Effects of beta-carotene supplementation on cancer incidence by baseline characteristics in the Physicians' Health Study (United States)

Objectives: The Physicians' Health Study (PHS) was a randomized trial of beta-carotene (50 mg, alternate days) and aspirin in primary prevention of cancer and cardiovascular disease among 22,071 US male physicians. This report updates results for beta-carotene and examines effect modification by baseline characteristics. Methods: Beta-carotene's effect on cancer over nearly 13 years was examined overall and within subgroups defined by baseline characteristics using proportional-hazards models. Results: 2667 incident cancers were confirmed, with 1117 prostate, 267 colon, and 178 lung cancers. There were no significant differences with supplementation in total (relative risk (RR) = 1.0, 95% confidence interval (CI) = 0.9–1.0); prostate (RR = 1.0, 95% CI = 0.9–1.1); colon (RR = 0.9, 95% CI = 0.7–1.2); or lung (RR = 0.9, 95% CI = 0.7–1.2) cancer, and no differences over time. In subgroup analyses, total cancer was modestly reduced with supplementation among those aged 70+ years (RR = 0.8, 95% CI = 0.7–1.0), daily drinkers of alcohol (RR = 0.9, 95% CI = 0.8–1.0), and those in the highest BMI quartile (RR = 0.9, 95% CI = 0.7–1.0). Prostate cancer was reduced with supplementation among those in the highest BMI quartile (RR = 0.8, 95% CI = 0.6–1.0), and colon cancer was reduced among daily drinkers of alcohol (RR = 0.5, 95% CI = 0.3–0.8). Conclusions: The PHS found no overall effect of beta-carotene on total cancer, or the three most common site-specific cancers. The possibility of risk reduction within specific subgroups remains.     

A prospective study of lifestyle factors and the risk of pancreatic cancer in Nord-Trøndelag, Norway

Objectives: Cancer of the pancreas is highly fatal and, despite extensive scrutiny, only cigarette smoking stands out as a likely causal agent in epidemiological studies. To explore to what extent different lifestyle factors are associated with the risk of pancreatic cancer, data from a large health screening survey in a county in Norway were analyzed. Methods: Our study included 31,000 men and 32,374 women initially free from any diagnosed cancer, and during 12 years of follow-up, 166 incident cases of pancreatic cancer were diagnosed at the Cancer Registry. Results: Compared with never smokers, we found a two-fold increased risk among current smokers, and a dose–response association with number of cigarettes (p for trend = 0.02 for both men and women) and with number of pack-years (p for trend = 0.02 for men and 0.01 for women). The risk among former smokers quitting more than 5 years before study entry was close to the risk of never smokers. Compared with persons who reported never or infrequently to be physically worn out after a day's work, the relative risk (RR) among those who nearly always became worn out was 2.9 (95% confidence interval (CI) = 1.4–5.8) for men and 3.8 (95% CI = 1.6–9.2) for women. Divorced or separated men had a risk of 3.1 (95% CI = 1.3–7.2) compared with married men. We observed a higher risk among women in occupations of high socioeconomic status (RR = 2.5; 95% CI = 1.2–5.2), and among men occupied in farming, agriculture or forestry (RR = 2.1; 95% CI = 1.1–4.0), compared with persons in occupations of low socioeconomic status. Conclusions: Our results confirm the findings of previous studies that indicate a causal role of cigarette smoking in pancreatic cancer. Moreover, we found that the risk of former smokers may approach the risk of never smokers within a few years subsequent to quitting.     

Employment as butcher and cancer risk in a record-linkage study from Sweden

Objective: To investigate the risk of cancer among butchers and other meat workers in a large record-linkage study from Sweden. Methods: The Swedish Cancer Environment Register III contains nationwide data on cancer incidence during 1971–1989 for all residents, by occupation and industry of employment as reported at the 1960 and 1970 censuses. We identified 25,049 men classified as butchers or meat workers at either census. We used as a comparison group the remaining part of the active male population, after exclusion of workers with direct contact with animals. Results: Butchers in the meat industry had a slight increase in the risk of cancer (relative risk [RR] 1.1, 95% confidence interval [CI] 1.0–1.3), which was due to an increased risk of cancers of the oral cavity and pharynx (RR 1.6, 95% CI 1.0–2.7), stomach (RR 1.6, 95% CI 1.1–2.7), larynx (RR 1.4, 95% CI 0.6–3.4), and lung (RR 1.4, 95% CI 1.1–1.9). The risk of stomach cancer was highest during the first 5 years of the study, and among butchers from urban areas. No temporal or geographic variations were seen for lung cancer risk, with elevations restricted to squamous cell carcinoma. An increased risk of stomach, laryngeal and lung cancers was present in butchers and meat workers outside the meat industry. There was no clear indication of an increased risk of other neoplasms. Conclusions: The increased risk of oral, laryngeal, lung and stomach cancers among Swedish butchers may be at least partly due to confounding by tobacco smoking, alcohol drinking, and other lifestyle factors. However, exposures in the meat industry (e.g., viruses, nitrosamines, polycyclic aromatic hydrocarbons) may contribute the elevated cancer risks.     

Hormonal risk factors for endometrial cancer: modification by cigarette smoking (United States)

Objectives: To evaluate whether smoking modifies the risk of endometrial cancer associated with body mass index (BMI), postmenopausal hormone use, and other hormonal factors. Methods: Using multivariate adjusted models we examined interview data from a population-based case–control study of Wisconsin women (n = 740 cases, n = 2372 controls). Results: The relative risk for endometrial cancer associated with current smoking was 0.8 (95% CI: 0.6–1.0) compared to never smokers. No clear dose–response relationship was evident for pack-years smoked. When examined according to smoking status the risk associated with the highest quartile of BMI seemed to be greater among non-smokers (OR = 3.6, 95% CI: 2.4–5.3) than among current smokers (OR = 2.8, 95% CI: 1.4–5.6). Among postmenopausal women the risk associated with current use of postmenopausal hormones appeared to be greater among non-smokers (OR = 3.3, 95% CI: 2.3–4.9) than among current smokers (OR = 2.7, 95% CI: 1.3–5.5). Risk for long-term use (10 or more years) compared with never users was 8.3 (95% CI: 4.6–15.1) among never smokers and 2.5 (95% CI: 0.8–7.9) among current smokers. The risk associated with non-insulin-dependent diabetes was greater among non-smokers (OR = 2.5, 95% CI: 1.7–3.6) than current smokers (OR = 1.1, 95% CI: 0.4–3.1). There was no modifying effect of smoking on the risk associated with parity. Conclusion: These results suggest that smoking moderates the risk associated with endometrial cancer among women at greatest risk, specifically women who are obese or who use postmenopausal hormones.     

A meta-analysis of asthma and risk of lung cancer (United States)

Objective: To review epidemiologic evidence that asthma is a risk factor for lung cancer, with particular attention to the role of smoking on this association. Methods: Through MEDLINE computer searches (January 1966 to April 2002) and a review of references, we identified studies with quantitative data on the relation of asthma to lung cancer. Summary estimates of relative risks (RR) were calculated using the fixed-effects model or the random-effects model as appropriate. Results: The combined results from five case–control studies – that presented data limited to individuals who had never smoked – showed a 1.8-fold increase in lung cancer risk among asthmatics (95% confidence interval (CI) = 1.3–2.3). The results from six case–control and four cohort studies, which controlled for smoking history also were significant (RR = 1.7; 95% CI = 1.3–2.2). Finally, to illustrate the potential confounding by tobacco, we combined data from seven cohort studies that did not control for smoking history and found an overall RR of 1.4 (95% CI = 1.2–1.6). Conclusions: The increased risk of lung cancer among never smoking individuals with asthma supports a direct relation between asthma and lung cancer. There is biological evidence to support this association.     

A pooled analysis of case–control studies of thyroid cancer: cigarette smoking and consumption of alcohol, coffee, and tea

Objective: To analyze the role of smoking, alcohol, coffee and tea in relation to thyroid cancer, we conducted a pooled analysis of 14 case–control studies conducted in the United States, Europe, and Asia. Methods: The sample consisted of 2725 thyroid cancer cases (2247 females, 478 males) and 4776 controls (3699 females, 1077 males). Conditional logistic regression with stratification on study, age at diagnosis, and gender was used to compute odds ratios and 95% confidence intervals. Results: Thyroid cancer risk was reduced in persons who had ever smoked. The relationship was more pronounced in current smokers (OR = 0.6, 95% CI = 0.6–0.7) than former smokers (OR = 0.9, 95% CI = 0.8–1.1). There were significant trends of reduced risk with greater duration and frequency of smoking. For consumption of wine and beer, there was a significant trend of decreasing thyroid cancer risk (p = 0.02) that was not maintained after adjustment for current smoking (p = 0.12). Thyroid cancer risk was not associated with consumption of coffee or tea. These findings were consistent in both gender-specific and histology-specific (papillary and follicular) analyses. Conclusions: Pooled analyses of these geographically diverse case–control data indicate a reduced thyroid cancer risk associated with current smoking. A reduced risk associated with alcohol was eliminated after adjustment for smoking, and caffeinated beverages did not alter thyroid cancer risk.     

Cigarette Smoking, Alcohol Consumption, and Endometrial Cancer Risk: A Population-based Study in Sweden

Objective: To assess effects of cigarette smoking and alcohol consumption on the risk of endometrial cancer among postmenopausal women. Methods: We performed a nationwide population-based case–control study among postmenopausal women aged 50–74 years in Sweden, including 709 incident endometrial cancer cases and 3368 controls. Results: Compared to never smokers, recent/current smokers had a decreased risk of endometrial cancer (multivariate OR 0.61, 95% CI 0.47–0.80), but former smokers presented no substantial difference in risk (multivariate OR 0.90, 95% CI 0.72–1.14). We observed a decreased risk of endometrial cancer for postmenopausal smoking, but there was no clear impact on risk for premenopausal smoking. The inverse association of smoking with risk was not explained by differences in body mass index between smokers and nonsmokers. Alcohol consumption was not clearly associated with risk of endometrial cancer. The multivariate OR for women consuming up to 1.6 g of alcohol per day was 1.12 (95% CI 0.88–1.44), and 0.92 (95% CI 0.70–1.20) for women consuming more than 4 g per day (p for trend over categories=0.44). Conclusions: Current cigarette smoking reduces the risk of postmenopausal endometrial cancer, but the inverse association dissipates after smoking cessation. Premenopausal smoking might not affect risk of postmenopausal endometrial cancer. Alcohol consumption is not materially associated with risk.     

A prospective study on active and environmental tobacco smoking and bladder cancer risk (The Netherlands)

Objective: In a prospective cohort study among 120,852 adult subjects the authors investigated the associations between cigarette, cigar, pipe, environmental tobacco smoking (ETS), and bladder cancer. Methods: In 1986 all subjects completed a questionnaire on cancer risk factors. Follow-up for incident bladder cancer was established by linkage to cancer registries until 1992. The case–cohort analysis was based on 619 cases and 3346 subcohort members. Results: Compared with lifelong non-smokers the age- and sex-adjusted incidence rate ratios (RR) for ex- and current cigarette smokers were 2.1 (95% CI 1.5–3.0) and 3.3 (95% CI 2.4–4.6), respectively. The RR for smoking duration was 1.03 (95% CI: 1.02–1.04) per 1-year increment. The RR per 10 cigarettes/day was 1.3 (95% CI 1.2–1.4). Tar and nicotine exposure increased bladder cancer risk only weakly. It appeared that associations of cigarette smoking characteristics with bladder cancer risk were largely attributable to cigarette smoking duration only. Smoking cessation, age at first exposure, filter-tip usage, cigar and pipe smoking, and ETS were no longer associated with bladder cancer risk after adjustment for frequency and duration of smoking. Conclusions: The authors conclude that current cigarette smokers have a three-fold higher bladder cancer risk than non-smokers. Ex-smokers experience a two-fold increased risk. About half of male bladder cancer and one-fifth of female bladder cancer was attributable to cigarette smoking. Other smoking types (cigar, pipe, or ETS) were not associated with increased risks.     

Effect of type of alcoholic beverage on the risks of upper aerodigestive tract cancers in Brazil

Objectives: We investigated the effects of alcohol consumption on the risk of cancers of the upper aerodigestive tract (UADT) in a hospital-based case–control study in Brazil. Methods: A total of 784 cases of cancers of the mouth, pharynx, and larynx and 1578 non-cancer controls matched on age, gender, hospital area, and admission period provided information on alcohol drinking, smoking, and other characteristics via interview. Using logistic regression, we evaluated the relative risks (RR) of UADT cancer for different beverage types based on cumulative ethanol content exposure and frequency of consumption. Results: Relative to nondrinkers of any alcohol, risks of UADT cancers varied across sites both with increased exposure to ethanol and by alcohol type. RRs at equivalent levels of ethanol consumption were highest for cancers of the mouth for hard liquor (6.9 for > 100 kg lifetime consumption, 95% confidence interval (CI) = 2.8–17.1) and cachaça (4.5 for 101–500 kg, 95% CI = 2.2–9.0). Although RRs increased with frequency of drinks per week, when evaluated against higher proportional alcohol intake, reductions in risk were observed for beer and wine. Conclusion: Although methods of measurement can influence the interpretation of the carcinogenic nature of alcohols, increased RRs persisted with continued exposure for all types.     

Alcohol Consumption and Breast Cancer among Black and White Women in North Carolina (United States)

Objective: The purpose of this study was to investigate the effects of alcohol consumption on breast cancer risk in black and white women. Methods: We used data from the Carolina Breast Cancer Study, a population-based, case-control study of black and white women in North Carolina. Interviews were conducted with 890 cases and 841 controls frequency-matched on age and race. Results: Overall, the prevalence of moderate to high levels of alcohol consumption was low. Compared with abstainers, the multivariate odds ratio for recent intake of one or two drinks per day was 1.4 (95% CI = 0.9 2.1) and two or more drinks a day was 1.0 (95% CI = 0.6–1.6); increasing consumption was not associated with risk (p for trend = 0.6). The associations were similar, but somewhat weaker, for average lifetime consumption. Among women who consumed 91 g/week or more of alcohol, a nonsignificant increased risk of breast cancer was observed for women reporting binge drinking (OR = 1.5; 95% CI = 0.9–2.3), but not for those who consumed less than 91 g/week reporting binge drinking (OR = 1.0; 95% CI = 0.6–1.5). Odds ratios did not differ meaningfully by race, age, menopausal status, exogenous hormone use, or body mass index. Conclusions: These data provide little evidence for an association between alcohol consumption and risk of breas cancer among either black or white women.     

Birth characteristics, maternal reproductive history, hormone use during pregnancy, and risk of childhood acute lymphocytic leukemia by immunophenotype (United States)

Objective: To investigate the associations of birth characteristics and maternal reproductive factors with risk of childhood acute lymphoblastic leukemia (ALL) by immunophenotypic subtypes. Methods: Data collected from a case–control study including 1842 ALL cases (age < 15 years) and 1986 individually matched controls were analyzed. Exposure information was obtained through telephone interviews of parents. Results: Factors associated with risk of ALL from all subgroups combined included high birth weight (OR = 1.4, 95% CI = 1.1–1.8), high birth order (OR = 2.0, 95% CI = 1.3–3.0 for fourth-born child compared to first-born child), young maternal age (<20 compared to 25–29, OR = 1.4, 95% CI = 1.1–1.9), advanced paternal age ( > 39 compared to 25–29, OR = 1.4, 95% CI = 1.0–1.9), induced abortion prior to the index pregnancy (OR = 1.2, 95% CI = 1.0–1.4), and oral contraceptive use during the index pregnancy (OR = 1.5, 95% CI = 1.0–2.2) with children under the age of 2 (OR = 5.1, 95% CI = 1.0–24.7) being the predominantly affected group. Risk of early pre-B-cell ALL increased with advanced paternal age (OR = 1.7, 95% CI = 1.1–2.7) and high birth order (OR = 2.0, 95% CI = 1.1–3.6), while risk of pre-B-cell ALL increased with both younger (OR = 3.4, 95% CI = 1.4–8.4) and advanced maternal age (OR = 2.6, 95% CI = 1.1–5.9). T-cell ALL was associated with high birth weight (OR = 2.4, 95% CI = 1.1–5.5) and history of induced abortion (OR = 2.4, 95% CI = 1.3–4.5). Conclusion: This study suggests that the association of ALL with birth characteristics and maternal reproductive factors varies with the immunophenotype of the ALL. Future studies are needed to better understand the effect of maternal hormone in the development of subtype of childhood ALL.     

Influence of the persistence of tobacco and alcohol use in the appearance of second neoplasm in patients with a head and neck cancer. A case–control study

Objective  To evaluate the influence of persistent tobacco and alcohol use on the risk of a second metachronous neoplasm in the aerodigestive tract in head and neck squamous cell carcinoma (HNSCC) patients. Methods  A matched case–control study was carried out in 514 patients with HNSCC. Case patients developed a second metachronous neoplasm in the aerodigestive tract after treatment of an index HNSCC. A patient free of second neoplasm was individually matched to every case patient by location of the index tumor, tumor stage, sex, previous tobacco and alcohol consumption, age, general health status, and treatment. Data about persistence in tobacco and alcohol consumption after treatment of the index tumor was collected retrospectively. A validation study was carried out to confirm the adequacy of this retrospective information. Results  Persistent tobacco smoking and alcohol drinking after treatment of a HNSCC contributed to the risk of appearance of second neoplasm. The odds ratio of a second neoplasm for patients who continued to smoke was 2.9 (95% CI OR 1.8–4.1), and for patients who continued to use alcohol it was 5.2 (95% CI OR 3.3–7.9). There was a strong association between persistence of tobacco and alcohol use after treatment of the HNSCC index tumor. According to the attributable risk estimation, persistent tobacco and alcohol consumption would be responsible for one-third of the second neoplasms in the patients with a HNSCC index tumor. Conclusions  Persistence of tobacco and alcohol use after treatment of a HNSCC had a significant influence on the appearance of a second neoplasm in the aerodigestive tract. Cessation of tobacco and alcohol use should be a major goal after treatment of a HNSCC.     

Cancer incidence among marine engineers,a population-based study (Iceland)

Objectives: Marine engineers are in their occupation exposed to different chemicals, organic solvents, exhaust gases, oils, and petroleum products, and were formerly exposed to asbestos. The aim was to study the cancer pattern, with particular attention to lung and bladder cancer, in an Icelandic cohort of marine engineers, indirectly controlling for their smoking habits. Methods: A cohort of 6603 male marine engineers was followed up from 1955 to 1998, a total of 167,715 person-years. The cohort was record linked by the engineers' personal identification numbers to population-based registers containing the vital and emigration status and cancer diagnosis. Standardized incidence ratios (SIRs) were calculated for all cancers and different cancer sites in relation to different lag time and year of graduation. Information on smoking habits was obtained by administering a questionnaire to a sample of the cohort (n = 1501). Results: In the total cohort 810 cancers were observed, whereas 794 were expected (SIR 1.0, 95% CI 1.0–1.1), and significantly increased risk of stomach cancer (SIR 1.3, 95% CI 1.0–1.5) and lung cancer (SIR 1.2, 95% CI 1.0–1.5) was found. Increased risk of all cancers (SIR 1.2, 95% CI 1.1–1.3), stomach cancer (SIR 1.5, 95% CI 1.1–1.9), lung cancer (SIR 1.4, 95% CI 1.2–1.8), pleural mesothelioma (SIR 4.8, 95% CI 1.3–12.3), and urinary bladder cancer (SIR 1.3, 95% CI 1.0–1.8) were observed when a 40-year lag time was applied. The engineers' smoking habits were similar to those in a sample of the general population. The predictive value for lung cancer was 1.03. Conclusions: The increased risk for mesothelioma is possibly attributable to the previous asbestos exposure. The excess of lung cancer could also be related to asbestos exposure. The high incidence of stomach cancer, lung cancer, and bladder cancer may be related to exposure to chemical risk factors, such as oils and petroleum products, as confounding due to smoking seems to be ruled out. In the light of the limited exposure information in the present study the importance of the different occupational exposures needs to be evaluated in further studies.     

Alcohol consumption,smoking, and other risk factors and prostate cancer in a large health plan cohort in California (United States)

Alcohol consumption and cigarette smoking have been suggested as possible causes of prostate cancer. We therefore examined this relation in a cohort of 43,432 men who were members of a prepaid health plan in northern California (United States) and who had received a health examination in the period from 1979 through 1985. Detailed information on demographic variables, alcohol consumption, smoking habits, medical complaints and conditions, occupation, and surgery (including vasectomy) was assessed. Symptoms of prostatism and a history of sexually transmitted diseases were abstracted from the medical records of all prostate cancer patients and of a matched subsample of randomly selected control-subjects. Alcohol consumption was associated with no elevated prostate cancer risk for the 238 men in our study in whom prostate cancer developed, but smoking one or more packs of cigarettes per day was associated with an adjusted relative risk (RR) of 1.9 (95 percent confidence interval [CI]=1.2–3.1). Prostate cancer risk for Black men was 2.2 (CI=1.6–3.1) when compared with that for White men, and education level was associated positively in an increasing trend (P<0.02) up to an RR of 1.4 (CI=0.9–2.1) among men with postgraduate education. Symptoms of prostate hypertrophy were not associated with elevated risk of prostate cancer if they occurred two or more years before the diagnosis. The finding that smoking increased the risk of prostate cancer confirms the observations of others but needs cautious interpretation because we were unable to adjust for the potential confounding effect of dietary and hormonal factors.Presented in abstract form at the 118th annual meeting of the American Public Health Association and related organizations, New York City, September 30–October 4, 1990. The research was supported by US National Cancer Institute Contract N01-CP-95606 and by the Alcoholic Beverage Medical Research Foundation, Baltimore, Maryland, USA.     

Laryngeal cancer risk associated with smoking and alcohol consumption is modified by genetic polymorphisms in ERCC5, ERCC6 and RAD23B but not by polymorphisms in five other nucleotide excision repair genes

Laryngeal cancer is known to be associated with smoking and high alcohol consumption. Nucleotide excision repair (NER) plays a key role in repairing DNA damage induced by these exposures and might affect laryngeal cancer susceptibility. In a population‐based case‐control study including 248 cases and 647 controls, the association of laryngeal cancer with 14 single nucleotide polymorphisms (SNPs) in 8 NER genes (XPC, XPA, ERCC1, ERCC2, ERCC4, ERCC5, ERCC6 and RAD23B) was analyzed with respect to smoking and alcohol exposure. For genotyping, sequence specific hybridization probes were used. Data were evaluated by conditional logistic regression analysis, stratified for age and gender, and adjusted for smoking, alcohol consumption and education. Pro‐carriers of ERCC6 Arg1230Pro showed a decreased risk for laryngeal cancer (OR = 0.53, 95% CI 0.34–0.85), strongest in heavy smokers and high alcohol consumers. ERCC5 Asp1104His was associated with risk in heavy smokers (OR = 1.70, 95% CI 1.1–2.5). Val‐carriers of RAD23B Ala249Val had an increased cancer risk in heavy smokers (OR = 1.6, 95% CI 1.1–2.5) and high alcohol consumers (OR = 2.0, 95% CI 1.1–3.4). The combined effect of smoking and alcohol intake affected risk, at high exposure level, for ERCC6 1230Pro carriers (OR = 0.47, 95% CI 0.22–0.98) and RAD23B 249Val carriers (OR = 2.6, 95% CI 1.3–4.9). When tested for gene–gene interaction, presence of 3 risk alleles in the XPC‐RAD23B complex increased the risk 2.1‐fold. SNPs in the other genes did not show a significant association with laryngeal cancer risk. We conclude that common genetic variations in NER genes can significantly modify laryngeal cancer risk. © 2009 UICC     

A multiethnic population-based study of smoking, alcohol and body size and risk of adenocarcinomas of the stomach and esophagus (United States)

Objectives: Since the 1970s incidence rates for esophageal and gastric cardia adenocarcinomas have risen substantially, particularly among white males in the United States. Reasons for the increase of these tumor types are not well understood. We sought to determine the role of smoking, alcohol use, and body size characteristics in the etiology of esophageal, gastric cardia, and distal gastric adenocarcinomas. Materials and methods: A population-based case–control study that included Whites, African-Americans, Latinos and Asian-Americans diagnosed with incident esophageal (n = 222), gastric cardia (n = 277), and distal gastric adenocarcinomas (n = 443), and 1356 control subjects was conducted in Los Angeles County. Unconditional logistic regression was used to calculate odds ratios (ORs), as an estimate of the relative risk, and corresponding 95% confidence intervals (CIs) for the three tumor types of interest. Results: After adjustment for age, gender, race, birthplace, and education, current cigarette smoking was a significant risk factor for all tumor types; the association was strongest for esophageal adenocarcinomas (OR = 2.80, 95% CI = 1.8–4.3), intermediate for gastric cardia adenocarcinomas (OR = 2.12, 95% CI = 1.5–3.1), and weaker for distal gastric adenocarcinomas (OR = 1.50, 95% CI = 1.1–2.1). For esophageal adenocarcinomas only, cigarette smoking had a long-lasting deleterious effect, even 20 years after smoking cessation. Alcohol use was not associated with an increased risk of these tumor types. Risks of esophageal and gastric cardia adenocarcinomas also increased statistically significantly in a dose-dependent manner with increasing body mass index measured at ages 20 and 40 years and recently. The positive associations with smoking and body mass index were generally consistent when evaluated separately for Whites, non-Whites, males, and females. Conclusions: Cigarette smoking and high body mass index are significant, independent risk factors for esophageal and gastric cardia adenocarcinomas. Studies designed to understand the mechanisms whereby smoking and high body mass influence these tumor types are needed.     

Tobacco,alcohol, and diet in the etiology of laryngeal cancer: a population-based case-control study

Cancer of the larynx constitutes an increasingly important problem in Polish males during the last 25 years. A population-based case-control study of laryngeal cancer among people under 65 years of age was conducted in Lower Silesia, a province in Southwest Poland, from 1986 to 1987, with 249 newly-diagnosed cancer cases and 965 controls. The estimated relative risk (RR) for smoking and alcohol are both very high: for smoking more than 30 cigarettes, RR=59.7 (95 percent confidence interval [CI]: 13.0–274); for drinking vodka regularly for more than 30 years, RR=10.4 (95 percent CI: 4.0–27.2). Exposures to alcohol and tobacco show a clear multiplicative effect in all categories of exposure. The risk was shown to be reduced by quitting smoking (RR=0.3, 95 percent CI: 0.14–0.64, after 10 years) or by having a history of intermittent smoking. Poor nutrition was also identified as a strong independent risk factor. However, data quality regarding this factor is not as high as for tobacco and alcohol. Smoking alone accounts in this study for an estimated 95.2 percent of all the cases of laryngeal cancer.This work was conducted within the framework of the PR-6 Cancer Control in Poland-National Cancer Program and was supported by the Humboldt Foundation.     

Alcohol consumption increases the risk of fatal breast cancer (United States)

Objective: To investigate the hypothesis that alcohol consumption increases the risk of breast cancer mortality. Methods: We examined breast cancer mortality in relation to self-reported alcohol consumption in women from the American Cancer Society Cancer Prevention Study (CPS)-II. After 14 years of follow-up, 1,442 eligible breast cancer deaths were observed among 242,010 women. Cox proportional hazards models were constructed for total alcohol consumption and for beer, wine, and liquor separately. Results: Total alcohol consumption was associated with increased risk of fatal breast cancer among post- but not pre- or perimenopausal women. Even less than one drink/day was associated with up to a 30% increase in breast cancer mortality among postmenopausal women compared to non-drinkers (RR = 1.3, 95% CI: 1.1–1.6 for women drinking 0.26–<1 drink/day). When examined separately, consumption of beer, wine, and liquor each increased the risk of breast cancer among postmenopausal women. We found no evidence that alcohol consumption was more deleterious among women at high risk for breast cancer compared to average-risk women. Conclusion: This study adds to the evidence that postmenopausal women can reduce their risk of breast cancer by avoiding or minimizing their use of alcohol.     

Parental occupational exposures and risk of neuroblastoma: a case–control study (United States)

Background: A case–control study was conducted with 183 histologically confirmed neuroblastoma cases aged 0–14 years diagnosed among residents of New York State, excluding New York City, between 1976 and 1987. Three hundred seventy-two controls were selected from the New York State live birth certificate registry and were matched to cases on year of birth. Methods: Parental occupational exposures at the time of each child's birth were obtained from maternal telephone interviews, successfully completed for 85% of cases and 87% of controls. Results: Odds ratios were significantly elevated for maternal occupation in the service (OR = 2.0, 95% CI = 1.0–4.1) and retail (OR = 2.0, 95% CI = 1.1–3.7) industries and paternal occupation in materials handling (OR = 3.8, 95% CI = 1.1–14.6). Odds ratios were also significantly elevated for maternal report of occupational exposure to acetone (OR = 3.1, 95% CI = 1.7–5.6), insecticides (OR = 2.3, 95% CI = 1.4–3.7), lead (OR = 4.7, 95% CI = 1.3–18.2) and petroleum (OR = 3.0, 95% CI = 1.5–6.1) and paternal exposure to creosote (OR = 2.1, 95% CI = 1.1–4.3), dioxin (OR = 6.9, 95% CI = 1.3–68.4), lead (OR = 2.4, 95% CI = 1.2–4.8), and petroleum (OR = 1.8, 95% CI = 1.1–2.8). Conclusions: Due to the uncertainty of the biologic plausibility of these associations and the possibility of alternative explanations, these results should be interpreted cautiously.     

Breast cancers among very young premenopausal women (United States)

Objective: To assess risk factors for breast cancer among very young compared to older premenopausal women. Methods: Between 1990 and 1992 a population-based case–control study conducted in Atlanta, GA, Seattle/Puget Sound, WA, and central NJ interviewed 3307 premenopausal women aged 20–54 years. Logistic regression models estimated adjusted relative risks (RR) and 95% confidence intervals (CI) for each of three 10-year age groups. Results: Among the youngest age group (<35 years, n = 545), significant predictors of risk included African-American race (RR = 2.66; 95% CI 1.4–4.9) and recent use of oral contraceptives (RR = 2.26; 95% CI 1.4–3.6). Although these relationships were strongest for estrogen receptor-negative (ER–) tumors (RRs of 3.30 for race and 3.56 for recent oral contraceptive use), these associations were also apparent for young women with ER+ tumors. Delayed childbearing was a risk factor for ER+ tumors among the older premenopausal women (p _trend < 0.01), but not for women <35 years in whom early childbearing was associated with an increased risk, reflecting a short-term increase in risk immediately following a birth. Family history of early-onset breast cancer was more strongly associated with risk among women <35 years (RR = 3.22) than those 45–54 years (RR = 1.51). Risk factors for premenopausal breast cancer not significantly modified by age at diagnosis included early age at menarche, low body mass index, and heavy alcohol consumption. Conclusion: These findings suggest the possibility that women who develop breast cancers at very young ages may be etiologically as well as clinically distinct.