The main causes of upper digestive tract cancers are smoking and alcohol drinking. It has been estimated, for the United States, that up to 80% of these cancers can be avoided by abstaining from alcohol drinking and smoking.1, 2, 3 Both alcohol and tobacco are independent risk factors for upper digestive tract cancers. Additionally, several epidemiologic studies have confirmed that alcohol and tobacco interact in a multiplicative way on cancer risk.4, 5, 6, 7, 8
The combined tumor-promoting effect of alcohol and tobacco smoking is poorly understood. Ethanol per se is not carcinogenic.9 However, acetaldehyde, the first metabolite of ethanol, is carcinogenic in animals.10, 11 There is convincing evidence for acetaldehyde being the ultimate carcinogenic compound behind alcohol intake also in humans.12, 13 After alcohol intake, acetaldehyde is locally formed in the oral cavity by oral mucosal alcohol dehydrogenases (ADHs) and by the oral microflora, both of which are able to oxidize ethanol to acetaldehyde.14, 15, 16 Also, tobacco smoke contains high levels of acetaldehyde, which is one of the most toxic compounds in cigarette smoke condensate.17
Acetaldehyde is highly toxic and mutagenic under various experimental conditions.18, 19, 20, 21 Epidemiologic and biochemical studies on Asian heavy drinkers with aldehyde dehydrogenase-2 (ALDH2) deficiency strongly suggest that acetaldehyde is a local and topical carcinogen also in humans.13 This deficiency results in the accumulation of acetaldehyde in saliva22 and in markedly increased risk for upper gastrointestinal (GI) tract cancers.23, 24, 25 Also, most studies concerning Caucasian individuals who are homozygous for the fast alcohol metabolizing alcohol dehydrogenase (ADH3*1) enzyme show increased salivary acetaldehyde levels related to increased risk of alcohol-related upper digestive tract cancers.26
Our earlier studies indicate that heavy alcohol drinking and smoking have an additive effect on in vitro acetaldehyde production in the saliva.27 Our aim was to examine the combined effect of alcohol drinking and tobacco smoking on in vivo acetaldehyde levels of saliva, to further elucidate the possible role of carcinogenic acetaldehyde in the pathogenesis of alcohol- and smoking-related upper GI tract cancers.
MATERIAL AND METHODS
Thirteen healthy volunteers (10 males, 3 females) took part in the study. Six were nonsmokers, mean age 28.8 ± 1 years (range 26–35), and 7 were permanent cigarette smokers, mean age 39.5 ± 3 years (range 26–61). Information about smoking status, alcohol consumption and nutritional habits was obtained by a self-administered questionnaire. All volunteers were moderate alcohol consumers, with a weekly average consumption of 70 g or less. Average cigarette consumption among smokers was 18.5 ± 2.1 (range 15–30), and all had a smoking history of more than 10 years. All volunteers were on a normal Western diet, and none was vegetarian. Exclusion criteria were as follows: treatment with antibiotics or oral antiseptic in the past month, recent food or fluid intake, smoking or tooth brushing during the previous 30 min. All participants were told to refrain from alcohol for at least 24 hr before the study.
Our study was approved by the ethical committee of the Department of Medicine, Helsinki University Central Hospital. Informed consent to participate in the study was obtained.
Volunteers were divided into 2 groups: smokers and nonsmokers. Two repeated measurements (2 study dates were separated by at least a 3-day interval) were done with the smokers, and the nonsmokers were used as controls.
Determination of the effect of ethanol ingestion on in vivo salivary acetaldehyde in smokers and nonsmokers
To measure the salivary acetaldehyde derived solely from ethanol, both parallel groups (smokers and nonsmokers) ingested 0.8 g ethanol/kg body weight in a standardized 10% v/v solution of absolute ethanol in distilled water/orange juice (50%/50%) within 30 min after baseline saliva collection. To remove local ethanol, subjects rinsed their mouths 3 times with water; thereafter, saliva samples were taken every 20 min for 160 min.
Determination of the effect of combined ethanol ingestion and tobacco smoking on in vivo salivary acetaldehyde in smokers
To measure the salivary acetaldehyde derived from the combined and simultaneous use of ethanol and tobacco, the above-mentioned procedure was repeated in smokers, after which subjects smoked one cigarette every 20 min and saliva samples were collected every 10 min for 160 min.
Determination of the effect of smoking on in vivo salivary acetaldehyde in smokers
To measure the salivary acetaldehyde derived solely from smoking, smokers smoked one cigarette (without ethanol ingestion) within 5 min after measuring the base level of salivary acetaldehyde. Thereafter, saliva samples were collected every 5 min for 15 min.
Collection and analysis of salivary acetaldehyde
To obtain the saliva samples, volunteers were told not to swallow the secreted saliva but to store it in their mouths. After 5 min, saliva was collected in a sample tube, and 450 μl were immediately transferred to a vial that contained 50 μl of perchloric acid (PCA). Acetaldehyde and ethanol levels were analyzed by headspace gas chromatography as previously described.28 Each in vivo measurement was done in duplicate.
All values are expressed as means ± SEM. Areas under the acetaldehyde concentration–time curves (AUC0–160min) were determined using NCSS2000 statistical software (Hintzel JL, Number Cruncher Statistical Systems, Kaysville, UT).
Statistical significance of the values obtained from nonsmokers and smokers was analyzed by the Mann-Whitney rank sum test. p < 0.05 was regarded as significant.
In vivo salivary acetaldehyde in smokers without concomitant smoking and in nonsmokers after ethanol ingestion
The mean in vivo salivary acetaldehyde in smokers, even without smoking, was about 2 times higher than in nonsmokers after ethanol ingestion throughout the follow-up period of 160 min (Fig. 1). The area under the curve of salivary acetaldehyde in smokers was significantly higher than in nonsmokers, 114.8 ± 11.5 vs. 54.2 ± 8.7 μM × hr, respectively (p = 0.002).
In vivo salivary acetaldehyde in smokers with concomitant smoking and in nonsmokers after ethanol ingestion
During the period of cigarette smoking, in vivo salivary acetaldehyde was increased 10-fold over levels derived from ingestion of ethanol alone. Salivary acetaldehyde increased immediately when smoking was started but also declined rapidly after cessation of smoking (Fig. 2). The area under the curve of salivary acetaldehyde in smokers was 7 times higher than in nonsmokers, and the difference was highly significant, 369.5 ± 12.2 vs. 54.2 ± 8.7 μM × hr, respectively (p < 0.001). Differences between acetaldehyde concentrations were significant at all time points from 40 to 160 min (p ≤ 0.05).
In vivo salivary acetaldehyde in smokers after smoking without concomitant ethanol ingestion
During active smoking, salivary acetaldehyde increased to 261.4 ± 45.5 μM from the basal level. Salivary acetaldehyde increased immediately when smoking was started but declined rapidly after cessation of smoking (Fig. 3).
Ethanol concentrations were equal in nonsmokers and smokers during the whole follow-up period.
Heavy alcohol consumption and tobacco smoking are the strongest and best-recognized risk factors for upper GI tract cancers in developing countries and account for over 75% of oral cancers in men in the United States, France and Italy.4 According to the epidemiologic data, alcohol interacts synergistically with tobacco smoke in the development of cancers of the oral cavity, pharynx, larynx and esophagus. In a survey from northern Italy, the odds ratios (ORs) for oral cavity and pharyngeal cancers at the highest levels of alcohol and tobacco intake were increased 80-fold relative to the lowest levels of both risk factors.3 In individuals drinking more than 121 g of alcohol and smoking over 25 cigarettes per day, the relative cancer risk for hypopharynx/epilarynx is over 135. In smokers without drinking and alcohol drinkers without smoking (equivalent amounts), the corresponding risks are 4.9 and 14.7.2 Similar and confirmatory results have been obtained also for laryngeal and esophageal cancers.2, 29, 30
There is convincing evidence for acetaldehyde being the ultimate carcinogenic compound derived from alcohol drinking both in experimental animals and in humans.13, 31 Studies among Japanese alcoholics who are deficient in the ALDH2 enzyme, which is responsible for the detoxification of acetaldehyde, show that the risk of upper digestive tract cancers is markedly increased (OR = 11.1 for oropharyngolaryngeal, 12.5 for esophageal, 3.5 for stomach and 54.2 for esophageal cancer concomitant with oropharyngolaryngeal and/or stomach cancer).23 The same enhanced cancer risk has been demonstrated among ALDH2-deficient social drinkers in Japan.32 However, these ALDH2-deficient subjects have after alcohol ingestion 2–3 times higher salivary acetaldehyde concentrations than control subjects with normal ALDH2 enzyme.22 In addition, epidemiologic studies among Caucasians have found that the frequency of the ADH3*1 allele is significantly increased in heavy drinkers with upper aerodigestive tract cancer.33 Also, these ADH3*1 homozygous individuals had significantly higher acetaldehyde concentrations in their saliva after alcohol ingestion.26 Hence, a plausible explanation for the increased cancer risk is the abnormally high acetaldehyde exposure in the upper GI tract after alcohol consumption.
Acetaldehyde is formed locally to the oral cavity mainly by the ADH enzyme of oral microbes, which oxidize ethanol to acetaldehyde in the presence of oxygen.34 Early examples of this kind of reaction included the increased breath acetaldehyde after ethanol ingestion and the production of acetaldehyde when human mouth and bronchopulmonary washings were incubated with ethanol in vitro.14, 15, 35 Since then, several in vivo studies have shown local acetaldehyde production in the oral cavity after ethanol intake. It has also been shown that heavy drinkers, smokers and patients with upper GI tract cancer have increased salivary acetaldehyde production in vitro.27, 34, 36 This microbial salivary acetaldehyde production shows high interindividual variation, but there is a significant positive correlation between salivary ethanol and acetaldehyde levels.28 Certain bacterial species identified from the oral cavity have very high ADH activity and acetaldehyde-producing capacity; furthermore, alcohol ingestion results in an increase of this bacterial strain.36
Acetaldehyde is also a known constituent of tobacco smoke.37 Undoubtedly, tobacco smoke contains many toxic compounds, but acetaldehyde is one of the most toxic compounds in cigarette smoke condensate.17 It is formed in the mainstream smoke during the burning process and then inhaled more or less to the aerodigestive tract during smoking.38 Previously, it was shown that smokers have elevated breath acetaldehyde levels after acute cigarette smoking and even endogenously after several weeks of smoking cessation.39, 40
In the present study, we demonstrate that smokers without smoking have 2 times higher salivary acetaldehyde levels than nonsmoking controls in vivo during ethanol challenge. Hence, the exposure to carcinogenic acetaldehyde after ethanol ingestion is markedly higher in smokers even without smoking than in nonsmokers. These findings are consistent with results reported earlier, which showed higher levels of breath acetaldehyde in smokers than in nonsmokers following an acute dose of ethanol.41, 42 It is unlikely that this acetaldehyde originated from the tobacco smoke since the smokers did not smoke during the test and, according to our results, the inhaled acetaldehyde derived from tobacco smoke disappears rapidly from the saliva. However, it has been suggested that smokers might have changes in their oral microbial flora, especially in high acetaldehyde producers, which could contribute to local acetaldehyde levels.27, 42 For example, certain aerobic species (Streptococcus salivarius, S. viridans, Corynebacterium sp., Stomatococcus sp. and yeasts) are significantly associated with higher acetaldehyde production in the oral cavity.27 Hence, the oral microbial flora of smokers might be more prone to produce higher levels of acetaldehyde.42, 43 There is also evidence for inhibition of the ALDH enzyme by smoking,44 which therefore could lead to less efficient acetaldehyde metabolism and, consequently, to higher acetaldehyde concentrations in the aerodigestive tract of smokers. In addition to oral microflora, the ADH and/or ALDH (especially in Asians) polymorphisms may have a marked effect on individual salivary acetaldehyde levels. This warrants further study, especially comparing smokers and nonsmokers.
The second finding of our study was that mucosal exposure to salivary acetaldehyde is markedly increased during smoking. In smokers, smoking alone increases salivary acetaldehyde to over 250 μM during active smoking, which is seen every time a cigarette is smoked. In our study, this acetaldehyde increase was 7 times higher (measured as AUC) in smokers with active smoking (every 20 min) than in nonsmokers after a dose of ethanol. In smokers, salivary acetaldehyde increased significantly more during active smoking with concomitant ethanol ingestion than without concomitant ethanol. After termination of smoking, salivary acetaldehyde concentration declines in 5 min to basal level, which still is twice as high as that in nonsmoking controls during ethanol challenge. This acetaldehyde increase during smoking is presumably due to deposition of acetaldehyde-containing smoke in the aerodigestive tract, where it is gradually dissolved in the saliva, metabolized locally or washed out via normal distribution from the oral cavity to the larynx, esophagus and digestive tract. This deposition of acetaldehyde mediated by saliva might explain the increased esophageal cancer risk observed in smokers.3, 37
The multifold increase in salivary acetaldehyde concentration may have a pronounced effect on the upper GI tract cancer risk of heavy drinkers, who often are chain smokers. In smokers, smoking frequency also largely influences the total exposure time to carcinogenic acetaldehyde because in chain smokers salivary acetaldehyde levels might be markedly higher due to the tobacco smoke, which contains acetaldehyde, than in heavy drinkers who do not smoke tobacco. We have previously reported that, after alcohol intake, up to two-thirds of the acetaldehyde can be removed from the oral cavity by a buccal L-cysteine tablet, to prevent the harmful effects of carcinogenic acetaldehyde.45 Furthermore, salivary acetaldehyde concentration after ethanol ingestion can be significantly reduced by antiseptic mouthwash (chlorhexidine) treatment, which subsequently leads to decreased acetaldehyde production by oral microflora.28 Therefore, to prevent the harmful effects of tobacco smoking and alcohol drinking on the aerodigestive tract, more investigations are warranted.
In conclusion, after a dose of ethanol, the mean in vivo salivary acetaldehyde exposure in smokers was 2 times higher without smoking and 7 times higher with active smoking than in nonsmokers. In addition, smoking alone increased markedly salivary acetaldehyde during active smoking. These findings may be due to microbial changes in the oral cavity of smokers, to inhibition of ALDH of the oral mucosa by tobacco smoke and to the significant deposition of acetaldehyde from tobacco smoke in the saliva. The markedly increased exposure to carcinogenic acetaldehyde caused by simultaneous smoking and drinking may explain the synergistic and multiplicative risk effect of alcohol drinking and tobacco smoking on upper GI tract carcinogenesis. Furthermore, the increased aerodigestive tract cancer risk observed in heavy smokers might be due to the salivary acetaldehyde from tobacco smoke.
We thank all of the volunteers who participated in the study.