The natural history and prognosis of oral cancer

The majority of oral cancers are squamous cell carcinomas of the oral mucosa. Oral cancer most commonly presents as an area of erythema, leukoplakia or ulceration or as a mass in the oral cavity or at a distant site due to metastasis. It may be symptomatic (pain, bleeding, dysphagia) or an incidental finding during a medical or dental examination.

The natural history of an oral cancer can be said to extend from early intracellular changes to the development of a clinically detectable lesion, to its treatment (Fig. 1). This whole process may take several years and the time before clinical detection may represent the vast majority of the total.1

Figure 1
figure 1

Time points in oral cancer

Full size image

Current understanding of carcinogenesis suggests that the longer a cancer is present then the larger it will become and the more likely it is to metastasise.1 Rates of progression are however very difficult to predict because of variation in the velocity of growth and aggressiveness of individual tumours. Factors which affect survival include the site of the lesion, the size of the lesion at diagnosis, its degree of differentiation, the involvement of regional lymph nodes and the presence of distant metastases.2,3

Factors implicated in the aetiology of oral cancer include smoking and alcohol consumption and therefore cancer prevention may be helped by raising public awareness of the dangers associated with these activities. The time taken for a tumour to develop from a single cell to one which is clinically detectable is one area which it is difficult to quantify or affect, however the time taken from its detection to its treatment is quantifiable and it is reasonable that attempts should be made to minimise it.

Diagnostic delays

A number of previous papers have reported delays in the diagnosis of oral cancer in the United Kingdom.4,5,6,7 These have consistently shown delays, both due to delays in patients seeking attention, and delays in medical and dental practitioners referring patients for diagnosis and treatment. Similar findings have been reported from Australia,8 Brazil,9 Canada,10 Denmark11 and Greece.12

Allison13 has reviewed the evidence regarding whether delays in the diagnosis of oral cancer once it is clinically detectable actually affects the prognosis of patients, and found that whilst there is evidence that those presenting with more advanced tumours fare worse, there is surprisingly little evidence specifically linking referral delay to tumour size. As this is the one area of the natural history of oral cancer which is potentially modifiable it is an important area to study further.

Delays in cancer diagnosis arise at a number of different levels (Fig. 1), and different definitions are used by authors on the subject.

Patient delay is generally defined as the time from the patient's first awareness of a symptom to seeking their first consultation with a healthcare professional.11,13

Professional delay is defined either as the time from the first consultation with a healthcare professional to the first consultation with a treating specialist,10,13 or to the definitive diagnosis being made,8,11,12 or to the patient being admitted for definitive treatment,9 or as the time from the first consultation until a referral letter is sent to a specialist unit.4,5,6,7

We would suggest the use of professional delay for the whole time from the patient's first consultation to their commencing definitive treatment. This is made up of referral delay (time from consultation to referral being made), appointment delay (time to appointment at specialist centre) and treatment delay (time from diagnosis to definitive treatment commencing).

What is considered to be excessive professional delay also differs between authors from two days to one month, due in part to the different definitions used for professional delay.

There is no consensus on what should be considered excessive delay at each stage. The Department of Health guidelines,14 for example, only mention the length of time after a symptom or sign is noticed that referral should be made and therefore referral time will be affected by patient delay.

Schnetler6 described referral delay as excessive where there was more than two days between consultation and referral and we feel that this is not in keeping with current guidelines.

In our audit we therefore used two weeks to define excessive referral delay which permitted GMPs and GDPs to consider potential confounding factors such as irritation from a denture or local trauma when examining a small ulcer for example. This allowance of time for review by the referring practitioner is in line with current guidelines from the BAOMS,15 BDA16 and DOH.14

Developing professional and public awareness of oral cancer

The British Association of Oral and Maxillofacial Surgeons (BAOMS) co-ordinated a National Oral Cancer Awareness Week (NOCAW) in October 1995 with one objective being to raise awareness of oral cancer amongst the general public and healthcare professionals. This generated some 280 newspaper articles and almost 90 radio broadcasts as well as mailings to some 40,000 general dental practitioners (GDPs), general medical practitioners (GMPs) and pharmacists.17 In recent years a number of other sources of information and education on oral cancer available to healthcare professionals and the general public has become available.

In 1998 the British Dental Association (BDA) published guidelines on the early detection of oral cancer18, which were circulated to its members. Included in this were guidelines on the appropriate course of action when a patient attends with a suspicious lesion. An updated version was circulated in April 2000.16

The advent of formal continuing professional development in its different forms for doctors and dentists can be expected to increase both the readership of journals and uptake of courses, some of which concentrate on or broach the subject of oral cancer in one way or another. Using a Medline search (performed November 2003) we identified 901 articles in the years 1992-2000 with search criteria of subject heading 'oral cancer', and keyword 'diagnosis', limited to 'Human' and 'English language'.

An internet search for 'oral cancer' in November 2003 revealed over 32,000 pages including a number of support groups. Highly public cases such as those of the writer John Diamond,19 who wrote a book on his experience, would also be expected to raise public awareness of oral cancer.

Whilst these different forms of publicity and education should be expected to raise awareness of oral cancer amongst the public and health professionals, it is important to try and assess how effective they are being. None of the previous studies on oral cancer referral delays have been easily comparable because they were set in different areas and used different methods. We therefore looked at delays in oral cancer referrals in two regions of England, Gloucestershire and Oxfordshire, that had previously been studied by Schnetler6 before the incept of NOCAW. We tried to copy the methods he described to allow some comparison of our results with those he reported. By repeating the audit in the same area we hoped to minimise confounding factors such as educational and socio-economic status of the population, access to health care and changes in hospital referral advice.

Comparison between referral delays in the United Kingdom and internationally

Table 1 gives on overview of published data on referral delays in oral cancer in the United Kingdom and Table 2 gives some published data from other countries. One immediate problem is the difficulty comparing results in different articles because of differences in definitions and presentations of the data. What is clear is that a substantial number of patients experience professional delays in the diagnosis of their oral cancer.

Table 1 Referral delays in oral cancer a United Kingdom perspective
Full size table
Table 2 Referral delays in oral cancer, an international perspective
Full size table

Referral delays: 1995-2000

Patients attending the Oral and Maxillofacial Surgery departments in the regions of Gloucestershire and Oxfordshire, with a first diagnosis of an Oral Cancer, between November 1995 and October 2000 were identified from pathology records and cancer databases. The referral letters and the record of the first clinical appointment was examined.

For the purposes of this audit tumours were identified as oral cancer when they originated from the floor of the mouth, tongue, lip, retromolar, lower or upper alveolus, cheek, or palate.

The following information was recorded for each patient: name, sex, age at presentation, referring practitioner (general dental practitioner, general medical practitioner or hospital doctor (HD)), diagnosis of the referring practitioner, referral delay, duration of lesion (patient delay), site and size of lesion and definitive diagnosis. Patients were excluded from the analysis if all the information was not available from the initial clinical records.

One-hundred-and-ninety-five cases of oral cancer were identified during the defined period. Of these there were 120 for which the notes could be obtained and contained the required information. Thirty-two referrals (27%) came from GDPs, 78 (65%) came from GMPs and 10 (8%) came from HDs.

One immediate problem in studies such as this is the continued poor quality of referral letters and record keeping.20 In our audit only 62% of the cases of oral cancer identified had all the required information available.

The median patient delay identified was of 13 weeks (range 1-104 weeks) (Fig. 2). This compares poorly with the 10 week patient delay identified by Schnetler.6 It is obviously disappointing that our audit has shown an increase in patient delay and raises questions as to the effectiveness of current strategies on health promotion.

Figure 2
figure 2

Median delay from patients and each referring group

Full size image

The median referral delay in our audit was one week (range 0-38 weeks) (Fig. 2). Sixty-one per cent of patients were referred within the recommended two weeks of first reporting the lesion to a healthcare professional. Hospital doctors were significantly more likely to refer patients within two weeks than other healthcare professionals (P<0.01) but this may simply represent their easier access to local specialist services. One note of caution that must be made when considering patient and professional delay is that some patients may have attended another healthcare professional prior to their diagnosis being made and the lesion not observed or no referral made and this would not be identified in a retrospective analysis such as this. There is anecdotal evidence of patients attending dentists for a check-up and having T4 tumours diagnosed within a few months by a specialist, suggesting a missed diagnosis. Neither could our audit identify any patients who sought advice from one healthcare professional about oral lesions, who were advised to seek advice from another healthcare professional, who subsequently referred them.

The 39% of patients in whom there was a referral delay appears to compare favourably with the 45% identified by Schnetler6 but as he used a shorter definition for referral delay it is not possible to compare the results directly. If we used a definition for referral delay of less than two days we find that 61% of patients had delayed referral, slightly more than what was found previously. Again interpretation of the results is difficult because as new guidelines are publicised, clinicians may change their practice. They may for example become more confident in reviewing a lesion before referring. The widely publicised guidelines by the DOH14 should have clarified the guidelines further for referral for healthcare professionals and it can only be hoped that they will have further impact on referrals after this date.

Diagnosing and referring oral cancer

In the year 2000 the DOH published set guidelines14 that all patients referred with suspected cancer should be seen by their local specialist unit within a period of two weeks, and contained within this are guidelines regarding the suggested duration of a lesion after which a referral should be made. The guidelines do not specify how soon after a practitioner sees a lesion they should refer it but as the average patient delay in our audit was 19 weeks then it is likely that the majority of patients should be referred within two weeks of first being seen by a healthcare professional. We did not collect data on the delay in patients being seen after their referral letter was received as it is normal practice in both Gloucestershire and Oxfordshire to offer patients, who are referred with a letter suggesting a malignancy, an appointment within one week of the referral being received. This is one potential source of delay as such urgent appointments would not normally be offered if the letter fails to mention information that makes the diagnosis of oral cancer likely.

We found that 75% of referral letters correctly suggested that the lesion was malignant or suspicious and this is slightly better than previous reports. No referring group was significantly more likely to make a correct diagnosis. Those making the correct diagnosis were however significantly less likely to delay referral (P<0.01).

The important role of GDPs in the prevention and detection of oral cancer is undeniable. Any dental appointment offers some opportunity for the practitioner to offer basic preventative advice including the avoidance of tobacco and heavy alcohol consumption.16 Disturbingly Warnakulasuriya and Johnson21 found that only 50% of dentists enquired about these habits and only 30% of those offered health education advice regarding these. They also found that only 84% of dentists claimed to perform screening of the oral mucosa routinely. Examination of the oral mucosa for benign and malignant lesions as well as oral manifestations of systemic disease should form a part of the dentist's routine clinical examination. Although early oral carcinoma may be quite subtle in appearance and mimic a number of other conditions a high index of suspicion is required and it is advisable that a low threshold for referral to a specialist centre be held.

The number of patients attending their GMP with oral cancer indicates also the importance of education amongst this group, and supports the need for Oral and Maxillofacial surgery to be included in the undergraduate medical curriculum.

Lydiatt22 reported on a number of cases of diagnostic delay in oral cancer and found that failure to refer was a common allegation and with delays of over three months it was more difficult to defend against an accusation of negligence.

Future directions

The time taken by patients with oral cancer to seek advice from health professionals remains the longest delay in them obtaining specialist advice and treatment. Our findings suggest that the publicity generated by the National Oral Cancer Awareness Week in 1995 and subsequent years, and other publications and events since then, have not had the desired effect in terms of improving patient awareness of the importance of seeking advice on oral lesions, as the mean patient delay has increased.

This raises questions as to whether the current style of campaigns chosen is best suited to the stated aims. The number of people smoking despite many years of consistent advice that there are health risks associated with the habit of smoking suggests that the general public are always willing to listen to health advice, particularly where it is associated with lifestyle choices that are not seen to be pleasurable. In this case the advice to attend the dentist is seen to be unpleasant by many people.

The delay in patients being referred after seeking initial advice is however still a long way short of what could be considered acceptable and again suggests that further efforts need to be made in the continuing education of healthcare professionals as to the presenting signs and symptoms of oral cancer and ensuring that current guidelines on referral are adhered to.

General dental practitioners play a crucial role in screening patients for oral cancer and education in this area should form a regular part of continuing professional development. A substantial number of patients also present via their family doctors and education on oral cancer should also be available for this group at both undergraduate and postgraduate level.

The treatment of oral cancer is becoming more centralised and Oral and Maxillofacial surgeons and other specialists working in district general hospitals, who are the first point of referral for most patients with oral lesions and will generally make the initial diagnosis, need to ensure that not only are patients given first appointments as early as possible but also that biopsy and arrangements for following up results is arranged to minimise delays. Referral pathways for definitive care need to be clearly defined, again to reduce any further delays before definitive treatment can be started. Table 3

Table 3 Number (%) of delayed referrals in each health care group. Comparison between Schnetler6 and our audit using two days and two weeks as referral delay
Full size table