‘Is this serious, doctor? Could it become cancer?’ These are difficult questions for the dental practitioner and they demand action.
‘I don’t think so but I will get a specialist to see you just to make sure,’ is the safe reply. Yet this approach will be counter-productive if the numbers of patients referred are excessive and thus extend waiting times. This not only delays treatment but also puts patients under great stress while they wait for the appointment.
It is well accepted that early detection of any cancer offers the patient the best chance of curative treatment and intra-oral squamous cell carcinoma (OSCC) is no exception in this regard.
In many respects, oral cancer has advantages over other malignancies, such as bowel cancer, in that the mouth is easily examined. Clearly, this puts an onus on dentists to be able to recognise the signs of malignant disease, or better still detect lesions that have a potential to become malignant. The latter, the detection of the so-called ‘pre-cancer’ stage, has been the cornerstone of recent strategies aimed at reducing the burden of OSCC. Lately, in the UK, it has even been coupled with target waiting times of two weeks from detection of a suspicious lesion to the patient being seen by an appropriate specialist (Singh P, Warnakulasuriya S, 2006). The primary care dentist is therefore faced not only with the responsibility of detecting oral cancer but also in making appropriate and timely referrals for patients with lesions that might become malignant.
Expertise in this is all a matter of clinical experience. In oral and maxillofacial and oral medicine departments, where there are a substantial number of patients referred with suspicious lesions, experience is high in diagnostic skills, including the interpretation of clinical and histological findings needed to identify lesions that present a significant risk of progressing to malignancy. The obverse is also true; primary care dentists see only a few such cases and therefore have limited experience in assessing the risk of malignant change. So, how can the primary care dentist get the right patient seen by the right specialist at the right time?
In healthcare parlance this means high sensitivity and low specificity in referred cases. Sensitivity means the number of referred cases that proved to be positive, i.e. are truly potentially malignant, and the objective would be to have as high a percentage as possible. Specificity means the reverse, i.e. of the cases adjudged innocent, the number that subsequently prove positive should clearly be as low a percentage as possible.
The latter is the nightmare of all clinicians – diagnosing a lesion as low or no risk only to find it progresses to malignancy. This must be a major factor in deciding whether or not to refer a patient. ‘Why take the risk? Best to be safe, get a second opinion,’ I hear you say but, as already stated, if the trigger for referral is set too low (poor sensitivity), then the number of cases referred becomes unmanageable and with that waiting times will increase. This ultimately defeats the objective; this is the dentist’s dilemma.
It would seem a simple problem that can be resolved by improving the sensitivity and specificity of referrals, thus getting the right patient seen on time.
The truth, however, is that although a great deal is known about the clinical appearance and histological characteristics of lesions such as leukoplakia, there are no reliable clinical or pathological criteria to predict the malignant change. Well-founded studies show the progression of leukoplakia to malignancy, particularly in cases where there is proven cytological abnormalities (dysplasia). However, percentages that transform are small, with numbers that are insufficient to deliver reproducible predictability. Hence the use of the term potentially malignant rather than pre-cancer, a term that implies that the lesion is progressing or will definitely progress to cancer (Cowan CG et al, 2001).
This, however, is not the final word and studies continue to test for reliable markers to identify the high risk lesion. These are both at the molecular and immuno-histochemical level and, hopefully, in time they will be able to resolve this key issue.
This situation is not unique to oral cancer as few, if any, cancer diagnostic pathways are completely secure. This is a concept that is hardly reassuring for patients or clinicians. The problem stems from the fact that the clinical symptoms and signs – ulceration, white or red lesions and swellings – indicative of potentially malignant disease are also common to many other innocent diagnoses, such as traumatic ulcers, frictional keratosis, candidosis and lichen planus.
In the interim, to resolve this conflict and get the right patient seen at the right time, clinicians must try to improve the sensitivity of referrals. This can be achieved to a significant degree by developing a more informed perspective of the patient and their lesion, but this in turn requires a greater understanding of key presenting features and how they ‘fit’ with the few established clinical characteristics that relate to early oral cancer.
A form of risk assessment
The high risk patient is well known: male over 40 years old, history of smoking and alcohol comsumption and more likely to be living in a large urban area. The risk is further increased if the patient comes from a disadvantaged section of society. Regarding smoking and alcohol, long-term data is important, i.e. how much for how long?
It is also important to recognise that the incidence is changing, with younger males and, in some studies, females now developing oral cancer. There is also a small but significant group of non- smoking patients developing malignancies.
It will be obvious that lesions such as leukoplakia, speckled leukoplakia and erythroplakia will be of potentially much greater significance in high risk patients, but they cannot be ignored when they present in the lower risk categories.
To evaluate the potential risk it is important to appreciate that in reacting to any stimulus the oral mucosa can only respond in a limited number of ways. It can get thicker and become white, get thinner and become red or even become so thin it breaks down to form an ulcer. Innocent lesions such as trauma can present like this, e.g. frictional keratosis (white), denture stomatitis (red) and any traumatic ulcer, sharp edge of a tooth etc. Similarly, if the stimulus is potentially carcinogenic the response can be leukoplakia (white), erythroplasia (red) and an ulcer may be malignant.
Clearly there is a wide spectrum of presentation (Napier SS et al, 2003; Scully C, Felix D, 2005). Lesions that are obviously innocent or clearly worrying are usually easy to separate, but many fall into the middle ground or ‘grey areas’ and they present the real problem. The key to improving detection lies in use of informed clinical observation, which will quickly improve sensitivity and specificity in referring cases.
This is dependent on appreciating the significance of a range of clinical factors: the type and colour of the lesion, its composition and texture, its site and its size. Interpretation of these key points is fundamental in making informed clinical comment.
Type and colour
Most studies demonstrate that only a minority of leukoplakias (15-20%) show dysplasia. Within this group only the minority progress to cancer, roughly stated 20% of 20% with proven dysplasia (Cowan CG et al, 2001). Erythroplasia, however, is always dysplastic and has a high incidence of malignant change. In many cases it is already malignant.
Clearly the latter is the most significant finding, although rarely found in advance of malignant disease, but what is of importance is the mixed red and white lesion, speckled leukoplakia or erythro-leukoplakia. This is of much greater significance than pure leukoplakia. So the ‘risk’ is high for erythroplakia and speckled lesions. Note that smokers may have ‘innocent’ lesions that have no risk of change, e.g. smokers’ palate (Figure 1), but this indicates change in the oral mucosa and it may be that the patient already has other significant lesions or these may develop later.
There is nothing really new here but if the clinician can extend his/her examination by evaluating the lesion in terms of its composition and texture, site and size then not only will sensitivity in referral be improved but the dentist will also be much better placed to inform and reassure patients.
Composition and texture
The composition and texture of the lesion are of great significance and are often missed out by clinicians when describing the lesions.
The whiteness in leukoplakia is a function of its thickness and can often be uniform throughout the area. This is described as a homogeneous leukoplakia; it indicates a degree of ‘order’ in its growth and is a favourable sign (Figure 2). The obverse of this is the heterogeneous lesion (Figure 3), which exhibits varying thickness and suggests disorder. It is typically seen in speckled leukoplakia, which is a more worrying type than the homogeneous variant.
Assessment of a potentially malignant lesion requires not only observation of the homo or heterogeneity but also palpation to determine any degree of induration (firmness) that would suggest aggressive behaviour – a worrying sign.
Site is of particular importance with potentially malignant lesions and this stems from two features of primary intra-oral squamous cell carcinoma.
Anatomical site is the first; clearly the most common sites for oral cancer to develop will be the most worrying in which to find potentially malignant lesions. It is well known that the dependent areas in the mouth, mucosa lying below the occlusal plane, constitute approximately 20% of the area of the oral mucosa yet it is the site for 80% of oral cancers.
The second modifying factor is the type of mucosa. There are two varieties in the oral cavity: keratinised or masticatory mucosa. Hard palate, dorsum of tongue and gingivae, and non-keratinised make up the remainder floor of mouth, ventral surface of tongue, retro molar area, buccal tissues and soft palate.
OSCC rarely arises from keratinised mucosa, probably because it is tough tissue designed to cope with the trauma associated with mastication whereas non-keratinised mucosa is thinner, more permeable to carcinogens and the common site for this type of malignancy. Obviously where the lesion extends to cover both types the involvement of the non-keratinised tissue is the key finding.
The size of the suspicious area does matter and the larger the lesion the greater the risk (Napier SS et al, 2003). In addition, the presence of multiple lesions is significant and can indicate field change where the entire mucosa is potentially at risk.
The objective is to build an individual profile for each case, starting with general ‘patient factors’: age, gender and tobacco and alcohol usage.
Lesions that are red or speckled and arise from non-keratinised mucosa in the key sites – ventral surface of tongue, floor of mouth, retro molar area and lower sulcus areas – have a significant potential for malignant change. This is increased when the lesion is large, heterogenous and any degree of induration is an additional and worrying sign. Conversely, white and red areas on keratinised sites have low potential for change. This is further enhanced when these lesion are small and heterogeneous.
In this short paper it is not possible to cover the pathology and differential diagnosis related to potentially malignant lesions and for further information readers are referred to standard texts and recent articles in the British Dental Journal (Scully C, Felix D, 2005).
It does however highlight the value of improving clinico-pathological based observation through which the dentist can add more substance to his/her interpretation of the risk of malignant change. This can then be used not only to trigger an urgent referral but also to assure quality information in the referral letter, which in turn will allow the specialist to make a much more informed choice on the timing of the appointment.
Informed clinical observation will also go a long way to help answer the questions that so often trigger the dentist’s dilemma: ‘Is this serious, doctor? Could it become cancer?’
Cowan CG, Gregg TA, Napier SS, McKenna SM, Kee F (2001) Primary intra-oral squamous cell carcinoma and potentially malignant oral lesions in Northern Ireland: a twenty-year population based perspective of malignant transformation. Oral Diseases 7: 18-24
Napier SS, Cowan CG, Gregg TA, Stevenson M, Lamey PJ, Toner PG (2003) Potentially malignant oral lesions in Northern Ireland: size (extent) matters. Oral Diseases 9(3): 129-37
Scully C, Felix D (2005) Oral medicine – update for the dental practitioner. Oral white patches. Br Dent J 199: 565-572
Scully C, Felix D (2005) Oral medicine – update for the dental practitioner. Oral red and pigmented lesions. Br Dent J 199: 639-645
Singh P, Warnakulasuriya S (2006) The two-week wait cancer initiative on oral cancer; the predictive value of urgent referrals to an oral medicine unit. Br Dent J 210: 717-720