be defined variously from atrophy, hyperplasia, to dysplasia. All frictional disorders (such as chronic cheek biting or benign alveolar ridge keratoses) are excluded by this definition. Two main clinical variants of leukoplakia are recognized: homogeneous leukoplakia with a low risk of MT and nonhomogeneous leukoplakia with a higher risk of MT [5]. The latter can be further subclassified into speckled leukoplakia (red and white but predominantly white), erythroleukoplakia (red and white but probably not predominantly white), nodular leukoplakia, verrucous leukoplakia, and proliferative verrucous leukoplakia [4]. The frequency of epithelial dysplasia, carcinoma-in-situ, or invasive SCC in leukoplakias varies from 8.6% to 60.0% [6–9]. MT of epithelial dysplasia or carcinoma-in-situ occurs in 13.6% to 36.4% of cases [6, 10], and the annual MT rate has been variably reported from 1 to 3% for all leukoplakia [6, 7, 10, 11]. It is well accepted that nonhomogeneous leukoplakia is associated with a higher risk (4- to 7-fold) for MT compared to homogeneous lesions [1–3]. The presence of an erythematous component (erythroleukoplakia) seems to convey a greater risk for MT. This is in agreement with the high malignant potential of pure red lesions (erythroplakia), which, despite its low prevalence ranging between 0.01% and 0.2%, is associated with a very high MT rate which approximates 55–65% in some studies [12]. Furthermore, the frequency of epithelial dysplasia, carcinoma-in-situ, or invasive SCC in erythroplakia is greater than 90% at first biopsy [12]. In addition, proliferative verrucous leukoplakia (PVL), a distinct entity with multiple verruciform white plaques showing a relentless tendency to expand and recur and a predilection to affect nonsmokers and especially women around 50–60 years, has been linked to a MT rate that may eventually approximate 100% [13]. Recently, optical diagnostic aids have been used to better define the clinical features of OPMD and to provide some insight into underlying cellular and molecular changes occurring in these lesions, as highlighted in the paper by Bhatia et al. in this special issue. Light-based devices at various wavelengths have been explored and show promise in assisting the clinician to detect and better visualize OPMD and oral cancer. Surgeons can also use this technology for assessment of tumour margins during surgical resection [14]. During surgical removal of malignancies, surgeons usually remove approximately 10 mm or more of normal appearing mucosal margins with the hope of achieving a margin clearance of 5 mm or more to compensate for fixation shrinkage of the formalin fixed resected specimens. Such clearance has been the routine standard used by surgeons in attempting to prevent recurrence from marginal areas with occult changes. There is thus a dependence on the pathologists' interpretation of surgical close and clear margins which have been used as predictors of tumour recurrence and survival. Despite this, there is still a high recurrence of primary tumours (up to 25%) which may result from the inability to correctly predict the molecular changes already occurring in these margins. This highlights the need to further explore adjunctive methods such as autofluorescence and narrow band imaging in a manner similar to that used for detection of OPMDs. The paper by Diajil et al. in the current issue adds credence to this approach, since laser excision of OPMD as determined by normal operatory light inspection resulted in a significant number of recurrences at the local site, and clinical resolution was most commonly seen with small and intermediate lesions compared to larger sized lesions. Furthermore, as outlined by Kudo et al. in the current issue, histology-based 3D reconstruction of serial tissue sections for evaluating tumour architecture has potential to better inform our understanding of cell invasion at the deep invasive front. Other than the importance of clinical subtyping of OPMD, the malignant potential of oral leukoplakia and erythroplakia appears to be affected by other parameters, such as site and size. The lateral border of the tongue and the floor of mouth have been correlated with the highest percentage for MT (as high as 44% and 24%, resp.) [1–3]. Despite the limited available data on the prognostic significance of the size of OPMDs, it appears that larger lesions (i.e., greater than 200 mm2) are associated with a higher risk (up to 5.4-fold) of MT [1]. Despite the aforementioned MT rates for leukoplakia, it has been known for some time now that so-called “benign hyperkeratosis” transforms to OSCC. As early as 1987, Silverman et al. [6] in the United States noted that 37 out of 235 cases of “benign hyperkeratosis” transformed to invasive carcinoma. Subsequently, Schepman et al. [7] from The Netherlands noted that MT occurred in 6 out of 20 (30%) cases of nondysplastic leukoplakia. Holmstrup et al. [15] in Denmark noted that 2% and 11% of patients with untreated or treated nondysplastic leukoplakia, respectively, developed invasive carcinoma and more recently in 2007 Hsue et al. [10] from Taiwan noted a MT rate of 3.6% in their cases although many of their patients also had submucous fibrosis from the use of betel quid. Several questions come to mind in this regard. (1) How can benign hyperkeratotic lesions transform to carcinoma? (2) Is there true “benign hyperkeratosis”? (3) Are there any features clinically or molecularly that can help to distinguish between “true” benign reactive keratosis that has no MT potential and nonreactive keratosis that does have a potential for MT? (4) If nondysplastic leukoplakia undergoes MT in at least 4% of patients, does this change the long term management of these patients?