An update on the clinical pathology of oral precancer and cancer

Aetiology

Oral cancer is a classic multifactorial disease, resulting from the interplay of a number of factors, most of which are environmental in nature.6 Genetic predisposition plays a part in a small number of patients. The relative importance of these factors varies in different geographical regions or ethnic groups.

Tobacco smoking (in its various forms) is still the main causative factor. The risk is greatest in heavy users (>20 per day) and the addition of alcohol gives a greater than multiplicative effect.7 The relative risk of oral cavity cancer for current smokers is 3.43, but this is markedly dependent on dose.8 Smoking cessation markedly reduces the cancer risk, with a complete loss of the enhanced risk over a 20-year period.9 Smokeless tobacco is important, particularly when used in paan (Betel leaf with ground areca nut, lime and tobacco), which is very common in Indian and South East Asian communities. Use of paan results in a particular pattern of disease where the mucosa in contact is affected (usually the buccal mucosa). More ‘sanitized’ forms of smokeless tobacco are readily available in parts of Europe and the USA. Despite claims of much reduced harm to users, these products are still carcinogenic.10

The other main causative factor is alcohol.11 In general, the amount is more important than the type and the risk is greatest when accompanied by tobacco use and at high levels of consumption, but there are many confounding factors. In one recent study, the odds ratio in non-smokers for three or more drinks per day versus never drinking was 2.04.12 This is of increasing importance in young patients, particularly with binge consumption.13

The most recent and important addition to the list of causative factors is human papilloma virus (HPV). HPV is well established as a causative factor in cancer of the cervix. The viral genome produces a number of proteins which disrupt the cell cycle and promote oncogenesis. There is now compelling evidence for a role for HPV in cancer of the oropharynx (particularly tonsil and base of tongue), but less so in oral lesions. HPV subtypes 16 and 18 (which are also important in cancer of the cervix) have been implicated and it is assumed that this is primarily sexual transmission to the oral tissues.14 Many predictions of the overall significance of HPV have been made, but it is not clear whether this is a ‘time bomb’ of disease or if the effect will be much less marked.15 The effect of the ongoing immunization of the teenage female population is also unclear. Other infective agents, such as fungi (largely Candida) may also have a supportive role in the development of oral cancer, but the mechanism of this is not clear.

Other factors also play their part. Ultraviolet light is an important cause of lip (skin) cancer (basal cell carcinoma, squamous cell carcinoma and melanoma). Poor diet and nutrition have an undefined role in the development of many human cancers. Other issues, including the molecular basis of head and neck cancer, have been the subject of a recent excellent review.16

The natural history and pathology of precancerous lesions

The multifactorial nature of oral cancer development implies that there are numerous damaging events which affect the DNA of the cells of the oral mucosa. Some of these may be made more likely in individuals with genetic susceptibility. In general, most of the damaging events have occurred at the precancer stage. These changes in the oral mucosa can affect large areas which may not be evident clinically or even on histopathological examination. This large ‘field’ of abnormal mucosa can give rise to other tumours, even after successful treatment of the first tumour.17

The main precancerous lesions are well known: leukoplakia (Figure 1) and erythroplakia (Figure 2). Both of these terms have strict WHO definitions which indicate that they are white or red patches (respectively) which do not have any definable cause (except tobacco use).

Leukoplakia has a prevalence of around 2.5–3%, and is generally described as homogeneous or non-homogeneous (a number of other terms may be used, such as speckled, nodular, verruciform, exophytic types). The rate of malignant transformation overall is about 1% per year, but for homogeneous lesions it is much less than this.18 Non-homogeneous lesions have a progression rate of about 20% in 5 years, but the site, texture, presence of Candida and degree of dysplasia all need to be taken into account.19^,20

The prevalence of erythroplakia is much less than for leukoplakia, although good studies are few and far between. It does, however, have a malignant transformation rate of 80%, with many lesions already SCC at first biopsy.21

In addition to these lesions are a number of precancerous conditions. These are disease states in which there is an increased risk of cancer. The important ones are: chronic hyperplastic Candidosis (Figure 3), submucous fibrosis (Figure 4), lichen planus and actinic keratosis (lip).

Screening, diagnosis and treatment

The case for population screening for oral cancer has not yet been convincingly made, although oral examination of high-risk individuals may be a cost-effective screening strategy.22 A number of tools have been developed for screening and to aid in the assessment of the oral mucosa. Toluidine Blue may be used as it preferentially stains precancerous and cancerous lesions and much less so normal oral mucosa, although the assessment is somewhat subjective. Commercial systems, such as VELscope, have built on this and use differences in the tissue fluorescence in areas of abnormal mucosa. The use of such aids is controversial as there is no proven benefit over careful clinical examination.23

The mainstay of a diagnosis of oral precancer (and indeed cancer) is a good incisional biopsy. This should incorporate adjacent normal tissue and the full depth of the lesion. In general, biopsy of suspicious lesions should not be undertaken in General Dental Practice, but should be referred for diagnosis and management (Table 1 and Figure 5). Brush biopsy is a technique for sampling the oral mucosa without the incisional biopsy. It is still a traumatic procedure as cells from the deep layers of the epithelium must be sampled. This will yield cells for cytological examination and also for a number of biomarker studies.24^,25

The histology of these lesions is very variable. The white lesions almost always demonstrate hyperkeratosis, but epithelial dysplasia is far more unpredictable.18 Epithelial dysplasia is a collective term which describes all of the abnormal features in the epithelium and is graded as mild, moderate or severe, depending on the extent of the abnormality seen. This grading process is subjective, but it is still the single most useful tool for prediction of the behaviour of abnormal epithelium.26 Only about 10–20% of dysplastic lesions will progress to cancer, with the rest regressing or staying the same.27

Management of precancerous lesions centres on cessation of risk factors and surgical removal of lesions with moderate (in some units this is controversial) and severe epithelial dysplasia. Medical (non-surgical) treatments, such as retinoid therapy, may result in resolution of the lesion, but relapse and side-effects are common.28

The natural history and pathology of oral cancer

Patients with oral cancer may have no symptoms at all, which probably contributes to the large number of patients who present with advanced lesions. Some will complain of an area of soreness or irritation. The features of paraesthesia, disruption of function and dysphagia indicate advanced disease. Clinical signs to look for are listed in Table 2. This requires careful and thorough examination of intra- and extra-oral tissues. The high risk sites to which particular attention needs to be paid include the lateral border of the tongue, floor of the mouth (Figure 6) and tonsillar pillars (Table 3).

Once a biopsy has confirmed the diagnosis, a number of investigations are undertaken to assess the extent of spread of the disease and to allow planning of treatment (Table 4). The final pathology report is a vital part of assessment of prognosis for the patient.29 Important factors include:

Assessment of prognosis

The overall poor prognosis in oral cancer, which was referred to at the start of the article, is due to the fact that 70% of patients present with advanced stage disease. The majority of patients have metastatic disease at first diagnosis. The important factors which determine prognosis and have to be taken into account for each patient are:

The 5-year relative survival rate for tumours of the anterior tongue varies from 75.9% (Stage 1) to 26.5% (Stage 4), with a similar pattern for floor of mouth. Tumours of the tonsil and oropharynx have an overall poorer prognosis, but this is much less affected by stage. Recent survival data relative to stage at diagnosis is shown in Table 5.

The role of the primary care dental team

The primary care dental team has a key role to play in the early detection of oral cancer and precancerous lesions.33 This includes, but is not limited to, prompt referral of suspicious lesions and modification of behaviours, such as tobacco-smoking, as part of a smoking cessation programme. However, the identification of asymptomatic lesions can be challenging and recent analyses of referral patterns has identified a clear need for further education and standardization of criteria to inform referral decisions better.34 A key component of the current UK Cancer Strategy is the two week urgent referral pathway which ensures rapid access to specialist services for suspected cancer patients, and the primary care team should be aware of how to access this locally.

Patients with precancerous lesions can be followed up in primary care after specialist diagnosis and assessment. The recall interval will vary depending on the lesion and continuance of risk factors. Clinical photography can be very useful for the longitudinal monitoring of these lesions, with any change in the lesion noted by the practitioner or reported by the patient prompting re-referral.

Patients who have completed treatment for oral cancer may also be returned to their primary care team for follow-up. These patients will require regular review and may also need other treatments, for example, patients who have radiotherapy-associated dry mouth will require intensive preventive dentistry in addition to the ongoing assessment of the soft tissues.

Conclusion

Despite the changes in oral cancer epidemiology reported over the last two decades, the key to improving the survival rates on oral cancer is early diagnosis. Patients with suspicious lesions should be referred to a local specialist for assessment, with those suspecting cancer marked as urgent, and placed into the 2-week wait system. A combination of this with early treatment and ongoing efforts to reduce the risk factors will reduce the incidence and improve outcomes.