Determinants of melanoma in a case-control study of twins (United States)

Objectives: To estimate the relative contribution of environmental and genetic factors disposing towards the development of melanoma. Methods: We investigated risk factors for melanoma in a case–control study conducted among 185 North American twin pairs in which one was diagnosed with melanoma and the other (the co-twin of the case) was not. We considered monozygous (MZ) and dyzygous (DZ) twins separately. Results: While greater risk of melanoma was associated with number of large nevi in DZ twins (adjusted OR = 26.6 (4.2–170.8) for three or more large moles), this was not the case for MZ twins (adjusted OR = 1.4 (0.6–3.2) for three or more large moles). Elevated risks of developing melanoma with site-specific sun exposures resulting in sunburn also appeared to be confined to DZ twins. Despite the number of identical twins reporting a difference in mole prevalence, we observed only a modest and inconsistent increase in melanoma risk attributable to that factor, whereas the increase within fraternal twin pairs was larger by an order of magnitude. Conclusions: If confirmed, this indicates that the significance of mole prevalence as a risk factor for melanoma is largely as a genetic, rather than an environmental, factor. We noticed an increased risk of developing melanoma among DZ twins who drank moderate amounts of beer, but we believe this result may be due to the sun-exposure activities of those people most likely to drink moderate amounts of beer.     

The relation of breast size to breast cancer risk in postmenopausal women (United States)

Objective: Breast size has been hypothesized to predict a woman's risk of breast cancer although studies in the main have not supported an association. In a large, population-based case-control study we examined whether breast size might emerge as a significant risk factor among very lean women in whom breast size might be a truer reflection of the volume of gland mass at risk for malignant change.Methods: The data derive from a population-based case- control study of women aged 50 to 79 years conducted in several New England states and Wisconsin. Incident cases of invasive breast cancer (n=2015) were identified through state tumor registries and controls (n=2556) were selected at random within age strata from population lists. Telephone interviews were conducted to obtain information on known and suspected risk factors which included bra dimensions (cup and back size) prior to a first birth, or at the age of 20 for nulliparous women.Results: We observed a significant positive association for cup size which was limited to women who were the most lean as young adults based on chest circumference. Among those reporting a chest size under 34 inches multivariate-adjusted relative risks were 1.34 (95% CI: 1.04 to 1.74) for cup size B, and 1.76 (95% CI: 1.04 to 3.01) for cup size C and larger, compared to a cup size smaller than B, and the trend for increasing cup size was statistically significant (P=0.005). There was no relation with breast size among women reporting an average or larger back circumference (34 inches or larger).Conclusion: Breast size before a pregnancy is a positive predictor of postmenopausal breast cancer, but this association is limited to those who were especially lean as young women.     

Weight change and risk of postmenopausal breast cancer (United States)

Objective: Although many studies have shown that higher weight increases the risk of postmenopausal breast cancer, some aspects of this association are unclear. In order to examine the risk associated with different patterns of weight change, we analyzed data from a large case–control study of postmenopausal breast cancer. Methods: Participants included women aged 50–79 years (n = 5031) who are newly diagnosed with invasive breast cancer in Massachusetts, New Hampshire, and Wisconsin. Similarly-aged population controls (n = 5255) were selected at random from driver's license files and Medicare beneficiary lists. Height, weight, and information on other breast cancer risk factors were ascertained by structured telephone interviews from 1992 to 1995, and logistic regression was used to estimate multivariable-adjusted odds ratios (OR) and 95% confidence intervals (CI). Results: Women in the top quintile groups for height at age 20, recent weight, and recent body mass index had significantly increased risks of breast cancer. Among women who reached their highest adult weight at younger ages (45 years), increasing weight loss since that age was associated with a reduced risk of postmenopausal breast cancer (OR 0.90, CI 0.84–0.98, per 5 kg). However, weight loss among women whose highest weight occurred after age 45 was not associated with risk (OR 1.00, CI 0.95–1.05, per 5 kg). Weight gain since the lowest adult weight increased risk by 8% for each 5 kg of gain (OR 1.08, CI 1.06–1.11). Temporary weight cycling (weight loss followed by weight gain) was not associated with increased risk. Conclusions: Weight gain clearly increased risk of postmenopausal breast cancer. These data lend further support to efforts aimed at helping women avoid weight gain as they age.     

Allergy and risk of breast cancer among young women (United States)

Objective: To investigate the relationship between allergy and risk of breast cancer in women 45 years of age and younger. Methods: Data were analyzed from a population-based case–control study of breast cancer in western Washington. Cases were women born after 1944 who were diagnosed with invasive breast cancer (n = 747) between January 1983 and April 1990. Controls (n = 958) were similarly aged women ascertained through random-digit dialing. Cases and controls were interviewed about their history of doctor diagnosed allergies, including detailed information on the specific types of allergies and the age of onset. Using logistic regression we examined the associations between allergy history and breast cancer. Results: A history of allergies was associated with a reduced risk of breast cancer for women older than 35 (odds ratio (OR) = 0.77; 95% confidence interval (CI) = 0.60–0.99), but not for women 35 years or younger (OR = 1.30; 95% CI = 0.94–1.81). There was little difference in effect when age of first allergy onset was examined. No specific type of allergy was associated with breast cancer risk. Conclusion: Our results provide some evidence that a history of allergy may be associated with a reduced risk of breast cancer for women who develop breast cancer between 35 and 45 years of age. Future studies are needed to verify the relationship between immune responses and breast cancer risk.     

Relation of serum insulin-like growth factor-I (IGF-I) and IGF binding protein-3 to risk of prostate cancer (United States)

Objective: Insulin-like growth factor I (IGF-I) exerts potent mitogenic and antiapoptotic effects on prostatic epithelial cells. Insulin-like growth factor binding protein-3 (IGFBP-3) modulates the effects of IGF-I, and independently induces apoptosis and inhibits cell growth. Previous studies have inconsistently associated IGF-I and IGFBP-3 with prostate cancer. To try and further clarify these potential associations, we undertook a sibling-matched case–control study. Methods: Serum IGF-I and IGFBP-3 were determined for 845 men (408 cases and 437 sibling controls). Conditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the association between the serum IGF levels and prostate cancer. Results: Among all study subjects, only the molar ratio of IGF-I to IGFBP-3 was associated with prostate cancer: comparing those in the highest to lowest quartiles gave an OR = 1.62 (95% CI = 1.02–2.57, trend-p = 0.04). Among men with clinically less aggressive disease, we observed positive associations between prostate cancer and high levels of IGF-I (OR = 2.78, 95% CI = 1.06–6.80, trend-p = 0.03), and IGFBP-3 (OR = 2.68, 95% CI = 1.08–6.80, trend-p = 0.04). Simultaneously modeling both left the IGF-I result essentially unchanged, while substantially weakening the IGFBP-3 association. Conclusions: We found that a high IGF-I to IGFBP-3 molar ratio was associated with an increased risk of prostate cancer. Furthermore, high IGF-I was associated with increased risk of prostate cancer among men with less advanced disease at diagnosis. These results lend support to the hypothesis that IGF-I, or the IGF-I to IGFBP-3 molar ratio, is an important risk factor for prostate cancer.     

Cigarette smoking,glutathione-s-transferase M1 and t1 genetic polymorphisms,and breast cancer risk (United States)

Objective: It has been suggested that functional polymorphisms in genes encoding tobacco carcinogen-metabolizing enzymes may modify the relationship between tobacco smoking and breast cancer risk. We sought to determine if there is a gene–environment interaction between GSTM1 (GSTM1A and GSTM1B), and GSTT1 genotypes and cigarette smoking in the risk of breast cancer. Methods: Cases and controls were recruited in a case–control study conducted in Connecticut from 1994 to 1998. Cases were histologically confirmed, incident breast cancer patients, and controls were randomly selected from women histologically confirmed to be without breast cancer. A total of 338 cases and 345 controls were genotyped for GSTM1 and GSTT1. Results: None of the GSTM1 genotypes, either alone or in combination with cigarette smoking, was associated with breast cancer risk. There was, however, a significantly increased risk of breast cancer among postmenopausal women with a GSTT1 null genotype (OR = 1.9, 95% CI 1.2–2.9). There were also indications of increased risk of breast cancer associated with cigarette smoking for postmenopausal women with GSTT1-null genotype, especially for those who commenced smoking before age 18 (OR = 2.9, 95% CI 1.0–8.8). Conclusion: Women with a GSTT1-null genotype may have an increased breast cancer risk, especially postmenopausal women who started smoking at younger ages.     

Prostate cancer screening and mortality: A case-control study (United States)

OBJECTIVE: We performed a case-control study at Kaiser Permanente Northwest to assess the association between digital rectal examination (DRE) and prostate-specific antigen (PSA) testing, separately and together, and prostate cancer mortality. METHODS: We identified 171 KPNW members who died as a result of prostate cancer from 1992 to 1999 and 342 randomly-selected KPNW members matched to the cases on age, sex, and length of plan membership. History of screening was determined from medical records and laboratory databases for cases and controls. RESULTS: DRE and/or PSA screening at any time up to and including the case diagnosis date had taken place among 69.0% of cases and 74.6% of controls. After using logistic regression analysis to adjust for matching variables and a provider diagnosis of benign prostatic hypertrophy (BPH), we found an inverse association between receipt of a prostate cancer screening test and prostate cancer mortality (odds ratio (OR): 0.70, 95% confidence interval (CI): 0.46 - 1.1). Most of the screening tests were DREs, and it was not possible to assess the separate influence of PSA screening. CONCLUSIONS: The results of this study suggest that men who have been screened for prostate cancer have a reduced risk of dying as a result of this disease.     

Reproductive factors in relation to breast cancer characterized by p53 protein expression (United States)

Objective: To evaluate the potential etiologic heterogeneity of breast cancer by examining whether associations with reproductive and other personal characteristics differed by p53 protein expression status. Methods: Data from the Carolina Breast Cancer Study, a population-based, case–control study of 861 cases and 790 controls, were utilized. Immunohistochemical staining for the p53 protein was performed on 638 archived tumor specimens; 46% of cases were classified as p53+. Two separate unconditional logistic regression models were used to calculate odds ratios (OR) and 95% confidence intervals (CI) for p53+ and p53– breast cancer relative to controls for reproductive and other personal characteristics. Analyses were performed separately for younger (45 years) and older (>45 years) women. Results: Risk factor profiles largely overlapped for p53+ and p53– breast cancer, with the exception of oral contraceptive (OC) use among younger women and a family history of breast cancer. Prolonged OC use was more strongly associated with p53+ breast cancer [OR 3.1 (95% CI: 1.2–8.1) than p53– breast cancer (OR 1.3 (95% CI: 0.6–3.2)] among younger women only. A first-degree family history of breast cancer was associated with p53+ breast cancer among younger women [OR 1.5 (95% CI: 1.0–2.2)] and older women [OR 1.4 (95% CI: 0.9–2.3)], but not p53– breast cancer in either age-group. Conclusions: These results provide little evidence of breast cancer heterogeneity as classified by p53 expression status. However, although not statistically significant, OC use among younger women and family history of breast cancer may operate through a pathway involving p53 alterations to increase risk of breast cancer.     

Antibiotic exposure and breast cancer in New Zealand

Summary Aims To investigate the hypothesis that use of antibiotics is related to subsequent development of breast cancer and also to apply this theory to other cancer types. Materials and methods A nested case–control study was conducted, using data linkage between the RNZCGP Research Unit database and the New Zealand Hospital Separation Diagnosis database. Cancer related hospital admissions were identified between 1998 and 2002, and prior antibiotic exposure in these patients was then found.Results A total of 6678 patients were identified with a newly diagnosed cancer in this time period. A slightly increased odds ratio (OR) (95% CI) for breast cancer was seen with penicillin, 1.07 (1.02–1.13). Penicillin was also associated with an increased OR with lung and respiratory cancer, 1.13 (1.06–1.21), and skin neoplasms, 1.05 (1.02–1.08). Significant associations were seen between macrolides and leukaemia, 1.15 (1.01–1.30), lung and respiratory cancers, 1.23 (1.10–1.38) and non-Hodgkin’s lymphoma, 1.26 (1.02–1.55). Tetracyclines were significantly associated with non-Hodgkin’s lymphoma, 1.12 (1.01–1.24). Cephalosporins only showed a significant association with leukaemia, 1.35 (1.06–1.71), sulphonamides with colorectal cancers, 1.12 (1.01–1.24), and ‘other‘ antibiotic classes with bladder and renal cancers, 1.34 (1.07–1.67). Conclusions It is most likely that antibiotic exposure represents a confounding factor rather than a causation for breast cancer and other cancer types.     

Association of mammographically defined percent breast density with epidemiologic risk factors for breast cancer (United States)

Objective: Mammographically defined percent breast density is an important risk factor for breast cancer, but the epidemiology of this trait is poorly understood. Although several studies have investigated the associations between reproductive factors and density, few data are available on the associations of breast density and waist-to-hip ratio (WHR), physical activity, education, alcohol and smoking. Methods: We investigated the associations of known and suspected breast cancer risk factors with breast density in a large breast cancer family study. Information was collected on members of 426 families through telephone interviews, mailed questionnaires and mammography. Mammographic films on 1900 women were digitized and breast density was estimated in discrete five-unit increments by one radiologist. Analysis of covariance techniques were used and all analyses were performed stratified by menopausal status. Results: Similar to other reports, nulliparity, late age at first birth, younger age and lower body mass index were associated with increased percent density in both premenopausal and postmenopausal women, and hormone replacement therapy among postmenopausal women. Higher levels of alcohol consumption and low WHR were associated with increased percent density among both premenopausal and postmenopausal women (differences of 3–11% between high and low categories). However, smoking and education were inversely associated with percent density among premenopausal (p = 0.004 and p = 0.003, respectively) but not postmenopausal women (p = 0.52 and p = 0.90). Physical activity was not associated with percent density in either stratum (p values > 0.25). Combined, these factors explained approximately 37% of the variability in the percent density measure in premenopausal women and 19% in postmenopausal women. Conclusions: Many of these factors may potentially affect breast cancer risk through their effect on percent breast density.     

High breast cancer incidence rates among California teachers: results from the California Teachers Study (United States)

Objective: To determine risk factor profiles and cancer incidence rates among participants in the California Teachers Study (CTS), a study designed to document high breast cancer incidence rates of California teachers and to investigate emergent hypotheses in the etiology of breast and other cancers. Methods: The CTS is a prospective study of 133,479 California female teachers and administrators, established in 1995–1996 with members of the California State Teachers Retirement System completing a detailed mailed questionnaire regarding possible risk factors for breast and other cancers. Cancer outcomes were identified by linkage with the California Cancer Registry. Results: CTS participants have a 51% higher age-standardized invasive breast cancer incidence rate and a 67% higher in-situ breast cancer incidence rate than would be expected based on race-specific statewide rates after three years of follow-up. CTS participants also have substantially elevated rates of endometrial cancer (rate ratio, RR = 1.72), ovarian cancer (RR = 1.28), melanoma (RR = 1.59), non-Hodgkin's lymphoma (RR = 1.53), and leukemia (RR = 1.28), but low rates of invasive cervix cancer (RR = 0.53) and lung cancer (RR = 0.66). Conclusions: CTS members have high rates of several major cancers, particularly breast cancer, and low rates of lung and cervix cancer. Although late age at first birth can explain a portion of the observed excess risk of breast cancer in this cohort, the unique risk factor profile of CTS members may account for much of their higher risk of breast and selected other cancers. The CTS offers a rich resource for future studies of cancer risk and of women's health, in general.     

Association of diabetes mellitus with prostate cancer: Nested case–control study (Prostate testing for cancer and Treatment study)

Observational studies suggest that diabetes is associated with a decreased risk of prostate cancer, but few are population based or have investigated associations with cancer stage or duration of diabetes. We report a case–control study nested within the population‐based Prostate testing for cancer and Treatment (ProtecT) study ISRCTN20141297. Men aged 50–69 years based around 9 UK cities were invited for a prostate‐specific antigen (PSA) test between June 2002 and November 2006. Amongst 55,215 PSA‐tested men, 1,966 had histologically confirmed prostate cancer; of these, 1,422 (72.3%) completed the questionnaire and 1,291 (65.7%) had complete data for analysis. We randomly selected 6,479 age‐ (within 5 years) and general practice‐matched controls. The prevalence of diabetes was 89/1,291 (6.9%) in cases and 555/6,479 (8.6%) in controls. Diabetes was associated with a reduced risk of prostate cancer (odds ratio = 0.78; 95% confidence interval: 0.61–0.99). There was weak evidence that the inverse association was greater for well‐ versus poorly differentiated cancers (p = 0.07). The magnitude of the inverse association did not change with increasing duration of diabetes (p for trend = 0.95). Diabetes is associated with a decreased risk of PSA‐detected prostate cancer. These data add to the evidence of the association of diabetes with prostate cancer in the PSA era.     

Nutrient intake and ovarian cancer: an Italian case-control study

Objective: The role of selected macronutrients, cholesterol, and fatty acids in the etiology of epithelial ovarian cancer was analyzed using data from a case–control study carried out in five Italian areas between January 1992 and December 1999. Methods: Cases comprised 1031 women with incident, histologically confirmed epithelial ovarian cancer, admitted to the major teaching and general hospitals of the study areas. Controls comprised 2411 women admitted for acute, non-neoplastic conditions to the same network of hospitals. Information on dietary habits was elicited using a validated food-frequency questionnaire including 78 food groups and recipes. Odds ratios (OR) and their corresponding 95% confidence intervals (CI) were computed by subsequent quintiles of nutrient intake. Results: Direct associations with ovarian cancer emerged for starch intake (OR = 1.4 in the highest vs the lowest quintile of intake; 95% CI 1.1–1.8), while inverse associations emerged for monounsaturated (OR = 0.7; 95% CI 0.5–0.9), and polyunsaturated (OR = 0.7; 95% CI 0.5–0.9) fatty acids. Among fatty acids, oleic (OR = 0.7; 95% CI 0.5–0.9), linoleic (OR = 0.7; 95% CI 0.5–0.9), and linolenic (OR = 0.8; 95% CI 0.6–1.0) acids were inversely related to ovarian cancer. When, however, six macronutrients were included in the same model, only the adverse effect of high starch intake remained significant. Results were consistent in separate strata of menopausal status, parity, and energy intake. Conclusions: Starch was directly associated, and unsaturated fatty acids were inversely associated, with ovarian cancer risk.     

Active and passive tobacco smoking and the risk of borderline and invasive ovarian cancer (United States)

Objective: A population-based case–control study was conducted to examine the hypothesis that active and passive tobacco smoking were associated with the risk of epithelial ovarian cancer. Methods: In-person interviews were obtained from 558 women with epithelial ovarian cancer (431 invasive, 127 borderline) and 607 population controls regarding active lifetime tobacco smoking, environmental tobacco smoke exposure in utero and during childhood, and other factors that may be related to the development of ovarian cancer. Results: No significant associations of ever or former tobacco smoking with the risk of invasive or borderline ovarian cancer were found, although long-term ex-smokers of 20 years or more were at significantly reduced risk of invasive cancer. Significant, positive dose–response relations of the number of cigarettes smoked per day and pack-years with the odds ratios for borderline cancer were evident. No association of active tobacco smoking with risk was found by histologic subtype of invasive ovarian cancer. Smokers were at significantly increased risk for borderline serous cystadenoma (OR: 1.91; 95% confidence intervals, CI: 1.09–3.34), but not for borderline mucinous cystadenoma. When we limited the analyses to current smokers, age-started smoking was significantly inversely related to the risk of invasive, but not borderline ovarian cancer. We found no association of gestational or childhood environmental tobacco smoke exposure with the risk of invasive or borderline ovarian cancer among never smokers. Conclusions: These findings do not support an association of active tobacco smoking with the risk of invasive ovarian cancer. An increased risk of borderline serous cystadenoma among smokers must be viewed with caution.     

Micronutrients and laryngeal cancer risk in Italy and Switzerland: a case-control study

Objective: To investigate the relation between various micronutrients and laryngeal cancer risk. Methods: A case–control study was conducted in Italy and Switzerland between 1992 and 2000. Cases were 527 patients with incident cancer of larynx, admitted to the major teaching and general hospitals of the study areas. Controls were 1297 subjects admitted for acute, non-neoplastic diseases to the same network of hospitals. Dietary habits were assessed using a validated food-frequency questionnaire. Odds ratios (OR) and their corresponding 95% confidence intervals (CI) were computed using multiple logistic regression. Results: Significant inverse relations emerged between laryngeal cancer risk and intake of vitamin C (OR = 0.2, for the highest versus the lowest intake quintile; 95% CI: 0.2–0.4), -carotene (OR = 0.2; 95% CI: 0.2–0.4), -carotene (OR = 0.3; 95% CI: 0.2–0.5), lutein/zeaxanthin (OR = 0.4; 95% CI: 0.3–0.6), vitamin E (OR = 0.4; 95% CI: 0.3–0.6), -criptoxanthin (OR = 0.4; 95% CI: 0.2–0.5), folic acid (OR = 0.4; 95% CI: 0.2–0.6), thiamin (OR = 0.4; 95% CI: 0.3–0.6), glutathione (OR = 0.5; 95% CI: 0.4–0.8), reduced glutathione (OR = 0.6; 95% CI: 0.4–0.8), vitamin B6 (OR = 0.6; 95% CI: 0.4–0.9) and potassium (OR = 0.6; 95% CI: 0.4–0.9). Direct associations were found with zinc (OR = 1.5; 95% CI: 1.0–2.2) and vitamin D (OR = 1.8; 95% CI: 1.2–2.6). Combining low intakes of vitamin C, carotene, vitamin E, and folate with heavy smoking and drinking led to ORs between 80 and 170. Conclusions: This study provides further support that, independently from smoking and alcohol consumption, the intake of several micronutrients, including selected antioxidants, is inversely related to laryngeal cancer risk.     

Association between alcohol consumption and postmenopausal breast cancer: results of a case-control study in Montreal,Quebec, Canada

OBJECTIVES: To determine the association between postmenopausal breast cancer and prior consumption of alcoholic beverages.METHODS: This case–control study, conducted in all Montreal hospitals between 1996 and 1997, included 556 postmenopausal women (age 50–75 years) who had a new histologically confirmed diagnosis of primary, malignant breast cancer. Control subjects (577) were selected from other histologically confirmed sites of cancer. A detailed history of alcohol consumption and other risk factors was obtained by interview. Indices reflecting alcohol consumption were developed and unconditional logistic regression was used to estimate adjusted odds ratios (OR) and 95% confidence intervals (CI).RESULTS: Current regular drinkers of any type of alcohol were at an increased risk of breast cancer (OR = 1.5; 95% CI 1.0–2.2). For all beverages considered, current regular drinkers showed higher risks than ever regular drinkers. The risk of breast cancer was highest among women who reported exclusive drinking of wine on a weekly or daily basis (e.g. current regular drinking: OR = 2.3; 95% CI 1.2–4.3). Women who started to drink wine on or before the age of 40 were at a 2.5 times increased risk (95% CI 1.4–4.4).CONCLUSIONS: Our findings provide further support for a positive association between the risk of postmenopausal breast cancer and alcohol consumption.     

Obesity and the risk of prostate cancer (United States)

The role of obesity in prostate cancer etiology remains controversial. A recent report suggested that obese men younger than age 60 may have a lower risk of developing prostate cancer than men the same age who are not obese. The current study used a nested, matched case–control study design and data collected in the General Practice Research Database between January 1991 and December 2001 to assess the association between body mass index (BMI) and the risk of incident prostate cancer. Seven hundred and thirty cases of prostate cancer with adequate information on BMI were identified and matched to 2740 controls on age, sex, general practice, and index date. Obese men (BMI ≥ 30.0 kilograms [kg]/square of height in meters [m²]) were at lower risk of developing prostate cancer (AOR=0.78, 95% CI: 0.56, 1.09) compared to normal weight men (BMI=23.0–24.9 kg/m²), and the data best fit an inverse quadratic model for the relation between BMI and the risk of prostate cancer. This study provides modest support for a protective association between obesity and the risk of incident prostate cancer.     

Diet and benign ovarian tumors (United States)

Objectives: The relation between benign ovarian tumors (BOTs) and nutrients, primarily dietary fat, was examined using case–control data. Methods: 746 cases were diagnosed from 1 January 1992 to 31 December 1993. The 404 age- and hospital frequency-matched community controls were identified by random digit dialing. Six hundred seventy-three cases and 351 controls provided dietary information. Results: The risk of BOTs was elevated for the highest vs. lowest quartile of intake of total, vegetable, saturated, monounsaturated, and polyunsaturated fat. The corresponding age-, hospital-, total energy-, and body mass index-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) are 1.3 (0.9–1.9), 1.7 (1.2–2.5), 1.2 (0.8–1.8), 1.3 (0.9–1.8), and 1.6 (1.1–2.3). After adjustment for polyunsaturated fat, the risk of BOTs only remained elevated for vegetable fat (highest vs. lowest quartile OR and 95% CI = 1.4 (0.8–2.3)). Elevated risks were observed for higher intakes of polyunsaturated fat with endometrioid, serous, and teratoma tumors. Higher intakes of vegetable fat, adjusted for polyunsaturated fat, increased the risk of endometrioid, mucinous, and serous tumors. Only the risk of serous BOTs was consistently lower for higher intakes of micronutrients, with the strongest reduction observed for sources of vitamin A. Estimates were not confounded by non-nutrient covariates. Conclusions: Polyunsaturated and vegetable fat may increase the risk of BOTs, while vitamin A may lower the risk of serous BOTs; however, these findings and lack of associations for other nutrients should be replicated.     

Induced and spontaneous abortion in relation to risk of breast cancer (United States)

The relation of induced and spontaneous abortion to the risk of breast cancer is evaluated in a hospital-based case-control interview study conducted in three cities in the United States from 1985 through 1995. Cases were 1,803 women aged 25 to 64 years with newly diagnosed invasive breast cancer; controls were 4,182 women of the same ages admitted for conditions unrelated to reproductive factors. Other breast cancer risk-factors were controlled through multiple logistic regression. The reference for allanalyses was women who had never had an abortion, either induced or spontaneous. Among parous women, the relative risk (RR) estimate was 1.1 (95percent confidence interval [CI] = 0.9-1.5) for induced abortion overall, 1.0(CI = 0.7-1.4) for abortion before the first birth, and 1.3 (CI = 1.0-1.8)for abortion after at least one birth. Among nulliparous women, the relative risk estimate for induced abortion was 1.3 (CI = 0.9-1.9). There was no trend of increased risk with number of abortions, nor was there consistent evidence of an increased risk in any particular subgroup. Spontaneous abortion was not associated with increased risk of breast cancer, either among nulliparous women or among parous women. These findings provide little support for the hypothesis that induced abortion increases breast cancer risk overall or in particular subgroups. This revised version was published online in July 2006 with corrections to the Cover Date.