Alcol e tumori con focus sulle basse dosi Carlo La Vecchia Department of Epidemiology, Istituto Mario Negri Department of Clinical Sciences and Community Health, Università degli Studi di Milano 1
Cancers of the mouth, pharynx, oesophagus, and larynx i) Consumption of alcoholic beverages increases the risk of cancers of the mouth pharynx, oesophagus, and larynx (apart from cancers, of the salivary glands and nasopharynx) ii) The risk increases approximately in proportion to the amount drunk iii) In each case, alcoholic beverages act synergistically with smoking, each agent approximately multiplying the effects of the other, and iv) The main component of alcoholic beverages that determines the risk of cancer is ethanol. 2
Alcohol and oral cancer and pharyngeal cancer 3
Alcohol and oral and pharyngeal cancer: meta-analysis 4
Alcohol and SCC esophageal cancer 5
Alcohol and SCC esophageal cancer: meta-analysis 6
Alcohol and esophageal adenocarcinoma Alcohol drinking is unrelated to esophageal adenocarcinoma 7
Alcohol and esophageal adenocarcinoma (Bagnardi et al., in press) 8
Alcohol and laryngeal cancer 9
Alcohol and laryngeal cancer: meta-analysis 10
Oral cancer Alcohol tobacco interaction Table 2. Odds ratio (OR) and rations of Ors and 95% confidence intervals (CI)* of oral cancer and pharyngeal cancer according to alcohol drinking and smoking habit. Italy and Switzerland, 1992-97 Alcohol drinking (drinks/wk) Smoking habit (cig./day) 0-20 OR (95% CI) 21-48 OR (95% CI) 49-76 OR (95% CI) 77 OR (95% CI) Alcohol-adjusted OR (95% CI) Oral cavity Never 1 3.1 (0.7-13.7) 6.8 (1.3-36.7) 1 1-14 2.3 (0.4-14.1) 8.3 (1.9-35.5) 53.9 (12.9-225.2) 92.2 (18.3-464.9) 3.8 (1.7-8.2) 15-24 3.4 (0.7-15.6) 25.3 (7.3-88.5) 81.5 (22.5-294.8) 163.0 (42.8-621.5) 7.6 (3.8-15.2) 25 5.8 (1.2-27.0) 24.8 (6.4-95.6) 119.5 (36.3-471.1) 274.5 (65.5-1149.8) 10.3 (4.9-21.8) Smokingadjusted 1 4.8 (2.5-9.4) 18.8 (9.3-37.9) 39.4 (18.4-84.2) Franceschi et al., 1999 11
Odds ratios Alcohol tobacco interaction H&N cancer (INHANCE) 14.2 (8.3, 24.4) 15 9.9 (6.4, 15.5) 4.8 (3.2, 7.2) 4.2 (2.4, 7.1) 10 3.1 (2.1, 4.5) 2.2 (1.6, 3.1) >20 cigarettes/day 1-20 cigarettes/day 5 1.9 (1.3, 2.9) 1.0 (0.8, 1.3) 1.0 Never 0 3 drinks/day 1-2 drinks/day Never Alcohol drinking Hashibe et al., 2009 12
Effect in the absence of smoking In Europe and North America, all cancers of the upper respiratory and digestive tracts are rare in the absence of smoking and only very few studies have included enough cases to provide useful information about the effect of alcohol by itself. 13
Effect in the absence of smoking H&N cancer (INHANCE) Hashibe et al., 2007 14
Effect in the absence of smoking There is no reason to suppose that tobacco smoke is the only carcinogenic agent to which the human upper respiratory and digestive tracts are exposed, and ethanol may be facilitating the effect of some other unrecognized carcinogenic agents in nonsmokers. Acetaldehyde is a carcinogenic metabolite of alcohol. Doll et al., 1999; IARC 2007; 2009 15
Interaction with nutrition There is now reason to believe that the risk from exposure to many carcinogenic agents can be reduced by the regular consumption of fruit and vegetables. Heavy drinking is commonly associated with poor nutrition and this increases the risk (particularly of pharyngeal and oesophageal, but also laryngeal cancers) in heavy drinkers. Doll et al., 1999; IARC 2003 16
A. Moderate versus Non-/Occasional Drinkers, Men and Women Author, year Sex Site Ca Exp Co Exp Ca NExp Co NExp RR (95% CI) Cancer of the large bowel. Moderate vs non/occasional drinkers Fedirko et al., 2011 Case-control studies Potter, 1986 W CR -- -- -- -- Potter, 1986 M CR -- -- -- -- Kune, 1987 W CR 94 86 189 195 Kune, 1987 M CR 186 183 103 121 Peters, 1989 M CR 39 38 61 63 Longnecker, 1990 M CR 179 224 168 281 Choi, 1991 M CR 36 103 30 105 Riboli, 1991 W CR 78 101 50 107 Riboli, 1991 M CR 42 79 8 16 Barra, 1992 MW CR 435 585 354 576 Peters, 1992 MW C 164 158 224 236 Gerhardsson de Verdier, 1993 MW CR 121 114 448 398 Hoshiyama, 1993 MW CR 58 236 82 232 Newcomb, 1993 W CR 124 369 169 395 Boutron, 1995 M CR 62 82 16 31 Chyou, 1996 M CR 236 3153 152 2804 Murata, 1996 M CR 19 32 34 98 Slattery, 1997 W C -- -- -- -- Slattery, 1997 M C -- -- -- -- Yamada, 1997 MW CR 43 85 11 23 Tavani, 1998 MW CR 893 1698 395 1059 Murata, 1999 W CR 9 20 112 238 Murata, 1999 M CR 87 147 50 77 Ji, 2002 W CR 18 17 838 659 Ji, 2002 M CR 72 88 503 462 Sharpe, 2002 M CR 178 80 106 59 Kim, 2004 MW CR 104 72 139 153 Murtaugh, 2004 M R 150 168 272 331 Hu, 2007 W CR 100 172 391 706 Hu, 2007 M CR 353 506 267 502 Stern, 2007 MW CR 25 47 234 965 Gao, 2008 M CR 103 78 73 124 Lightfoot, 2008 MW CR 265 289 140 294 Benedetti, 2009 M CR 330 226 125 120 Kim, 2009 MW CR 279 185 317 324 Morita, 2009 MW CR 413 467 272 311 Wernli, 2009 W CR 128 128 633 552 Yamamoto, 2010 MW CR 9 21 5 27 All case-control studies (I-squared = 65.5%, p = 0.0001) 0.0001) Cohort studies Wu, 1987 W CR -- -- -- -- Wu, 1987 M CR -- -- -- -- Klatsky, 1988 MW CR 96 2921 36 944 Stemmermann, 1990 M CR -- -- -- -- Goldbohom, 1994 MW CR 101 2508 82 2449 Flood, 2002 W CR 36 3344 301 26776 Otani, 2003 M CR 72 71933 65 74123 Pedersen, 2003 MW CR 231 10877 124 5712 Shimizu, 2003 M CR 154 93793 13 8101 Sanjoaquin, 2004 MW CR 26 3506 30 3141 Su & Arab, 2004 MW C 26 1382 63 3811 Wei, 2004 W CR -- -- -- -- Chen, 2005 MW CR 57 156713 134 359616 Wakai, 2005 W CR 7 6253 199 193562 Wakai, 2005 M CR 98 41446 54 33018 Akhter, 2007 M CR 54 39854 36 34553 Ferrari, 2007 MW CR 508 698399 224 409104 Tsong, 2007 MW CR 70 25177 658 443968 Bongaerts, 2008 MW CR 677 12367 487 11447 Kabat, 2008 W CR -- -- -- -- Lim & Park, 2008 MW CR 8 3645 74 48414 Thygesen, 2008 M CR 330 229757 67 71855 Toriola, 2008 M CR 17 8082 5 8802 Allen, 2009 W CR 1493 2190000 1543 2180000 All cohort studies (I-squared = 49.2%, p = 0.004) 0.004) All studies (I-squared = 60.1%, p = 0.000).25.5 1 1.5 2 3 4 5 6 1.76 (0.96, 3.22) 0.76 (0.53, 1.07) 1.13 (0.79, 1.61) 1.20 (0.86, 1.67) 1.00 (0.53, 1.90) 1.39 (1.05, 1.85) 1.74 (1.02, 2.97) 1.65 (0.84, 3.23) 1.06 (0.26, 4.27) 1.39 (1.10, 1.75) 1.05 (0.76, 1.45) 0.91 (0.68, 1.23) 0.44 (0.26, 0.73) 1.20 (0.89, 1.62) 1.40 (0.71, 2.76) 1.41 (1.14, 1.73) 1.72 (0.86, 3.42) 1.00 (0.78, 1.27) 0.87 (0.75, 1.00) 1.00 (0.42, 2.37) 1.20 (1.03, 1.39) 0.96 (0.42, 2.16) 0.85 (0.55, 1.32) 0.90 (0.44, 1.81) 0.86 (0.61, 1.20) 1.39 (0.97, 2.00) 1.30 (0.82, 2.07) 1.04 (0.78, 1.38) 0.93 (0.70, 1.25) 1.29 (1.03, 1.62) 2.19 (1.32, 3.64) 2.25 (1.46, 3.46) 2.02 (1.51, 2.69) 1.31 (0.96, 1.80) 1.54 (1.21, 1.97) 1.01 (0.82, 1.25) 0.87 (0.67, 1.13) 2.31 (0.67, 7.94) 1.18 (1.07, 1.29) 1.45 (0.80, 2.61) 2.42 (1.30, 4.50) 2.03 (1.27, 3.25) 1.39 (1.10, 1.76) 0.97 (0.64, 1.47) 1.00 (0.70, 1.42) 1.30 (0.89, 1.89) 1.03 (0.83, 1.28) 1.40 (0.76, 2.56) 1.53 (0.87, 2.69) 1.69 (1.03, 2.78) 1.08 (0.92, 1.27) 1.11 (0.74, 1.67) 1.32 (0.61, 2.86) 1.55 (1.11, 2.17) 1.34 (0.88, 2.05) 1.13 (0.95, 1.34) 1.84 (1.31, 2.58) 1.10 (0.93, 1.29) 1.06 (0.88, 1.27) 1.13 (0.52, 2.45) 1.40 (1.08, 1.83) 3.50 (1.22, 10.00) 1.07 (1.01, 1.13) 1.24 (1.14, 1.34) 1.21 (1.13, 1.28) 17
Ca Co Ca Co Author, year Sex Site Exp Exp NExp NExp RR (95% CI) Case-control studies Peters, 1989 M CR 45 45 61 63 0.89 (0.52, 1.51) Longnecker, 1990 M C R 51 57 168 281 1.63 (1.05, 2.53) Choi, 1991 M C R 64 212 30 105 3.53 (2.58, 4.82) Riboli, 1991 M C R 95 156 8 16 1.22 (0.32, 4.59) Barra, 1992 MW C R 469 979 354 576 1.01 (0.83, 1.23) Peters, 1992 MW C 116 76 224 236 1.67 (1.13, 2.47) Boutron, 1995 M CR 31 45 16 31 1.30 (0.59, 2.86) Cancer of the large bowel. Heavy vs non/occasional drinkers Murata, 1996 M C R 9 12 34 Tavani, 1998 MW CR 269 567 395 Murata, 1999 M CR 97 71 50 Ji, 2002 M C R 209 176 503 Sharpe, 2002 M C R 111 32 106 All case-control studies (I-squared = 83.4%, p = 0.0001) Cohort studies Otani, 2003 M CR 90 71194 65 Pedersen, 2003 MW C R 69 2821 124 98 1059 77 462 59 74123 5712 2.16 (0.84, 5.59) 0.95 (0.76, 1.18) 1.96 (1.20, 3.20) 1.17 (0.91, 1.50) 2.05 (1.28, 3.30) 1.49 (1.13, 1.96) 1.70 (1.20, 2.40) 1.18 (0.87, 1.60) Wakai, 2005 M C R 97 47460 54 33018 1.56 (1.11, 2.19) Akhter, 2007 M CR 138 75969 36 34553 1.91 (1.32, 2.77) Ferrari, 2007 MW CR 101 81939 224 409104 1.66 (1.27, 2.16) Lim & Park, 2008 MW CR 10 4291 74 48414 1.11 (0.40, 3.06) Thygesen, 2008 M CR 59 28425 67 71855 1.75 (1.21, 2.53) All cohort studies (I-squared = 0.0%, p = 0.468) 1.57 (1.38, 1.80) All studies (I-squared = 76.4%, p = 0.000) 1.52 (1.27, 1.81) Fedirko et al., 2011.25.5 1 1.5 2 3 4 5 6 18
Cancer of the large bowel. Dose-risk relation. Fedirko et al., in press 19
Conclusions. Cancers of the large bowel. Cohort and case-control studies are consistent in suggesting some direct relation between alcohol consumption and colorectal cancer. The relation, however, is moderate, and a two-fold risk for both colon and rectum cancer can be excluded, even with high levels of alcohol consumption. Doll et al., 1999; IARC 2007, 2009 20
Liver cancer Alcohol drinking is strongly related to cirrhosis. Alcohol drinking is associated with primary liver cancer, although the relation is difficult to investigate in epidemiological studies, since most alcohol-related liver cancers follow a cirrhotic degeneration, which may lead to a reduction of alcohol drinking. 21
1959 1965 1971 1977 1983 1989 1995 2001 Standardised rate per 100,000 Population Liters of pure alcohol per adult (15+) Alcohol and mortality from cirrhosis. Italy Italy, 1959-2001 50 45 40 35 30 25 20 15 10 5 0 30 26 22 18 14 10 6 Year 22
1959 1965 1971 1977 1983 1989 1995 2001 Standardised rate per 100,000 Population Liters of pure alcohol per adult (15+) Alcohol and mortality from cirrhosis. France France, 1959-2001 60 50 40 30 20 10 0 32 30 28 26 24 22 20 18 16 14 12 10 Year 23
1959 1965 1971 1977 1983 1989 1995 2001 Standardised rate per 100,000 Population Liters of pure alcohol per adult (15+) Alcohol and mortality from cirrhosis. Hungary Hungary, 1959-2002 180 160 140 120 100 80 60 40 20 0 20 18 16 14 12 10 8 Year 24
Liver cancer Table II. Relation of hepatocellular carcinoma with smoking and drinking habits: Milan, Italy, 1984-1987. Hepatocellulcar carcinoma Controls Relative risk estimates (95% CI) M-H MLR Total alcoholic beverage consumption (drinks per day) <4 89 652 1 1 4-6 23 216 0.76 (0.46-1.26) 1.10 (0.67-1.79) >6 39 183 1.51 (0.96-2.36) 1.43 (0.83-2.46) 2 1 2.05 (p=0.15) 1.83 (p= 0.18) La Vecchia et al., 1988 25
Liver cancer (Bagnardi et al., in press) 26
Liver cancer Alcohol Cohort Case-Control Cancer site intake Pooled RR (95% CI) Pooled RR (95% CI) P a Liver Light 0.85 (0.74,0.97) 1.31 (0.97,1.78) 0.226 Moderate 1.00 (0.87,1.17) 1.15 (0.97,1.35) Heavy 1.12 (1.02,1.23) 2.79 (2.00,3.87) (Bagnardi et al., in press) 27
Liver cancer Apparently moderate assocation, possibility underestimated. Turati et al, subm 28
Alcohol and pancreatic cancer Men and Women Tramacere et al., 2010 29
Alcohol and pancreatic cancer Tramacere et al., 2010 30
Alcohol and pancreatic cancer There is convincing evidence for the absence of a role of moderate drinking in pancreatic carcinogenesis, coupled to an increased risk for heavy alcohol drinking. Given the moderate increase in risk and the low prevalence of heavy drinkers in most populations, alcohol appears to be responsible only for a small fraction of all pancreatic cancers. Tramacere et al., 2010 31
Breast cancer Collaborative re-analysis Collaborative re-analysis, 2002 32
Alcohol and breast cancer Meta-analysis (Bagnardi et al., in press) 33
Alcohol, kidney cancer and lymphomas No excess risk, and possible inverse relationships 34
Alcohol and kidney cancer (Bagnardi et al., in press) 35
Alcohol and lymphomas (Bagnardi et al., in press) 36
Alcohol and cancer Low doses We evaluated the association between light drinking (<1 drink day) and cancer of the colorectum, breast, larynx, liver, esophagus, oral cavity and pharynx, through a metaanalytic approach. We searched epidemiological studies using PubMed, ISI Web of Science and EMBASE, published before December 2010. Bagnardi et al., 2012 37
Alcohol and cancer Low doses Bagnardi et al., 2012 38
Alcohol and cancer Low doses We included 222 articles comprising 92 000 light drinkers (<1 drink day) and 60 000 non-drinkers with the cancer sites of interest. Bagnardi et al., 2012 39
Alcohol and cancer Low doses <1 drink/day Bagnardi et al., 2012 40
Alcohol and cancer Low doses Alcohol 1 drink day and cancer Low doses Bagnardi et al., 2012 41
Alcohol and cancer Low doses Light drinking was associated with the risk of oral and pharyngeal cancer [relative risk, RR = 1.17; 95% confidence interval (CI) 1.06 1.29], esophageal squamous cell carcinoma (SCC) (RR = 1.30; 95% CI 1.09 1.56) and female breast cancer (RR= 1.05; 95% CI 1.02 1.08). Bagnardi et al., 2012 42
Low doses and ALDH2 in Asian populations Asian individuals with the ALDH2 Lys allele experience a marked elevation in blood acetaldehyde after alcohol ingestion. In a Japanese case-control study of oral cancer, the RR for intermediate alcohol drinking /low folate was 2.4 for ALDH2 Lys+ vs 1.2 for those with ALDH2 Glu/Glu, a compared with those with low alcohol/high folate (Matsuo et al, 2012). 43
Total mortality and ischemic heart disease The consumption of alcohol appeared to reduce the risk of ischaemic heart disease, largely irrespective of amount. Among regular drinkers, mortality from all causes combined increased progressively with amount drunk above 21 units a week. Alcohol-related causes, including cancers, increased above 14 units per week. Doll et al., 1994 44
Doll et al., 1994 45
Alcohol in Mediterranean countries Alcohol consumption was traditionally high in Mediterranean countries, particularly in France and Italy, but substantial declines have been observed over the last three decades. The pattern and problem of alcohol drinking is also peculiar in these countries, with regular (rather than binge) drinking at meals, and with wine being the most common type of alcoholic beverage in most regions. 46
Alcohol consumption in Italy Castro et al., 2013 47
Alcohol in Mediterranean countries Alcohol drinking has major implications for the risk of cirrhosis and a few cancer sites, including those of the upper digestive and respiratory tract, whose incidence and mortality were exceedingly high in the 1970 s and 1980 s in France, northern Italy and Spain. Now, the highest rates are in central and eastern Europe. 48
Conclusions 1 In terms of risk assessment, high levels of alcohol consumption (i.e., more than four drinks per day) result in a substantial risk of cancer at several sites. Lower levels of consumption result in a moderately increased risk for selected cancers. Moderate alcohol consumption can have protective effects against certain types of heart disease. 49
Conclusions 2 A substantial number of cancer cases and cancer deaths are attributable to alcohol drinking. Over 85% of these are avoidable by moderating consumption. 50
Conclusions 3 Taking into account favourable and unfavourable effects of alcohol on health, a sensible advice should be given as for recommended limits to alcohol drinking. These limits should not exceed 30 g of ethanol per day (i.e. about to two drinks of beer, wine or spirits a day, meals included) for men and 15 g (one drink) for women. 51
52