Smoking and risk of non-Hodgkin's lymphoma and multiple myeloma

Population-based case-control interview studies of 622 White men with non-Hodgkin's lymphoma and 820 controls from Iowa and Minnesota (United States) and 173 White men with multiple myeloma and 452 controls from Iowa offered the opportunity to investigate the relationship of these cancers with smoking. Risks were significantly elevated for all lymphoma (odds ratio [OR]=1.4), high-grade lymphoma (OR=2.3), and unclassified lymphoma (OR=2.8) for cigarette smokers. Dose-response gradients were not seen with intensity of cigarette use, but risks for these subtypes were greatest for cigarette smokers of longest duration. Similar elevations in risks were seen for tobacco users. The risk of multiple myeloma was not significantly elevated for either tobacco users or cigarette smokers. The findings from this study confirm the lack of an association between smoking and multiple myeloma and provide some support for an association between tobacco use and certain subtypes of non-Hodgkin's lymphoma.Ms Brown and Dr Blair are with the Epidemiology and Biostatistics Program, National Cancer Institute. Dr Everett is with the Department of Internal Medicine, Orlando Regional Medical Center, Orlando, FL, USA. Drs Gibson and Schuman are in the Department of Epidemiology, University of Minnesota, Minneapolis, MN, USA. Dr Burmeister is in the Department of Preventive Medicine, University of Iowa, Iowa City, IA, USA. Address correspondence to Ms Brown, Epidemiology and Biostatistics Program, National Cancer Institute, Executive Plaza North, Room 415C, Bethesda, MD 20892, USA. This work was supported in part by a grant from the National Institute of Environmental Health Sciences (ES 03099).     

Reproductive factors and risk of brain,colon, and other malignancies in Iowa (United States)

The influence of parity on the risk of cancers of the female breast and reproductive organs is well established. However, non-reproductive sites have received less attention. Mail questionnaire data gathered from incident female cases (169 brain; 332 colon; 260 rectal; 145 kidney; and 169 pancreas cancers), and 821 populationbased controls in Iowa (United States) were used to measure the effect of parity and age at first birth on risk of these malignancies. Relative to nulliparous women, ever-parous women were at significantly decreased risk of brain cancer (odds ratio [OR]=0.44, 95 percent confidence interval [CI]=0.3–0.7) and of colon cancer (OR=0.67, CI=0.5–0.97), after adjustment for age and other risk factors. The OR for the other sites did not differ significantly from 1.0. The lower risk of brain cancer among parous women was similar in younger and older age groups, in patients diagnosed with glioblastoma and astrocytoma, and among ever- and never-smokers. The findings for colon cancer are consistent with observations from other studies. In the context of limited laboratory and clinical evidence implicating hormones in brain neoplasia, these findings may suggest a role for hormonal factors in brain cancer etiology. Hormonal factors deserve more detailed future consideration as risk factors in brain cancer.Dr Cantor is with the Environmental Epidemiology Branch, National Cancer Institute, Bethesda, MD, USA. Dr Lynch and Ms Johnson are with the Department of Preventive Medicine and Environmental Health, University of Iowa, Iowa City, IA, USA. Address correspondence to Dr Cantor, Environmental Epidemiology Branch, National Cancer Institute, Executive Plaza North, Suite 443, Bethesda, MD 20892, USA. Supported in part by United States National Cancer Institute research contracts (NCI-NO1-CP-51026 and NCI NO1-CP-85614) and by a Public Health Service Preventive Oncology Academic Award (5 KO7 CA01181-05).     

Cigarette smoking and leukemia: results from the Lutheran Brotherhood Cohort Study

In a 20-year follow-up (1966–86) of 17,633 White males who described tobacco use in a mailed questionnaire sent in 1966, there were 74 deaths from leukemia (including 30 myeloid, 30 lymphatic, and 14 other and unspecified leukemia). Among men who ever smoked cigarettes, increased risks were observed for lymphatic (relative risk [RR]=2.7), and other and unspecified leukemia (RR=1.5); risks rose with increasing number of cigarettes smoked, although the dose-response relationship was statistically significant only for total leukemia. Mortality from myeloid leukemia was not elevated, except among those smoking over a pack of cigarettes per day. Results from this cohort support a relationship between cigarette smoking and leukemia. Further studies are needed to elucidate subtype associations with cigarette smoking.Drs Linet, McLaughlin, Hsing, Wacholder, and Blot are with the Epidemiology and Biostatistics Program, National Cancer Institute. Dr Co-Chien is at Westat, Inc., Rockville, Maryland, USA, Dr Schuman is at the University of Minnesota, Minneapolis, Minnesota, USA. Dr Bjelke is with the Center for Epidemiologic Research, University of Bergen, Norway. Address correspondence to Dr Linet, Epidemiology and Biostatistics Program, National Cancer Institute, Executive Plaza North Room 415B, Bethesda, MD 20892, USA.     

Cancer risk following polymyositis and dermatomyositis: a nationwide cohort study in Denmark

Polymyositis and dermatomyositis (PM/DM) have been associated with cancer, although the long-term risks are poorly understood. To evaluate the risk of cancer by time periods subsequent to PM/DM diagnosis, a cohort of 539 patients hospitalized with PM/DM in Denmark between 1977 and 1989 was identified from the Danish Central Hospital Discharge Register. Cancer incidence among cohort members was ascertained by linkage to the Danish Cancer Registry using a unique personal-identification number. The overall cancer risk was elevated significantly among patients with DM (standardized incidence ratio [SIR]=3.8, 95 percent confidence interval [CI]=2.6–5.4) and to a lesser extent PM (SIR=1.7, CI=1.1–2.4). Significant excesses were observed for cancers of lung, ovary, and lymphatic and hematopoietic system. However, the excess cancer incidence declined steadily with increasing years since initial diagnosis of PM/DM. The cancer risk was increased about sixfold (SIR=5.9, CI=3.8–8.7) during the first year, but was lower during the second year (SIR=2.5, CI=1.1–4.8), with no significant excesses in subsequent years of follow-up. These findings confirm that PM/DM may occur as a paraneoplastic syndrome that calls for steps aimed at early cancer detection and treatment. Among long-term survivors of PM/DM, however, there is little evidence to warrant extensive preventive and screening measures beyond those recommended for the general population.Drs Chow, McLaughlin, and Fraumeni, and Ms Gridley are with the Epidemiology and Biostatistics Program, Division of Cancer Etiology, National Cancer Institute, Bethesda, MD, USA. Dr McLaughlin is currently with the International Epidemiology Institute, Rockville, MD. Ms Mellemkjær and Dr Olsen are with the Division for Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark. Address correspondence to Dr Chow, National Cancer Institute 6130 Executive Blvd, EPN/415, Rockville, MD 20852, USA.     

Bladder cancer,parity, and age at first birth

The excess of bladder cancer in males (M:F ratio of 4:1 in the United States) is not explained fully by gender differences in smoking habits or occupational exposures. Laboratory studies suggest that some androgenic hormones stimulate (or do not inhibit) oncogenesis in bladder tissue, and that estrogenic hormones have the opposite effect. These observations suggest that bladder cancer risk in females may be modified by sex hormones which undergo profound changes during and following pregnancy. Mail-questionnaire data from 317 incident female cases, and 833 population-based controls in Iowa were used to measure the effect of parity and maternal age at first birth on bladder cancer risk. Parous women were at decreased risk relative to nulliparous women (odds ratio [OR]=0.67, 95 percent confidence interval [CI]=0.44–1.00), after adjustment for age, tobacco use, and previous bladder infection. The overall risk reduction was restricted to women who had never smoked (OR=0.51, CI=0.30–0.88), with no apparent effect of parity among ever-smokers (OR=0.93, CI=0.49–1.77). Risk appeared to decrease with increasing age at first birth, but did not vary with increasing parity after the first birth. Our findings are consistent with the hypothesis that oncogenesis in transitional cell tissue of the human bladder is influenced by sex hormones, and that hormonal changes related to pregnancy thereby can decrease risk.Dr Cantor is with the Environmental Epidemiology Branch, National Cancer Institute. Dr Lynch and Ms Johnson are in the Department of Preventive Medicine, University of Iowa School of Medicine, Iowa City, Iowa. Address correspondence to Dr Cantor at NCI, Executive Plaza North, Suite 443, Bethesda, MD 20892, USA. This research was supported in part by National Cancer Institute research contracts NCI-NO1-CP-51026 and NCI-NO1-CP-85614, and by a Public Health Service Preventive Oncology Academic Award, 5 KO7 CA01181-04.     

Prostate cancer and prediagnostic levels of serum vitamin D metabolites (Maryland,United States)

An hypothesis has been forwarded linking prostate cancer to low serum levels of vitamin D metabolites. We sought to test this hypothesis using sera obtained in a large, prospective cohort study. A serum bank in Washington County, Maryland (United States) has stored sera obtained from 20,305 county residents during a blood collection campaign undertaken in August through November 1974. We studied sera obtained from 61 residents who were diagnosed with prostate cancer during the period 1980 through 1992. Each prostate cancer case was matched to two controls on age (±1 yr) and race. Controls had donated blood in the same blood-collection campaign and had not been diagnosed with prostate cancer through 1992. Serum levels of vitamin D metabolites did not differ significantly between cases and controls. Mean 25-hydroxyvitamin D (25-D) levels were 34.3 ng/ml and 33.2 ng/ml, and mean 1,25-dihydroxyvitamin D (1,25-D) levels were 41.0 pg/ml and 40.1 pg/ml, in cases and controls, respectively. No statistically significant trends or differences between cases and controls were found in an analysis by quintile of serum level. We also did not observe the association of vitamin D metabolites with prostate cancer to be strongest among older men with more severe disease, as previously has been reported. In summary, although our study's power was limited, our findings provide little support for the hypothesis that vitamin D metabolite levels are associated strongly with subsequent risk for prostate cancer.Dr Braun is with the Epidemiology and Biostatistics Program, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA. Authors are also affiliated with the Department of Epidemiology, Johns Hopkins University School of Hygiene and Public Health, Baltimore, MD, USA (Drs Helzlsouer and Comstock), and the Department of Pediatrics, Medical University of South Carolina, Charleston, SC, USA (Dr Hollis). Address correspondence to Dr Braun, Epidemiology and Biostatistics Program, National Cancer Institute, EPN 443, Bethesda, MD 20892-7374, USA.     

Risk factors for small intestine cancer

Small intestine cancer is relatively rare. Clinical reports have suggested that several medical conditions may predispose to increased occurrence of this cancer, but otherwise its etiology is unknown. In one of the first case-control studies of this cancer, we compared questionnaire responses provided by next-of-kin of 430 persons who died of small intestine cancer cf921 controls who died of other causes. Subjects were identified from decedents included in the 1986 United States National Mortality Followback Survey. The questionnaires sought information on demographic and lifestyle characteristics, including diet and use of tobacco and alcohol. Tobacco and alcohol consumption were unrelated to risk of small intestine cancer, but weekly or more frequent consumption of red meat and monthly or more frequent intake of salt-cured/smoked foods were associated with two-to three fold increases in risk. The findings suggest that dietary factors probably are involved in risk of small intestine cancer, but additional research in other settings is required to clarify the determinants of these rare cancers.Drs Chow, Linot, McLaughlin, Hsing, and Blot are with the Epidemiology and Biostatistics Program, US National Cancer Institute, Bethesda, MD, USA. Mr Co Chien is with Westat, Inc., Rockville, MD, USA. Address correspondence to Dr Chow, Epidemiology and Biostatistics Program, Division of Cancer Etiology, 6130 Executive Blvd, EPN Room 403, Rockville, MD 20852, USA.     

A cohort study of smoking,alcohol consumption,and dietary factors for pancreatic cancer (United States)

Risk factors for pancreatic cancer were evaluated in a cohort study of 17,633 White men in the United States who responded to a mailed questionnaire in 1966 and were followed-up through 1986 for mortality. Cigarette smoking and alcohol consumption were found to be important risk factors for pancreatic cancer. Risks increased significantly with number of cigarettes smoked, reaching fourfold for smokers of 25 or more cigarettes per day relative to nonsmokers. Alcohol intake also was related significantly to risk, with consumers of 10 or more drinks per month having three times the risk of nondrinkers, but dose-response trends among drinkers were not smooth. Coffee consumption was unrelated to risk. Dietaryanalyses revealed a rising rate of pancreatic cancer mortality with increasing consumption of meat after adjustment for other risk factors. Men in the highest quartile of meat intake had about three times the risk of those in the lowest quartile. No consistent association, however, was observed for consumption of fruits, vegetables, or grains. This study confirms cigarette smoking as an important risk factor for pancreatic cancer, and provides evidence that elevated intake of alcohol and meat may increase the risk of this fatal malignancy.Drs Zheng (at the time of this study), McLaughlin, Gridley, Silverman, Wacholder, Blot, and Fraumeni Jr. are with the Epidemiology and Biostatistics Program, National Cancer Institute, Bethesda, MD, USA. Dr Zheng is currently with the School of Public Health, University of Minnesota, Minneapolis, MN, USA, as is Dr Schuman. Dr Bjelke is with the Center for Epidemiologic Research, University of Bergen, Bergen, Norway. Mr Co-Chien is with Westat, Inc., Rockville, MD, USA. Address correspondence to Dr McLaughlin, Biostatistics Branch, National Cancer Institute, 6130 Executive Blvd., Room 415, Bethesda, MD 20892, USA.     

Incidence of childhood cancer in twins

The incidence of childhood cancer in twins was evaluated by linking a roster of 30,925 twins born in Connecticut (United States) between 1930 and 1969 with the Connecticut Tumor Registry. Cancer, exclusive of nonmelanoma skin cancer, was identified in 19 females and 12 males under 15 years of age. The incidence rate among twins was 7.9 cancers per 100,000 person-years (PY) overall, and 9.7 and 6.1 per 100,000 PYs for females and males, respectively. Four of 13 leukemias occurred in two female twin pairs, representing concordance rates of 18 percent overall and 29 percent for like-sex pairs, which are somewhat higher than values reported previously. The number of cancers expected was computed on the assumption that twins experienced the same sex-, age-, and calendar time-specific cancer rates as recorded for all Connecticut-born children. Because active follow-up of individuals was not conducted, an adjustment to person-years of observation was made to account for childhood mortality, including the high perinatal mortality characteristie of twins. Childhood cancer was 30 percent less frequent than expected (standardized incidence ratio [SIR]=0.7; 95 percent confidence interval [CI]=0.5–0.9), a deficit that is marginally greater than those found in previous studies. Both leukemia (SIR=0.8; CI=0.4–1.4), and all other cancers combined (SIR=0.6; CI=0.3–0.9) occurred less often than expected. The deficit was greater among males (SIR=0.5; CI=0.2–0.8) than among females (SIR=0.9; CI=0.5–1.4) and was especially pronounced among males younger than five years (SIR=0.2; CI=0.0–0.7). The data support the view that twins, particularly male twins, have a lower risk of childhood cancer than single-born children. Any added risk for twins associated with their greater frequency of exposure to prenatal X-rays appears to have been insufficient to offset an effect of twinning per se. Possible explanations for this finding include (i) the low birthweight distribution of twins, or (ii) selective early mortality of twin fetuses or neonates who would otherwise have developed a clinical cancer.Drs Inskip, Boice, Stone, and Fraumeni are with the Epidemiology and Biostatistics Program, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA. Dr Harvey was in the Epidemiology and Biostatistics Program at the time of this research and is now with Sterling Drug, Malvern, PA, USA. Dr Matanoski is in the Department of Epidemiology, Johns Hopkins School of Hygiene and Public Health, Baltimore, MD, USA. Dr Flannery is with the Connecticut Tumor Registry, Hartford, CT, USA. Address correspondence to Dr Inskip, Radiation Epidemiology Branch, National Cancer Institute, Executive Plaza North, Room 408, Rockville, MD 20852, USA. This study was supported in part by Contract N01-CPO-1047 with the National Cancer Institute, US Public Health Service.     

Occupational risk factors for lung cancer among nonsmoking women: a case-control study in Missouri (United States)

Occupationally related risk of lung cancer among women and among nonsmokers has not been widely studied. A recently conducted population-based, case-control study in Missouri (United States) provided the opportunity to evaluate risk of lung cancer associated with several occupational factors. Incident cases (n=429) were identified through the Missouri Cancer Registry for the period 1986 through 1991, and included 294 lifetime nonsmokers and 135 ex-smokers who had stopped at least 15 years prior to diagnosis or had smoked for less than one pack-year. Controls (n=1,021) were selected through driver's license and Medicare files. Risk was elevated among women exposed to asbestos (ever: odds ratio [OR]=3.5, 95 percent confidence interval [CI]=1.2–10.0; >9 yrs: OR=4.6, CI=1.1–19.2) and pesticides (ever: OR=2.4, CI=1.1–5.6; >17.5 yrs: OR=2.4, CI=0.8–7.0). Risk also was elevated among dry cleaning workers (ever: OR=1.8, CI=1.1–3.0; >1.125 yrs: OR=2.9, CI=1.5–5.4). Occupational risks for lung cancer among women merit further study.Drs Brownson and Chang are with the Missouri Department of Health, Columbia, MO, USA. Dr Alavanja is with the Epidemiology and Biostatistics Program, National Cancer Institute, Rockville, MD, USA. Dr Chang directs the Missouri Cancer Registry with the Missouri Department of Health. Address correspondence to Dr Brownson, Division of Chronic Disease Prevention and Health Promotion, Missouri Department of Health, 201 Business Loop 70 West, Columbia, MO 65203, USA. This study was supported in part by US National Cancer Institute contracts NO1-CP7-1096-01 and NO1-CP7-1096-02.     

Race and sex differences in associations of vegetables,fruits, and carotenoids with lung cancer risk in New Jersey (United States)

We used data from a case-control study conducted in New Jersey between 1980 and 1983 to evaluate race and sex differences in associations of vegetable, fruit, and carotenoid consumption with lung cancer. Cases included 736 White males, 860 White females, 269 Black males, and 86 Black females with incident, histologically confirmed, primary cancer of the trachea, bronchus, or lung. Controls were identified through drivers' license and Health Care Financing Administration files and included 548 White males, 473 White females, 170 Black males, and 47 Black females. Usual intakes of vegetables (predominantly yellow/green) and fruit (predominantly yellow/orange) as well as other food sources of carotenoids were ascertained by a food frequency questionnaire. White females showed significant inverse associations of lung cancer with vegetables, fruit, and carotenoids. White males showed nonsignificant inverse associations with vegetables and carotenoids, and Black females just with vegetables. No inverse associations were found for Black males. Vegetable consumption was associated with risk of all histologic types of lung cancer, but the pattern of increasing risk with decreasing intake was limited to smokers. We infer that consumption of yellow/green vegetables and carotenoids may confer protection from lung cancer to White male and White female smokers. Further studies are needed to clarify the effect in Blacks.Drs Dorgan and Shaw are with the Division of Cancer Prevention and Control, and Drs Ziegler and Hartge, and Ms Falk are with the Division of Cancer Etiology, National Cancer Institute, Bethesda, MD, USA. Authors also are affiliated with the Special Epidemiology Program, New Jersey State Department of Health, Trenton, NJ, USA (Ms Schoenberg and Mr Wilcox) and Information Management Services, Inc., Silver Spring, MD, USA (Ms McAdams). Address correspondence to Dr Dorgan, Division of Cancer Prevention and Control, National Cancer Institute, Executive Plaza North, Room 211, Bethesda, MD 20892, USA.     

Leukemia,lymphoma, and multiple myeloma following selected medical conditions

The role of selected prior medical conditions in the etiology of hematopoietic malignancies was examined in a case-control study of members of two regional branches of the Kaiser Permanente Medical Care Program (USA). Past history of chronic infectious, autoimmune, allergic, and musculoskeletal disorders was abstracted from medical records for leukemia (n = 299), non-Hodgkin's lymphoma (NHL, n = 100), and multiple myeloma (n = 175) cases and matched controls (n = 787). Little difference was found between cases and controls for most of the chronic conditions evaluated, including sinusitis, carbuncles, urinary tract infections, pelvic infections, herpes zoster, asthma, rheumatoid arthritis, psoriasis, bursitis, and gout. Only three statistically significant elevated risks were found, i.e., with combined disc disease myeloma among patients with prior eczema and disk and other musculoskeletal conditions, and NHL following tuberculosis. Only two of these associations showed consistent patterns by sex and geographic region (myeloma with eczema and with musculoskeletal conditions). While prior history of eczema and musculoskeletal conditions may slightly increase risk of myeloma, this study provided little if any support for an association of chronic infectious, autoimmune, allergic, and musculoskeletal conditions with subsequent occurrence of the leukemias or NHL. Additionally, these data did not support a role for chronic antigenic stimulation, as defined in previous epidemiologic studies, in the etiology of hematopoietic malignancies.Ms Doody and Drs Linet, Pottern, Boice, and Fraumeni are with the Epidemiology and Biostatistics Program, National Cancer Institute. Dr Glass is with the Kaiser Permanente Medical Care Program, Northwest Region, Portland, Oregon, USA. Dr Friedman is with the Kaiser Permanente Medical Care Program, Northern California Region, Oakland, California, USA. Address correspondence to Ms Doody, Radiation Epidemiology Branch, National Cancer Institute, Executive Plaza North, Room 408, Bethesda, MD 20892, USA. This research was supported in part by National Cancer Institute contracts NO1-CP-01047, NO1-CP-01054, NO1-CP-11009, NO1-CP-11037, NO1-CP-31035, and NO1-CP-61006.     

Breastfeeding and breast cancer risk

A population-based case-control study of breast cancer with a focus on premenopausal women under 45 years of age, conducted in three geographic regions of the United States, enabled the evaluation of risk in relation to varying breastfeeding practices. Among premenopausal parous women (1,211 cases, 1,120 random-digit-dialing controls), a history of breastfeeding for two or more weeks was associated with a relative risk (RR) of 0.87 (95 percent confidence interval [CI]=0.7–1.0). This relationship was not altered substantially by removing from the reference group women who had problems with breastfeeding in the first two weeks, including those with insufficient milk production. Risk was not related substantially to number of children breastfed or length of breastfeeding, although a relatively low risk was observed among those breastfeeding for the longest duration examined (RR=0.67, CI=0.4–1.1 for an average period per child of 72 or more weeks). Women who began to breastfeed at a young age (<22 years) experienced the greatest reduction in risk, but other timing parameters (e.g., interval since first or last breastfeeding) were not predictive of risk. Risks were not modified substantially by age or menopause status, although the number of menopausal subjects examined was limited. Use of medications to stop breast milk was unrelated to risk (RR=1.04). The results of this study do not support the notion that breastfeeding substantially reduces breast cancer risk; however, this may reflect the fact that most of our study subjects breastfed only for limited periods of time (average breastfeeding per child of 30 weeks). Further studies are needed to clarify the relationship of breastfeeding to breast cancer risk, and to determine possible etiologic mechanisms underlying any observed associations.Drs Brinton, Potischman, and Swanson are with the Environmental Epidemiology Branch, National Cancer Institute, Betbesda, MD, USA. Authors also are affiliated with the Special Epidemiology Program, New Jersey State Department of Health, Trenton, NJ, USA (Ms Schoenberg); Rollins School of Public Health, Emory University, Atlanta, GA, USA (Dr Coates); the Division of Epidemiology, Columbia University School of Public Health, New York, NY, USA (Dr Gammon); and the Fred Hutchinson Cancer Research Center, University of Washington, Seattle, WA (Drs Malone, Stanford, Daling). Address correspondence to Dr Brinton, Environmental Epidemiology Branch, National Cancer Institute, Executive Plaza North, Room 443, Bethesda, MS 20892, USA.     

The relation of body size to plasma levels of estrogens and androgens in premenopausal women (Maryland,United States)

We analyzed data from a cross-sectional study of 107 premenopausal women to evaluate the relations of height, weight, and body mass index (BMI) with plasma hormone levels. Participants were 20- to 40-year old women residing in Maryland (United States), whose reported menstrual cycle lengths were not more than 35 days and whose measured weights for height were 85 to 130 percent of desirable based on 1983 Metropolitan Life Insurance tables. Fasting blood specimens were collected on each of days 5–7, 12–15, and 21–23 of every participant's menstrual cycle and pooled to create follicular, midcycle, and luteal phase samples, respectively, for analysis. Adjusted for age, taller women had significantly higher follicular-phase plasma-estradiol levels percent difference/cm=1.5, 95 percent confidence interval [CI]=0.3–2.7, and heavier women had significantly lower plasma sex-hormone binding globulin (SHBG) levels averaged across the menstrual cycle phases (percent difference/kg=–1.2; CI=–1.9–0.6). Body weight within the range studied, however, was not related significantly to the concentration of SHBG-bound estradiol during any phase of the menstrual cycle. The results of this cross-sectional study suggest a possible mechanism by which height may influence breast cancer risk.Drs Dorgan, Brown, Schatzkin, Albanes, Taylor, and Mr Campbell are with the Division of Cancer Prevention and Control, National Cancer Institute, Bethesda, MD, USA. At the time of this research Dr Reichman was with the Division of Cancer Prevention and Control and now is with MERJRE Associates, Bethesda, MD. Authors also are affiliated with Beltsville Human Nutrition Research Center, ARS, USDA, Beltsville, MD (Dr Judd); Departments of Obstetrics and Gynecology and Medicine, University of Massachusetts Medical School, Worcester, MA, USA (Dr Longcope and Ms Franz); and Information Management Services, Inc., Silver Spring, MD (Ms Kable). Address correspondence to Dr Dorgan, Division of Cancer Prevention and Control, National Cancer Institute, Executive Plaza North, Room 211, Bethesda, MD 20892, USA.     

Oral contraceptives and primary liver cancer among young women

The association of oral contraceptive use with liver cancer was examined in a study of 76 deaths from primary liver cancer, 22 deaths from cancer of the intrahepatic bile ducts, and 629 controls among women aged 25 to 49 years. The subjects in the study are from the 1986 National Mortality Followback Survey, which included a questionnaire sent or administered to the next-of-kin of almost 20,000 deceased individuals in the United States. Information on a number of lifestyle factors was collected, including questions on oral contraceptive use. Increased risks of primary liver cancer were found for ever-users (odds ratio [OR]=1.6, 95 percent confidence interval [CI]=0.9–2.6), and for long-term (10 years) users (OR=2.0, CI=0.8–4.8) of oral contraceptives. When the analysis was restricted to subjects whose spouse or parent was the respondent, more pronounced risks were seen for ever-users (OR=2.7, CI=1.4–5.3) and long-term users (OR=4.8, CI=1.7–14.0). No clear excess risk was found for cancer of the intrahepatic bile ducts. This study, the largest to date, adds to the number of investigations demonstrating an increased risk of primary liver cancer with use, particularly long-term use, of oral contraceptives.Authors are with the Epidemiology and Biostatistics Program, Division of Cancer Etiology, National Cancer Institute, with the exception of Mr Co-Chion who is at Westat Inc., Rockville, MD, USA. Address reprint requests to Dr Hsing at the National Cancer Institute, Executive Plaza North, Room 415, Bethesda, MD 20892, USA.     

Dietary associations in a case-control study of endometrial cancer

Despite the established role of obesity in the etiology of endometrial cancer, limited data are available from analytical epidemiologic studies on the association of risk with dietary factors. A case-control study of 399 cases and 296 controls conducted in five areas of the United States from 1 June 1987 to 15 May 1990, enabled evaluation of risk related to dietary intakes adjusted for potential confounders. Caloric intake was associated modestly with increased risk (odds ratio [OR]=1.5,95 percent confidence interval [CI]=0.9–2.5 for highest cf lowest quartiles of intake), with the principal contributors being fat and protein calories. After adjustment for other risk factors, including body mass, increased risk was associated with higher intakes of fat. Several components of fat investigated were associated with increased risk, although associations were slightly stronger for saturated fat (OR=2.1, CI=1.2–3.7) and oleic acid (OR=2.2, CI=1.2–4.0) than for linoleic acid (OR=1.6, CI=0.9–2.8). Food-group analyses showed intake of complex carbohydrates—and specifically of breads and cereals—associated with reduced risks (OR=0.6, CI=0.4–1.1), whereas animal fat and fried foods were associated with elevated risks (OR=1.5 and 1.7, respectively). The relations of endometrial cancer with animal fat and complex carbohydrates were independent. No consistent associations were noted for intakes of cholesterol, fiber, vitamins A and C, individual carotenoids, or folate-rich foods. These data imply an etiologic role for a diet rich in total fat and/or animal fat and low in complex carbohydrates with endometrial cancer. These associations are consistent with a hormonal mechanism and were independent of the associations of obesity and other risk factors.Drs Potischman, Swanson, Brinton, and Hoover are with the Environmental Epidemiology Branch, National Cancer Institute, Bethesda, MD, USA. Authors are also affiliated with Information Management Services, Inc., Silver Spring, MD (Ms McAdams), and the Departments of Obstetrics and Gynecology at Bowman Gray School of Medicine, Winston-Salem, NC (Dr Barrett); University of California at Irvine Medical Center, Irvine, CA (Dr Berman); Milton S. Hershey Medical Center, Hershey, PA (Dr Mortel); University of Minnesota Medical School, Minneapolis, MN (Dr Twiggs); Rush Medical College, Chicago, IL (Dr Wilbanks). Address correspondence to Dr Potischman, Nutritional Epidemiology Section, Division of Cancer Etiology, National Cancer Institute, Executive Plaza North, Suite 443, Bethesda, MD 20892, USA.     

Cancer risk in patients with diabetes mellitus

Cancer incidence was ascertained in a population-based cohort of 51,008 patients in Uppsala, Sweden, who were given a discharge diagnosis of diabetes mellitus during 1965–83. Complete follow-up through 1984 with exclusion of the first year of observation showed that the observed number of cancers in females (1,294) was eight percent higher than expected (relative risk [RR]=1.1, 95 percent confidence interval =11.0–1.1), whereas in males the observed number (1,123) was close to the expected (RR=1.0, 0.9–1.1). Significantly increased risks of pancreatic (RR=1.4, 1.2–1.7), primary liver (RR=1.5, 1.2–1.7), and endometrial (RR=1.5, 1.2–1.8) cancers and a lower than expected number of prostatic cancers (RR=0.7, 0.7–09) were found in this cohort of diabetic patients. The excess risk of pancreatic cancer was similar in females and males and evident both during one through four years (RR=1.7, 1.4–2.1) and five through nine years (RR=1.3, 0.9–1.7) of follow-up, but not thereafter. A similar pattern was found for primary liver cancer, but the RRs were generally higher in males than in females.Drs Adami and Ekbom are with the Cancer Epidemiology Unit, University Hospital, Uppsala, Sweden; Dr Ekbom is also in the Department of Surgery. Drs McLaughlin and Silverman are with the Biostatistics Branch, National Cancer Institute, Bethesda, Maryland, USA. Dr Berne is in the Department of Internal Medicine, University Hospital, Uppsala, Sweden. Mr Hacker is with Information Management Services, Silver Spring, Maryland, USA. Dr Persson is in the Department of Obstetrics and Gynaecology, University Hospital, Uppsala, Sweden. Address correspondence to Dr Adami, Cancer Epidemiology Unit, University Hospital, S-751 85 Uppsala, Sweden. This research was supported by grants from the Swedish Cancer Society.     

Attributable risk of lung cancer in lifetime nonsmokers and long-term ex-smokers (Missouri,United States)

A population-based, case-control study of incident lung cancer among women in Missouri (United States) who were lifetime nonsmokers and long-term ex-smokers was conducted between 1986 and 1992. The study included 618 lung cancer cases and 1,402 population-based, age matched controls. Information on lung-cancer risk factors was obtained by interviewing cases, next-of-kin of cases (36 percent and 64 percent of the cases, respectively) and controls. Year-long radon measurements also were sought in every dwelling occupied for the previous five to 30 years. Population attributable risks (PAR) for specific risk factors were computed for all subjects, for lifetime nonsmokers, for long-term ex-smokers, by histologic cell type (i.e., adenocarcinoma cf nonadenocarcinoma) and for direct interviews with case (for living cases) and for next-of-kin interviews (for dead cases or cases too ill to complete an interview). The mean age at lung cancer diagnosis was 71 years, and nearly 50 percent of the lung cancers were histologically confirmed adenocarcinomas. Almost 40 percent of all lung cancers among lifetime nonsmokers and almost 50 percent of lung cancers among all subjects could be explained by the risk factors under study. Dietary intake of saturated fat and nonmalignant lung disease were the two leading identified risk factors for lung cancer among the lifetime nonsmokers, followed by environmental tobacco smoke, and occupational exposures to known carcinogens. A small nonsignificant risk was found for study subjects exposed to median domestic radon concentration of 4 pCi/l (25-year time-weight average). Since only a small fraction of the population is exposed at this level, it is estimated that the PAR for domestic radon was less than two percent in Missouri. The risk for saturated fat intake was similar for lifetime nonsmokers, ex-somkers, adenocarcinoma cases, and nonadenocarcinoma cases; however, the increased risk was much more pronounced for next-of-kin interviews (PAR=31 percent) than for interviews with the study subjects (PAR = nine percent). A similar pattern of PAR was identified among ex-smokers but, in this group, the lingering effect of a history of smoking was also very important. Along with saturated fat intake (PAR=20 percent), the combined effect of previous active and passive smoking even after 15 years of cessation of active smoking was responsible for more lung cancer than any other risk factor under study (PAR=59 percent).Drs Alavanja, Benicbou, Swanson, and Boice are with the Epidemiology and Biostatistics Program, National Cancer Institute, Bethesda, MD, USA. Dr Brownson is with the Department of Community Health, Saint Louis University School of Public Health, St Louis, MO, USA. Address correspondence to Dr Alavanja, Epidemiology and Biostatistics Program, National Cancer Institute, EPN/543, 6130 Executive Blvd, Bethesda, MD 20892, USA.     

Racial differences in the risk of invasive squamous-cell cervical cancer

To investigate reasons for the higher rates of invasive squamous-cell cervical carcinoma among Blacks than Whites in the United States, we examined data from a case-control study of cervical cancer conducted in five geographic areas of the US, supplemented by incidence data from the Surveillance, Epidemiology, and End Results (SEER) Program, and hysterectomy prevalence data from the Cancer and Steroid Hormone Study. We observed only minor differences between Blacks and Whites in the magnitude of relative risks associated with a long interval since last Pap smear, multiple sexual partners, cigarette smoking, a higher number of births, and low levels of income and education. Thus, differences in the strength of associations contributed little to the higher incidence rate in Blacks, but the prevalence of these risk factors, except for cigarette smoking, was higher in Blacks than Whites. The SEER incidence rate ratio of 2.3 for Blacks compared to whites was increased to 2.7 when incidence rates utilized denominators corrected for prevalence of hysterectomy, while the rate difference increased from 14.9 to 25.8 cases per 100,000 person-years (PY). We estimated further that, after adjustment for prevalence of hysterectomy, the incidence rate for women at the lowest levels of exposure to the risk factors evaluated was 2.2 times higher in Blacks than Whites, but that the corresponding rate difference was only 2.2 cases per 100,000 PYs. Thus, our results suggest that racial differences in the prevalence of exposure to identified risk factors account for most of the difference in incidence rates. It remains to be determined what, as yet unidentified, aspects of lower socioeconomic status contribute to the higher incidence rate in Blacks.Authors are with the Epidemiology and Biostatistics Program, Division of Cancer Etiology, National Cancer Institute. Address correspondence to Ms Schairer, Environmental Epidemiology Branch, National Cancer Institute, National Institutes of Health, Executive Plaza North, Room 443, Bethesda, MD 20892, USA.     

Temporal trends in the incidence of cutaneous malignant melanoma among Caucasians in the San Francisco-Oakland MSA

Temporal changes in the incidence of cutaneous malignant melanoma (CMM) were examined in the San Franscisco-Oakland (California, United States) Metropolitan Statistical Area (MSA) between 1976 and 1987, using data from the population-based cancer registry. This analysis was conducted after the completion of a project designed to eliminate bias in the reporting of CMM due to changes in medical practice. The incidence of CMM is higher in the San Francisco-Oakland MSA than nationally. From 1976 through 1987, the incidence of invasive CMM increased from 9.8±0.9 to 16.5±1.1 per 100,000 (P=0.0001) among men and from 9.3±0.8 to 12.7±0.9 per 100,000 (P=0.001) among women. Age-specific, histologic-specific, and anatomic site-specific trends were also evaluated. The temporal patterns of CMM suggest that the recent increases are not accounted for solely by ascertainment bias due to reporting practices. The observed trends are consistent with early detection efforts and with changes in the prevalence of risk factors.Authors are at the Northern California Cancer Center, Alameda, CA, USA. Dr Holly is also with the Department of Health Research and Policy, Stanford University School of Medicine, and the Department of Epidemiology and Biostatistics, University of California, San Francisco, School of Medicine, USA. Address correspondence to Dr Horn-Ross, Northern California Cancer Center, 1420 Harbor Bay Pkwy, Suite 260, Alameda, CA 94501, USA. This research was supported by contract N01-CN-05224 from the Surveillance, Epidemiology, and End Results Program, National Cancer Institute, contract N01-CP-05681 from the National Cancer Institute, and subcontract 050E-8708 from the California Tumor Registry, California State Department of Health Services.