Approach

Oral leukoplakia as defined by the World Health Organization is a predominantly white lesion of the oral mucosa that cannot be characterised as any other definable lesion.[1][2][3]​ Thus, diagnosis is one of exclusion of all other possible diagnoses of known white lesions. For many lesions an alternative diagnosis may be suggested following a thorough history and clinical examination, with confirmation by biopsy if clinical doubt exists.[59][60]

A thorough investigation and accurate diagnosis is essential because, while many leukoplakias may be benign, some have a significant potential for malignant transformation to oral squamous cell carcinoma.[61][62][63][31]​​ As such, the clinician should be aware of the importance of leukoplakia and its range of behaviours, as early diagnosis of potentially malignant lesions can also reduce morbidity and mortality.[64] Histological examination is recommended to assess the presence and degree of any epithelial dysplasia, which has conventionally been the method used to predict malignant potential.[65]

In parts of the world, dentists may not be trained to perform a biopsy; therefore, referral to a specialist for such a procedure is advised.

Leukoplakias with malignant potential

Potentially malignant oral mucosal lesions (oral pre-cancers) mainly include some leukoplakias and erythroplakia (also known as erythroleukoplakia). Erythroplakia, though much less common than leukoplakias, has higher malignant potential. Erythroplakia presents as a velvety red plaque, and at least 85% of cases show frank malignancy or severe epithelial dysplasia.[66]​ Leukoplakia, particularly where admixed with red lesions (speckled leukoplakia), is potentially malignant. By utilising a binary microscopic classification system, where lesions were classified as low-risk dysplasia or high-risk dysplasia, the latter was a significant indicator for evaluating malignant transformation.[67] Lesions including proliferative verrucous leukoplakia, sublingual leukoplakia (sublingual keratosis), and candidal leukoplakia also have a higher malignant potential, with proliferative verrucous leukoplakia having a more significant risk of malignant transformation compared with the other leukoplakia subsets. A large case-cohort study of US adults aged 65 years and older found patients with oral cancer and prior leukoplakia had a lower risk of death related to this cancer than those who developed oral squamous carcinoma without a leukoplakia precursor. The study concluded that leukoplakia identification and analysis can lead to earlier detection of cancer and improve levels of survival.[68][Figure caption and citation for the preceding image starts]: ErythroplakiaCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@55d140f6[Figure caption and citation for the preceding image starts]: Speckled leukoplakiaCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@3fd010ea[Figure caption and citation for the preceding image starts]: Proliferative verrucous leukoplakiaCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@4b5aae43[Figure caption and citation for the preceding image starts]: Sublingual leukoplakiaCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@88d4d02[Figure caption and citation for the preceding image starts]: Candidal leukoplakiaCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@377e9e59

Clinical examination

A thorough examination of the oral cavity and the regional lymph nodes is essential. Many potentially malignant oral lesions (including leukoplakia) and cancers can be detected visually but can easily be overlooked, especially if the examination is not thorough.[59][60][69]​ The possibility of widespread mucosal alterations ('field change') or a second neoplasm mandates that the whole oral mucosa be examined.[70] Molecular changes indicative of malignant potential may not produce clinically evident lesions and may extend well beyond the clinically identifiable lesion, which in part explains the increased risk for second primary tumours in patients with oral pre-cancer and cancer.[71][72][73][74][75]

Some common conditions that may cause diagnostic confusion include Fordyce's granules and geographical tongue. Collections of debris (materia alba) or fungi (candidiasis) may also look white, but these can usually easily be wiped off with a gauze pad. Other lesions appear white usually because they are composed of thickened keratin. Oral hairy leukoplakia, a lesion of viral aetiology, is seen mainly in immunosuppressed people (e.g., HIV infection) and has no known malignant potential.

White lesions are usually painless but can be focal, multifocal, striated, or diffuse, and these features may guide the diagnosis. For example:

  • Focal lesions are often caused by cheek biting, at the occlusal line

  • Multifocal lesions are common in thrush (pseudomembranous candidiasis) and lichen planus; striated lesions are typical of lichen planus

  • Diffuse white areas are seen in the buccal mucosa in leuko-oedema and in the palate in stomatitis nicotina.

Other causes of white oral lesions must be considered. Following exclusion of these causes, a diagnosis of leukoplakia can be made.[Figure caption and citation for the preceding image starts]: Homogeneous leukoplakiaCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@7cce62d[Figure caption and citation for the preceding image starts]: Oral hairy leukoplakiaCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@524fde30[Figure caption and citation for the preceding image starts]: Lip-cheek bitingCourtesy of Dr James Sciubba; used with permission [Citation ends].com.bmj.content.model.Caption@3e8cbee4

Causes of white oral lesions

[Figure caption and citation for the preceding image starts]: Causes of white oral lesionsCreated by the BMJ Knowledge Centre team [Citation ends].com.bmj.content.model.Caption@1215747d

Tests to exclude other diagnoses

Although, in many cases, such causes can be diagnosed following a careful history and physical examination, others may be confirmed only following a representative biopsy of the lesion. Additional laboratory investigation may also aid in excluding alternative diagnoses. These may include Treponema pallidum serology, and autoantibody testing where mucosal lupus erythematosus is considered possible.

First-line testing

The difficulty is in determining the malignant potential of the leukoplakia, which requires an appreciation of specific clinical forms and attendant likelihood for transformation. However, clinical diagnosis alone is insufficient in establishing and excluding malignant and pre-malignant disease; even experienced, highly trained professionals cannot identify all pre-cancers and early stage oral squamous cell carcinoma (OSCC) by visual inspection alone.[76] Furthermore, the recognised classic features of OSCC, such as ulceration, induration, elevation, bleeding, and cervical lymphadenopathy, are features of advanced disease, not early stage disease.[77] Unfortunately, clinical data suggest that there is often a substantial delay in obtaining a biopsy specimen, even when oral lesions display characteristics of frank cancers.[78][79]

Biopsy

  • An incisional biopsy is invariably indicated for most leukoplakias and should be sufficiently large and representative. An excisional biopsy should be avoided, because this is unlikely to have excised an adequately wide margin of tissue if the lesion is proven malignant, and may limit for the surgeon the clinical evidence of the site and character of the lesion.

  • False-negative results are occasionally possible from incisional biopsy and, even where dysplasia has been excluded in a leukoplakia by incisional biopsy, studies have shown that the lesions if wholly excised may prove to contain OSCC in up to 10% of patients.[80][81]​ This is explained in part by data showing that early malignant changes can be scattered through and beyond a potentially malignant clinical lesion.[72][82]​ Since red rather than white areas are most likely to show any dysplasia, a biopsy should include the former.

Pathology

  • Histopathological analysis by an experienced pathologist is a crucial aspect of management, as interobserver and intra-observer variability between pathologists is significant.[83][84][85][86][87] If, following an initial biopsy, clinical doubt remains as to whether a representative sample has been taken (e.g., if the pathology report denies malignancy and yet clinically this is suspected), a repeat biopsy is indicated.

  • On routine histological analysis leukoplakias demonstrate a wide range of histological features ranging from benign to dysplastic changes of varying degrees. Benign histological features include ortho/parakeratosis with no sign of keratin in areas deep to the surface (dyskeratosis, a feature of dysplasia). Additionally, a thickening or increase in the overall volume of the spinous or prickle cell layer (acanthosis) is commonly observed. Notably, in the majority of leukoplakias (over 60%) only hyperkeratosis with or without acanthosis will be found on microscopical analysis.[88]​ Definitive cellular atypia within the basal and more superficial layers of the epithelium are often absent. However, one study in a Thai population found 10.6% of leukoplakias were dysplastic in nature, while 4.9% were diagnosed as squamous cell carcinoma.[88]

  • Where present, the assessment of epithelial dysplasia should be graded (e.g., mild, moderate, or severe).[89]​ Intraepithelial neoplasia, a concept created for the uterine cervix and extended to other mucosae, has been adapted to the oral mucosa (oral intraepithelial neoplasia) and used by some as a diagnostic term.[90][89]​​ Microscopic epithelial changes structured under architectural changes or cellular atypical features and associated with pre-malignancy or epithelial dysplasia include the presence of the following:[5]

  • Architectural features

    • Irregular epithelial stratification

    • Loss of polarity of basal cells

    • Drop-shaped rete ridges

    • Increased number of mitotic figures

    • Abnormally superficial mitoses

    • Premature keratinisation in single cells (dyskeratosis)

    • Keratin pearls within rete ridges

    • Loss of epithelial cell cohesion.

  • Cytological features

    • Abnormal variation in nuclear size

    • Abnormal variation in nuclear shape

    • Abnormal variation in cell size

    • Abnormal variation in cell shape

    • Increased nuclear-cytoplasmic ratio

    • Atypical mitotic figures

    • Increased number and size of nucleoli

    • Hyperchromasia.

A binary grading system classifying dysplasias as low and high risk has also been proposed.[89]​ 

Adjunctive diagnostic techniques

There are a number of adjunctive diagnostic aids to assist in the clinical assessment of oral mucosal pathology. However, evidence of efficacy is lacking.[91][92]​​​

Based on an evidence-based clinical practice guideline issued by the American Dental Association (ADA), these include:[93]

  • Cytological testing: brush biopsy

  • Autofluorescence

  • Tissue reflectance

  • Vital staining

  • Salivary adjuncts

  • Additional adjuncts of interest.

A description of each of these triaging tools is given below.

Brush biopsy

  • An oral brush biopsy may be used to exclude dysplasia among common, harmless-appearing oral lesions that do not appear suggestive enough to warrant a scalpel biopsy. This test uses a small nylon brush to gather cell samples of a suspicious area, which are then sent for computer analysis. Specimen collection is simple, causes little or no pain or bleeding, and requires no anaesthetic; however, accurate sampling of the abnormality is necessary. A print-out reporting any abnormal cells and a written pathologist's report are returned to the clinician with a recommendation for a conventional incisional biopsy if significant abnormalities are detected.

  • This test has been reported to have a sensitivity greater than 92% to 96% and specificity of more than 90% to 94% in detecting epithelial dysplasia and OSCC, with a positive predictive value of 38% to 44%.[94][95][96][97][98]​ Brush biopsy has been used to detect OSCC missed in scalpel biopsy.[99] However, case reports of OSCC in patients where brush biopsy results were negative have been reported.[100]

  • Outcomes of brush biopsy may be improved with the use of molecular techniques (e.g., p53 mutations, DNA cytometry, and AgNOR [nucleolar organiser region protein counts visualised by the argyrophil technique] counts).[101][102][103][104]

Vital staining with toluidine blue

  • Toluidine blue (TB) is a metachromatic stain that is easily available, economical, and has a high affinity for DNA and RNA. It rapidly stains abnormal tissues, but not normal mucosa. Several studies have demonstrated the ability of TB to detect oral pre-malignant lesions, including oral leukoplakia and oral cancers, with a high sensitivity. However, the stain is also taken up by other ulcerative conditions and, therefore, the specificity of the technique is low.[105]

Optical systems

  • Interest has been increasing in the potential use of optical spectroscopy systems to provide tissue diagnosis in a real-time and non-invasive fashion.[106][107][108]

  • Autofluorescence-based imaging systems are simple, manual hand-held devices based on the direct visualisation of tissue fluorescence. Abnormal tissue typically appears as an irregular, dark area that stands out against the otherwise normal, green fluorescence pattern of surrounding healthy tissue.[109][110]

  • Another technique involves an acetic acid oral rinse, after which the mucosa is examined using a blue-white light source. The light source is generated by a chemical reaction, hence the term chemiluminescence. The basis of the test is that oral leukoplakia has a differential tissue reflectance, and the patch may appear 'whiter' with enhanced visibility and improved sharpness of the lesion during the examination. However, the specificity of the test is low.[109]​ Such optical adjuncts may assist in identification of mucosal lesions and in selection of biopsy sites.[111]​ It may also assist in identification of cellular and molecular abnormalities not visible to the naked eye on routine examination, and provide additional information on tissue adjacent to the mucosal lesion.[112][113]

  • Data are conflicting as to the clinical benefits; studies have shown detection rates (for dysplastic lesions) similar to those of conventional clinical diagnosis and biopsy site selection.[114][115][116][117][118] However, oral lesions visualised with these optical adjuncts require formal biopsy to determine the underlying pathological changes.

  • Photodynamic diagnosis (PDD) using 5-aminolevulinic acid (ALA-PDD) enables the visualisation of leukoplakia lesions as red fluorescence and has been shown to be successful in detecting oral disorders.[119]​ Compared with other optical systems, an improvement in the sensitivity and specificity to detect higher grades of dysplasia is described. However, more research is needed before recommending this for routine practice.

Molecular and chromosomal markers

  • The ability to predict malignant transformation subsequent to the establishment of a clinical and histological analysis of pre-malignant conditions relates to identification of a set of molecular markers that are easily and economically applicable. These techniques are useful at the necessary routine follow-up visits following diagnostic work-up.

  • Studies have yielded several molecular markers that are of potential value in helping predict future behaviour of these lesions.[21][17]​ For example, in assessment of leukoplakia involving so-called high-risk sites, there were greater levels of genetic alterations associated with an elevated risk of progression to carcinoma, by way of high losses of heterozygosity on chromosome 3p and/or 9p sites.[120] Podoplanin, a transmembrane glycoprotein, could be a valuable biomarker in the future for risk assessment of malignant transformation in patients with oral leukoplakia.[121] Additionally, the development of gene array analysis technology and other molecular tools holds promise regarding the predictability of leukoplakia lesions progressing to a neoplastic phenotype.[122]

  • DNA ploidy status remains an attractive option in evaluating oral leukoplakia where there has been demonstrated aneuploidy in associated increased risk of progression to squamous carcinoma.[23]​ Additionally, S-phase fractions of leukoplakia lesions with dysplasia demonstrated aneuploid rates more than double those of leukoplakias without dysplasia.[123] Further adding to the progress in understanding molecular carcinogenic changes in oral leukoplakia has been the addition of microarray analysis utilising reverse transcriptase polymerase chain reaction results, with candidate biomarkers identified in those cases resulting in cancer formation.[124]

Salivary biomarkers

  • Novel non-invasive tests using saliva as a body fluid for the detection of biomarkers are in development. Human salivary proteome analysis using mass spectrometry, supported by bioinformatics tools, are reported. The most studied proteins S100A7, S100P, CD44, and COL5A1 have shown increased expression in leukoplakia as compared with healthy controls.[125]​ Of these proteins, CD44 in oral rinses has been shown to have potential as a candidate marker for the early detection of oral cancer for patients who show leukoplakia lesions.[126]​ Several pro-angiogenic and pro-inflammatory cytokines such as IL-1, IL-6, IL-8, and tumour necrosis factor-alpha are also reported to be elevated in saliva samples of patients with oral leukoplakia, and elevated levels of these cytokines may indicate the carcinogenic transformation from pre-cancer to oral cancer.[127]

The ADA expert panel does not recommend vital staining, autofluorescence, tissue reflectance, or salivary adjuncts as triage tools for use in primary care.[93]

Use of this content is subject to our disclaimer