Oral cancer - risk factors

Cigarette, cigar and pipe smoking are all are causes of oral cancer. 3, 79 In the earlier parts of the last century, pipe smoking was associated with lip cancer (most lip cancers arise on the lower lip) and its decline in popularity may be linked with some of the decrease in lip cancer. A recent case control study in Spain showed an increased risk of lip cancer when smokers were in the habit of leaving the cigarette on the lip. 4 Since the 1920s smoking cigarettes has been the main form of tobacco use in the UK. According to a meta-analysis, on average current smokers have a three-fold increased risk of oral cancer. 62 The risk of oral cancer associated with smoking is both dose and duration dependent while smoking cessation leads to a fall in risk. 1, 5, 6 However, a recent study showed that it takes 20 years or longer for the risk to reduce to that of never smokers. 73

A study published in December 2011 estimated that around 70% of oral and pharyngeal cancers in men and around 55% in women in the UK in 2010 were caused by smoking tobacco.81

In 2009, 2% of British men smoked at least 1 cigar a month, while less than half of 1% of men said they smoked a pipe. 8 Very few women in the UK smoke pipes or cigars. In one case control study of oral cancer carried out in Cuba, the odds ratio associated with smoking 30 or more cigarettes a day was comparable with that for smoking 4 or more cigars a day. Indian women who practice reverse chutta smoking, with the lighted end of the cigar inside the mouth, have particularly high rates of oral cancer of the palatal mucosa. 9 Smoking bidi(s) which are made of hand-rolled tobacco wrapped in tendu leaf also increases the risk of oral cancer. 10

There is evidence that exposure to secondhand smoke (SHS) may increase oral cancer, with a 63% risk increase shown for never smokers exposed to SHS at home or at work. Among those exposed at home and at work for more than 15 years, there was an 84% risk increase.79

A recent evaluation by the International Agency for Research on Cancer (IARC) has confirmed that smokeless tobacco is also carcinogenic. 11, 12 Risk varies according to the composition of smokeless tobacco used in different countries. A recent meta-analysis showed a more than doubling in risk of oral cancer with use of smokeless tobacco in the United States and Canada, a five-fold risk increase for India and other Asian countries and a seven-fold risk increase in Sudan. No risk increase for oral cancer was shown with smokeless tobacco use in the Nordic countries. 63 In the UK and Europe (with the notable exception of Sweden) the use of smokeless tobacco is rare except in minority ethnic groups (see next paragraph). 13 In the USA it is a major problem with a reported 6% of the adult male population as regular users 14. In some areas, particularly the southern states, the prevalence is much higher with up to a third of young men using smokeless tobacco. 15

The primary cause of the very high incidence of oral cancer in South Asia is the widespread habit of chewing betel quid (or paan) and related areca nut use (areca nut is the seed of the fruit of the oriental palm, Areca catechu). 16 Chewing betel is thought to date back at least 2000 years and worldwide an estimated 200-400 million people practice the habit. 17 The components of the betel quid vary between different populations but the main ingredients are the leaf of the vine, Piper betel, areca nut, slaked lime (calcium hydroxide) and spices. 18 Tobacco was introduced to South Asia in the seventeenth century. Areca nut is carcinogenic to humans and the risk of oral cancer is increased with chewing paan without tobacco, although the risk is higher for paan containing tobacco. 19-21,64 As with smoking tobacco, risk is dependent on dose and duration of use. 22 Among Asian communities in the UK, Bangladeshis are the most likely to retain the habit of betel quid chewing as Figure 4.1 shows with 9% of men and 16% of women using smokeless tobacco. 23 The most commonly used chewing tobacco product is betel quid with tobacco. 24

Download this chart (32.5KB)

Alcohol is a major risk factor for oral cancer and a study published in December 2011 estimated that, in 2010, around 37% of oral and pharyngeal cancers in men and 17% in women in the UK were linked to alcohol.82 A meta-analysis reported risk ratios for alcohol intake of 25 grams/day, 50 grams/day and 100 grams/day after adjustment for smoking of 1.8, 2.9 and 6.1, respectively. 65 One study showed a doubling in risk for people drinking 14 grams of alcohol/day. 66 People who both drink and smoke have a much higher risk of oral cancer than those using only alcohol or tobacco as Figure 4.2 demonstrates using US data. 28

Download this chart (27KB)

Heavy drinkers and smokers have 38 times the risk of abstainers from both products. 67 The effect of alcohol consumption in non-smokers differs by subsite, with a recent pooled analysis showing no increase in risk of oral cavity cancer in this group, although there was an increase in risk of cancers of the oropharynx/hypopharynx, two other subtypes of oral cancer. 75 Results from the Million Women Study in the UK also supported this finding. 76 Risk of alcohol consumption also interacts with risk of smokeless tobacco, with a combined risk of regular alcohol consumption and tobacco chewing of 24 shown in one study. 74

It has been suggested that it is the total amount of ethanol ingested rather than the type of product (beer, wine, spirits) which is important. 68 . The rising trends in oral cancer mortality in Europe have been related to increasing levels of alcohol consumption. For example, in Denmark the alarming increase in oral cancers has been attributed predominantly to greater alcohol consumption. 30 An exception to this rise in alcohol consumption is seen in France, where a decrease in alcohol consumption has been linked to the fall in oral cancer mortality rates in the 1980s. 31

In the UK, consumption of alcohol has more than doubled since the middle of the last century, from 3.9 to 8.6 litres of pure alcohol per head per year- see Figure 4.3. 32, 33

Download this chart (30.5KB)

In Britain there has been a decline since 2000 in the proportion of adults exceeding the recommended weekly guideline of 21 units for men and 14 units for women and in 2009, 26% of men and 18% of women exceeded these levels. 8 On average, men in Britain drank 16 units a week and women 8 units a week in 2009. The age group with the highest alcohol consumption was 45-64 for men (average of 19 units a week) and 16-24 for women (average of 10 units a week).8 There have also been falls in heavy drinking since 2000, especially in those aged 16-24. 8

At present the UK level of drinking (8.6 litres per year) is lower than in most European countries, for example, France (10.7 litres), Portugal (11.0 litres), Spain (9.9 litres) and Germany (10.6 litres). However, whereas consumption is either falling or stabilising in most of these countries, in the UK it is rising quickly. It is estimated that if current trends continue, the UK could rise to near the top of the consumption table within the next ten years. 33 Co-ordinated and funded action is needed to tackle the UK’s complex drinking problems. 35

Concern has also been expressed about the use of mouthwashes, particularly those with high alcoholic content. 36 However, the majority of studies show no increase in risk with use of alcoholic mouthwash. 59-61,69-71

A meta-analysis showed a significant risk reduction of about 50% for each additional daily serving of fruit or vegetables. 37 A large prospective study, published since this meta-analysis, showed a smaller significant risk reduction for oral cavity cancer of 26% for each additional serving of vegetables, but no association for fruit intake. 38 Results may vary by smoking status, with a large case-control study showing risk reductions with the highest intake of fruit and vegetables among smokers and alcohol consumers but not among people who had never smoked or drunk alcohol. 39 A study published in December 2011 estimated that, in 2010, around 57% of oral and pharyngeal cancers in men and 54% in women in the UK were linked to people eating fewer than five portions a day (400g/day) of fruit and vegetables. 83 However, there is considerably more uncertainty about the links between diet and oral and pharyngeal cancer than for other risk factors, such as smoking and alcohol.80,83

Risk of oral cancer appears to fall with increasing body mass index (BMI). A recent case-control study in Spain reported a significant reduction in risk of oral cancer with higher BMI at diagnosis and two years prior to diagnosis, after adjustments for smoking, drinking, fruit and vegetable intake, although the association was not significant among people who had never smoked. 43 Since then, a case-control study showed a lower risk with higher BMI two years prior to the study among never and ever smokers. 72

Solar irradiation is a risk factor for cancer of the lip. 77 Lip cancer is three times more common in men than women which may be an effect of occupation, smoking and sun-exposure. 4 The use of sunscreens and protective clothing would significantly reduce exposure.

There is evidence that infection with high-risk human papillomaviruses (HPV) increases risk of oral cancer, particularly HPV-16. 44-47 The association is strongest for cancers of the oropharynx. 71 It has been estimated that, in the UK, 8% of cancers of the oral cavity and 14% of cancers of the oropharynx are linked to HPV infection.84

Studies show an increased risk of oral cancer in women with previous HPV-associated anogenital cancer, providing more evidence of a link with HPV infection. 48-49 In addition, a history of more sexual partners or oral sex partners has been associated with an increased risk of oropharyngeal cancer, and a younger age at sexual debut or history of oral sex with an increased risk of cancer of the tonsil and base of the tongue, again pointing to a role of sexually-transmitted HPV.78 An increased risk of oral cancer has been shown in individuals with HIV/AIDS or people who have undergone organ transplants, supporting a role of immunosuppression. 50

Several oral lesions and conditions precede oral carcinoma and the most common of these are leukoplakia and erythroplakia. Other rare precancerous conditions include lichen planus, oral submucous fibrosis, syphilitic glossitis and sideropenic dysphagia. Leukoplakia has many clinical variants but is much less likely to progress to malignancy than erythroplakia. The precise prevalence of these conditions in the UK is unknown. Estimates of leukoplakia prevalence outside the UK range from 0.2 to 11.7% of the population and the prevalence of erythroplakia is considerably less. 51 It has recently been estimated that the annual transformation rate of oral leukoplakia to oral squamous cell carcinoma may not exceed 1%. 52 Erythroplakia is rare and mainly occurs in people aged over 60. 53

The rising incidence and mortality rates in young and middle-aged adults is incontrovertible, but there has been debate over the causes of this increase and whether their disease is inherently more aggressive than that occurring in older patients. 48, 54-56 A series of studies in southern England looking at risk factors for patients under 45 years concluded that most young patients are exposed to the traditional risk factors of tobacco smoking and alcohol while consumption of fresh fruit and vegetables is protective. 48, 57, 58 However, the relatively short duration of exposure to these known risk factors suggests that other factors may also be involved and there was a small sub-group of patients who had little, if any, exposure to the major risk factors.

People with a previous oral and pharyngeal cancer have a more than 30-fold increased risk of second oral and pharyngeal cancer, and risk remains 20-fold higher 10 or more years after the first diagnosis. 25 An almost seven-fold increase in risk of oral and pharyngeal cancer has been shown after a diagnosis of squamous cell carcinoma of the oesophagus, with risk remaining higher five or more years after the first diagnosis. 26