Determinants of breast cancer detection among Wisconsin (United States) women, 1988–90

Early detection is advocated widely as the best method to reduce the high rate of breast cancer mortality in women. The purpose of this study was to describe the detection histories of women with breast cancer and to identify factors related to the method of detection. During the period 1988–90, 3,197 women with invasive breast cancer, identified through the Wisconsin (United States) tumor registry, were interviewed. The method of cancer detection (classified as self, screening mammography, or clinical breast examination [CBE]) was analyzed using polychotomous logistic regression. Fifty-five percent (1,754/3,197) of the women found their own cancers, while 35 percent (1,122/3,197) were detected by screening mammography. Compared with self-detection, the likelihood of non-localized disease was significantly lower for tumors detected by mammography (odds ratio [OR]=0.3, 95 percent confidence interval [CI]=0.2–0.4) and CBE (OR=0.6, CI=0.4–0.7). The likelihood of cancer being detected by screening mammography increased with increasing age, education, number of prior mammograms, family history, and body mass index (weight/height²) (BMI). Women in the highest BMI quintile were 2.3 times (CI=1.7–3.0) more likely than women in the lowest BMI quintile to have their cancers diagnosed by mammography. This association most likely results from breast tumors being more difficult to palpate in heavier women.This work was undertaken while Dr Reeves was an Epidemic Intelligence Officer, assigned to the Wisconsin Division of Health from the Division of Field Epidemiology, Centers for Disease Control and Prevention, Atlanta, GA, USA.     

Strenuous physical activity in young adulthood and risk of breast cancer (United States)

The epidemiologic data on the relation between strenuous physical activity and breast cancer are limited and inconsistent. Because risk of breast cancer may be influenced by ovarian function which, in turn, is modulated by physical activity, the hypothesis that exercise may be associated with a reduced risk of breast cancer merits further investigation. We, therefore, conducted a large case-control study in 1988–91, and interviewed 6,888 women (17 to 74 years of age) with breast cancer in Maine, Massachusetts, New Hampshire, and Wisconsin (United States). Interviewed controls (9,539 women, 18 to 74 years of age) were selected randomly from lists of licensed drivers (for younger women) or from a roster of Medicare enrollees (for older women). We used multivariate adjusted odds ratios (OR) and 95 percent confidence intervals (CI) from logistic regression models to estimate relative risks between self-reported physical activity when 14 to 22 years of age and breast cancer. When compared with sedentary controls, women who reported any strenuous physical activity during ages 14 to 22 years had a modest reduction in the risk of breast cancer (OR=0.95, CI=0.93–0.97). However, those who exercised vigorously at least once a day had a 50 percent reduction in risk of breast cancer (OR=0.5, CI=0.4–0.7). These data support the hypothesis that women who are physically active have a reduced risk of breast cancer.This project is funded by grants (R01 CA 47147 and R01 CA 47305) from the US Public Health Service.Dr Mittendorf was also supported by National Research Service Award No. 5 T32 ES07069.     

Ionizing radiation and breast cancer in men (United States)

The purposes of this study were to determine whether exposure of the vestigial male breast to ionizing radiation is associated with an increase in risk of breast cancer and, if so, to determine whether the apparent effects on risk in men are similar to those reported for women. A population-based case-control study of breast cancer in men was conducted in 10 geographic areas of the United States. Information on possible prior exposure to ionizing radiation, and on other potential risk factors for breast cancer, was obtained from personal interviews of 227 cases and 300 controls who were recruited from October 1983 to September 1986. Evidence from this study that ionizing radiation can cause breast cancer in men includes: a modest trend of increasing risk with frequency of chest X-rays; an increase in risk in men with three or more radiographic examinations, especially if received prior to 1963; and an increase in risk in men who received X-ray treatments to the chest and adjacent body areas. Risk was increased only from 20 to 35 years after initial exposure from either radiographic examinations or X-ray treatments, and declined after three to four decades since last exposure, suggesting a wave of increased risk of finite duration following exposure. The doses of radiation received could not be estimated precisely, but those from diagnostic procedures were likely similar to those received by prepubertal females in prior studies, and the results of those and the present investigation are compatible. The carcinogenic effects of ionizing radiation may be similar in the male and prepubertal female breast.This study was funded by grant No. R01 CA 35653 from the US National Cancer Institute.     

A population-based study of contralateral breast cancer following a first primary breast cancer (Washington, United States)

To evaluate predictors of contralateral breast cancer risk, we examined data from a nested case-control study of second primary cancers among a cohort of women in western Washington (United States) diagnosed with breast cancer during 1978 through 1990 and identified through a population-based cancer registry. Cases included all women in the cohort who subsequently developed contralateral breast cancer at least six months after the initial diagnosis, but prior to 1992 (n=234). Controls were sampled randomly from the cohort, matched to cases on age, stage, and year of initial breast cancer diagnosis. Information on potential risk factors for second primary cancer was obtained through medical record abstractions and physician questionnaires. Women who were postmenopausal due to a bilateral oophorectomy (i.e., a surgical menopause) at initial breast cancer diagnosis had a reduction in contralateral breast cancer risk compared with premenopausal women (matched odds ratio [mOR]=0.25, 95 percent confidence interval [CI]=0.09–0.68), whereas no reduction in risk was noted among postmenopausal women who had had a natural menopause (mOR=0.90, CI=0.39–2.09). Among postmenopausal women, there was a suggestion of a lower risk associated with relatively high parity (2+). A family history of breast cancer was associated with an increased risk (mOR=1.96, CI=1.22–5.15) and varied little by menopausal status. Having an initial tumor with a lobular component (c.f. a ductal histology) was not related strongly to risk (mOR=1.47, CI=0.79–2.74). The results of the present and earlier studies argue that we have limited ability to predict the occurrence of a contralateral breast tumor. Better predictors will be required before diagnostic and preventive interventions can be targeted to subgroups of patients with unilateral breast cancer.Authors are with the Department of Epidemiology, University of Washington, Seattle, WA, USA (Drs Cook, White, Schwartz, Daling, Weiss); with the Fred Hutchinson Cancer Research Center, Seattle, WA (Drs Cook, White, Schwartz, McKnight, Daling, Weiss); and the Department of Biostatistics, University of Washington, Seattle, WA (Dr McKnight). Address correspondence to Dr Cook, MP-381, Fred Hutchinson Cancer Research Center, 1124 Columbia Street, Seattle, WA 98104, USA. This research was supported in part by grants from the US National Cancer Institute (R35 CA 39779), the Agency for Health Care Policy and Research (1 RO3 HS08004-01), and by the Cancer Surveillance System of the Fred Hutchinson Cancer Research Center, which is funded by Contract No. N01-CN-05230 from the Surveillance, Epidemiology and End Results (SEER) program of the National Cancer Institute with additional support from the Fred Hutchinson Cancer Research Center.     

Early body size and subsequent weight again as predictors of breast cancer incidence (Iowa,United States)

We examined whether associations of adult weight gain with the risk of postmenopausal breast cancer vary by stature, waist-hip ratio (WHR), and early adult size in a cohort of 37,105 Iowa (United States) women. Both low body mass index (kg/m²) (BMI) at age 18 and high subsequent weight-gain were associated independently with increased risk of incident postmenopausal breast cancer. After stratifying on BMI at age 18, high weight gain was associated with increased risk irrespective of whether early BMI was low (relative risk [RR]=1.92, 95 percent confidence interval [CI]=1.45–2.53) or high (RR=1.59, Ci=1.19–2.12). Women with lower BMI at 18 were at a higher risk at all levels of weight change, but having low BMI at age 18 and low subsequent weight gain conferred no significantly excess risk over those with high BMI at 18 and low gain. An inconsistent increase in risk was associated with taller stature; there was no additional risk associated with high WHR. Part of the observed risk from lower early size may reflect greater weight gain by lighter women. Limiting adult weight gain thus may be a feasible method to avoid increasing an individual's risk of breast cancer. Reasons for different effects of early cf late weight gain are not established, but benefits of a greater size at age 18 are likely to be offset by increased risks of other weight-related diseases at older ages.This work was supported by grant R01-CA 39724 from the US National Cancer Institute. Dr Barnes-Josiah was supported by a US Public Health Service Award (5 T32 CA 09607).     

A nested case-control study of mammographic patterns,breast volume,and breast cancer (New York City,NY, United States)

The relations of Wolfe mammographic patterns, quantitative mammographic densities, and mammographically estimated breast size to breast cancer risk were investigated prospectively in a case-control study nested in the New York University Women's Health Study, a cohort of 14,291 women in New York City, NY (United States). The archived mammograms of 197 breast cancer cases who were identified during the first 5.5 years of the study and of 521 individually matched controls from the same cohort were retrieved and classified according to Wolfe parenchymal patterns and mammographic densities by two expert radiologists. Breast size and volume were estimated on the mammogram's cranio-caudal projection. In both premenopausal and postmenopausal subjects, the risk of breast cancer increased progressively with increasing density and percent density area. A significantly increased risk was found also for Wolfe pattern DY in premenopausal women and P2 pattern in postmenopausal subjects. In premenopausal women, mammographically determined breast volume and breast height also were associated positively with breast cancer risk. Although the results of the present study confirmed that mammographic parenchymal patterns and densities were important predictors of breast cancer risk, their practical use in screening seems limited due to the high prevalence of high risk patterns.This work was supported by Grants CA51921, CA34588, CA13343, and CA16087 from the US National Cancer Institute and by Grant ES00260 from the National Institute of Environmental Health Sciences.     

Recent oral contraceptive use and risk of breast cancer (United States)

We examined the association between recent oral contraceptive (OC) use and the risk of breast cancer in data from a large population-based case-control study in the United States. Cases (n=6,751) were women less than 75 years old who had breast cancer identified from statewide tumor registries in Wisconsin, Massachusetts, Maine, and New Hampshire. Controls (n=9,311) were selected randomly from lists of licensed drivers (if aged under 65 years) and from lists of Medicare beneficiaries (if aged 65 through 74 years). Information on OC use, reproductive experiences, and family and medical history was obtained by telephone interview. After adjustment for parity, age at first delivery, and other risk factors, women who had ever used OCs were at similar risk of breast cancer as never-users (relative risk [RR]=1.1, 95 percent confidence interval [CI]=10–1.2). Total duration of usealso was not related to risk. There was a suggestion that more recent use was associated with an increased risk of breast cancer; use less than two years ago was associated with an RR of 1.3 (CI=0.9–1.9). However, only among women aged 35 to 45 years at diagnosis was the increase in risk among recent users statistically significantly elevated (RR=2.0, CI=1.1–3.9). Use prior to the first pregnancy or among nulliparous women was not associated with increased risk. Among recent users of OCs, the risk associated with use was greatest among non-obese women, e.g., among women with body mass index (kg/m²) less than 20.4, RR=1.7, CI=1.1–2.8. While these results suggest that, in general, breast cancer risk is not increased substantially among women who have used OCs, they also are consistent with a slight increased risk among subgroups of recent users.Authors are with the University of Wisconsin Comprehensive Cancer Center, Madison, WI, USA (Dr Newcomb, Ms Trentham Dietz); NIEHS Epidemiology Branch, Research Triangle Park, NC (Dr Longnecker); Fred Hutchinson Cancer Research Center, Seattle, WA (Dr Surer); Department of Obstetrics and Gynecology, Pritzker School of Medicine, The University of Chicago, Chicago, IL (Dr Mittendorf); Department of Community and Family Medicine, Dartmouth Medical School, Hanover, NH (Dr Baron); Boston University, School of Public Health, Boston, MA (Dr Clapp); Department of Epidemiology and Department of Nutrition, Harvard School of Public Health, and Channing Laboratory, Harvard Medical School and Department of Medicine, Brigham and Women's Hospital, Boston, MA (Dr Willett). Address correspondence to: Dr Polly A. Newcomb, University of Wisconsin-Madison Comprehensive Cancer Center, 1300 University Ave., #4780, Madison, WI 53706, USA. Supported by Public Health Service (National Cancer Institute) grants R01 CA 47147 and R01 CA 47305.     

Leisure-time physical activity in relation to breast cancer among young women (Washington, United States)

It has been hypothesized that women who participate in vigorous physical activity may have lower risk of breast cancer due to lower lifetime exposure to ovarian hormones. A population-based case-control study was conducted to investigate the association between leisure-time physical activity and risk of breast cancer among women aged 21 to 45 years. Cases were 747 women diagnosed with invasive breast cancer between 1983 and 1990 in three counties of western Washington state (United States), and were identified through the Seattle-Puget Sound Surveillance, Epidemiology, and End Results (SEER) registry. Controls were 961 women selected from the same area by random-digit telephone dialing. Physical activity was assessed through personal interview, with questions on frequency and duration of each type of recreational activity during the two-year period immediately prior to reference date (date of diagnosis for cases and a comparable assigned date for controls) and between ages 12 and 21. For the two-year time period before diagnosis, there was no association with frequency of activity (age-adjusted odds ratio [OR]=0.93, 95 percent confidence interval [CI]=0.71-1.22 for four or more episodes per week cf none), total hours spent in physical activity (age-adjusted OR=0.92, CI=0.71-1.22 for four or more hours per week cf none) or MET (metabolic equivalent energy expenditure unit) (age-adjusted OR=0.95, CI=0.73-1.23 for 18 or more METs per week cf none), nor any trend in risk with increasing activity levels. Similarly, there was no association between leisure activity during adolescence and breast cancer risk. These results were not confounded further by body mass index (wt/ht2), age at menarche, age at first full-term pregnancy, parity, family history of breast cancer, or other measured health behaviors. Our findings do not support a protective effect of leisure-time physical activity either in the adolescent years or in adulthood on breast cancer in young women.     

Physical activity in usual occupation and risk of breast cancer (United States)

We have used data from a large population-based case-control study inthe United States to evaluate the effect of occupational physical activity onbreast cancer risk. Women diagnosed with breast cancer identified from fourstate cancer registries, and controls randomly selected from lists oflicensed drivers or Medicare beneficiaries, were interviewed by telephone forinformation on usual occupation and other factors. We classified usualoccupation into one of four categories of physical activity. After excludingsubjects for whom a strength rating could not be assigned, we had a finalsample size of 4,863 cases and 6,783 controls. Using conditional logisticregression models, we calculated adjusted odds ratios (OR) and 95 percentconfidence intervals (CI) for occupations having light, medium, and heavyactivity compared with sedentary ones. Women with heavy-activity occupationshad a lower risk of breast cancer than women with sedentary jobs (OR = 0.82,CI = 0.63-1.08), as di d women with jobs with medium activity (OR = 0.86, CI= 0.77-0.97) or light activity (OR = 0.92, CI = 0.84-1.01). There was asignificant decreasing trend in the ORs from sedentary to heavy work (P =0.007). Although limited by exposure misclassification, these data areconsistent with the hypothesis that physical activity reduces the risk ofbreast cancer.     

Age at menarche, time to regular cycling, and breast cancer (North Carolina, United States)

Objectives: Later menarche, and a longer time until onset of regular cycling, are considered markers of lower lifetime exposure to circulating ovarian hormones. While later age at menarche is associated with reduced breast cancer risk, evidence for the relationship between time until onset of regular cycles and breast cancer is inconsistent. We evaluated both associations with breast cancer risk.Methods: We used data from the Carolina Breast Cancer Study, a population-based case-control study of breast cancer among White and Black women aged 20 to 74 years, residing in central and eastern North Carolina (United States). Cases were diagnosed between May 1993 and June 1996. Unconditional logistic regression models were limited to women with complete data for the risk factors considered (n = 830 cases, 758 controls).Results: We observed an inverse relationship between age at menarche and breast cancer risk, but found little support for the hypothesis that a longer time until onset of regular menstrual cycling was associated with reduced risk of breast cancer (odds ratios=1.0 [95 percent confidence interval (CI) = 0.7-1.5], and 1.2 [CI = 0.8-1.6], respectively, for 1-4 years and < 1 year until onset of regular cycling, relative to 5+ years). There was little relationship between age at menarche and time until regular cycling. We found strong evidence that delays in onset of regular cycling were associated with increased frequency of irregular cycles throughout young adulthood.Conclusions: Given the inconsistent findings regarding the links between menstrual cycle characteristics and breast cancer, and recent recommendations to delay menarche and alter the patterns of cycles of young women in order to reduce breast cancer risk, this topic calls for further, innovative study.     

Laterality of breast cancer in the United States

Breast cancer laterality of over 250,000 cases from the Surveillance, Epidemiology, and End Results (SEER) program in the United States was studied in relation to gender, race, tumor stage, histology, age at diagnosis, year of diagnosis, estrogen receptor status, and marital status. The data, which include all invasive and in situ breast cancer cases in the SEER program during the years 1973–92, confirm results from other studies of an overall five percent excess of left-sided disease in women. The excess occurs for all races and stages of disease, and for invasive disease, the excess increases with age. There was no significant variation in the laterality of invasive disease over time, though for in situ tumors, the left-sided excess was significantly greater during the years 1978–82 than in other periods. No excess of left-sided breast cancer was observed among men. There is no evidence that detection bias plays a major role, and although the left breast is slightly larger, on average, than the right, there is little evidence that breast size is associated with breast cancer risk. The reason for the left-sided excess among women remains unclear.Authors are with the Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD, USA. Address correspondence to Dr Weiss, Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Executive Plaza North Rm 443, 6130 Executive Blvd., Bethesda, MD 20892-7374, USA.     

Interaction of family history of breast cancer and dietary antioxidants with breast cancer risk (New York,United States)

We sought to determine if specific dietary antioxidants may be particularly effective in reducing breast cancer risk for women reporting family history (FH) of breast cancer in a first-degree relative. Interviews regarding usual diet, health, and family histories were conducted with 262 premenopausal and 371 postmenopausal women with incident, primary breast cancer from western New York (United States). These women were frequencymatched by age and county of residence with community controls. Among premenopausal women, there was a significant interaction between FH and -tocopherol; -tocopherol was associated with significantly decreased risk among FH+ women (adjusted fourth-quartile odds ratio [OR]=0.01, 95 percent confidence interval [CI]=0.0–0.3). This association was much weaker for FH-women [OR=0.7, CI=0.4–1.2]. For FH-women, a significant inverse association was observed between -carotene and premenopausal breast-cancer risk (OR=0.4, CI=0.3–0.5), but not for FH+ women (OR=0.5, CI=0.1–4.0). Similar relationships, although not as strong, were noted among postmenopausal women. Although limited by small numbers, these results suggest that biologic mechanisms of tumorigenesis may differ in FH+ and FH-women, and that -tocopherol may be a potential chemopreventive agent for women with a family history of breast cancer, particularly premenopausal women.This research was conducted by the Department of Social and preventive Medicine, State University of New York at Buffalo. This publication is supported in part by grants CA11535 and 5 R25 CA1820117 from the US National Cancer Institute and PDT-434 from the American Cancer Society. Dr Freudenheim is a recipient of a Research Career Development Award from the National Cancer Institute (CA01633). This work is solely the responsibility of the authors and does not necessarily represent the views of the NCI.     

A prospective study of oral contraceptive use and risk of breast cancer (Nurses' Health Study, United States)

Results of previous epidemiologic studies have provided reassurance that there is little, if any, increase in risk of breast cancer with oral contraceptive (OC) use in general. However, in several studies, an increased risk of breast cancer has been observed in two subgroups, young women who used OCs for extended durations and in women who used OCs prior to a first-term pregnancy. We evaluated these relationships using data from the ongoing Nurses' Health Study cohort (United States). We documented 3,383 cases of breast cancer from 1976 to 1992 among 1.6 million person-years of follow-up. We observed no overall relationship between duration of OC use and breast cancer risk, even among women who reported using OCs for 10 or more years (multivariate relative risk [RR]=1.11, 95 percent confidence interval [CI]=0.94-1.32). Among women less than 45 years of age, the multivariate RR for using OCs for 10 or more years was 1.07 (CI=0.70-1.65) compared with never-users. The risk associated with five or more years of OC use prior to a first full-term pregnancy compared with never-use was 0.96 (CI=0.65-1.43). Among women less than 45 years of age, we observed no evidence of an increased risk with OC use before a first full-term pregnancy (use for five or more years: RR=0.57, CI=0.24-1.31). Because of the age distribution of our cohort, we were unable to evaluate these relationships among women less than 40 years of age. Our study provides considerable evidence that long-term past OC use, either overall or prior to a first full-term pregnancy, does not result in any appreciable increase in breast cancer risk in women over 40 years of age.     

Incomplete pregnancies and risk of ovarian cancer (Washington, United States)

Because of the reduced risk of ovarian cancer related to prior full-term pregnancies, we sought to determine whether there was any association with a history of one or more incomplete pregnancies. White female residents of three counties in Washington State (United States) diagnosed with ovarian cancer during 1986–88 (n=322), and a random sample of control women selected from these same counties (n=426), were interviewed regarding their pregnancy and childbearing histories. Among women who had given birth to at least one child, an additional incomplete pregnancy was not associated with the risk of ovarian cancer (relative risk [RR]=1.1, 95 percent confidence interval [CI]=0.8–1.6, adjusting for age, oral contraceptive use, and number of births). For those who had never given birth, a somewhat smaller proportion of cases had a history of incomplete pregnancy than controls (RR=0.8, CI=0.4–1.7). In an analysis restricted to ever-pregnant women, a prior induced or spontaneous abortion was not found to be associated with the incidence of ovarian tumors (RR=1.0, CI=0.6–1.7, and RR=1.3, CI=0.8–1.9, respectively). Other studies of the possible relation between incomplete pregnancies and ovarian cancer generally have observed either a weak negative association or no association at all. It is possible that if incomplete pregnancies do affect the risk of ovarian cancer, their impact might be too small to be identified reliably through epidemiologic studies.This research was supported in part by a grant from the US National Cancer Institute (R35 CA39779), and by the Cancer Surveillance System of the Fred Hutchinson Cancer Research Center, which is funded by Contract No. N01-CN-05230 from the Surveillance, Epidemiology and End Results (SEER) program of the National Cancer Institute with additional support from the Fred Hutchinson Cancer Research Center.     

Menopausal estrogen and estrogen-progestin replacement therapy and risk of breast cancer (United States)

This study examines the relationship between menopausal estrogen and estrogen-progestin replacement therapy and risk of breast cancer, focusing on whether associations differ according to whether the tumors arein situ or invasive. Data are from a prospective study conducted 1980–89 on 49,017 selected participants in the Breast Cancer Detection Demonstration Project, a five-year screening program conducted between 1973 and 1980 in the United States. Overall, the rate ratio for estrogen-only use compared with no-hormone use was 1.0, and that for the estrogen-progestin combination was 1.2 (95 percent confidence interval [CI]=1.0–1.6). However, the associations differed according to whether the tumors werein situ or invasive. The rate ratios ofin situ breast cancer associated with use of estrogens alone and the combination regimen were 1.4 (CI=1.0–2.0) and 2.3 (CI=1.3–3.9), respectively. Duration of estrogen-only use also was associated with risk ofin situ tumors, with users for 10 or more years at twice the risk of nonusers (P-value for trend test =0.02). Duration of use was not associated with risk of in vaisve cancer. Our results are consistent with the hypothesis that hormone replacement therapy is related to earlier-stage breast cancer; however, the possibility that the results reflect increased breast cancer surveillance among those taking hormones cannot be ruled out.     

Parental ages at birth in relation to a daughter's risk of breast cancer among female participants in the Framingham study (United States)

Data from the Framingham Heart Study, collected in Framingham, MA (United States) during 1948–86, were used to evaluate the relation of parental age at birth to the risk of breast cancer among daughters. After 38 years of follow-up, 149 breast cancer cases occurred among 2,662 women. All but two cases were confirmed by histologic report. The rate of breast cancer increased among daughters with increasing maternal age at birth up to the mid-30s, where the rate levelled off. A similar pattern was observed with paternal age. After adjustment for other confouding factors and paternal age, the rate ratios for breast cancer in daughters whose mothers were aged 26 to 31 years and 32 or more years at their birth, relative to women whose mothers were aged 25 years or younger, were 1.5 (95 percent confidence interval [CI]=1.0–2.4) and 1.3 (CI=0.8–2.2), respectively. However, there was no longer an association between paternal age at birth and risk of breast cancer after controlling for maternal age and other risk factors.Drs Zhang, Cupples, and Coulton are with the Department of Epidemiology and Biostatistics, School of Public Health, Boston University, Boston, MA, USA, Dr Rosenberg is with the Slone Epidemiology Unit, Boston University School of Medicine, Brookline, MA. Dr Kreger is with the Section of Preventive Medicine and Epidemiology, The Evans Memorial Department of Clinical Research, Boston University Medical Center, Boston, MA. Address correspondence to Dr Zhang, Department of Epidemiology and Biostatistics, School of Public Health, Boston University, 80 East Concord Street, Boston, MA, 02118-2394, USA.     

Dietary relationships with early onset (under age 45) breast cancer in a case-control study in the United States: influence of chemotherapy treatment

Methodologic investigations have addressed selection and recall biasin case-control studies of diet and breast cancer, whereas the effect ofdisease progression and medical treatment on estimates of dietary intake hasbeen largely overlooked. In a multicenter, population-based case-controlstudy of breast cancer in the United States, 1,588 newly diagnosed cases and1,451 controls completed a self-administered food-frequency questionnaire.Initial evaluation suggested increased risk related to high intakes ofcalories, carbohydrates, fat, and protein. All nutrient associations werediminished after adjustment for calories. Evaluation by stage of diseaserevealed no relation of calories to risk among women with in situ disease,but elevated risks among women with localized (odds ratio [OR] = 1.33, 95percent confidence interval [CI] = 1.0-1.7 highest cf lowest quartile) orregional and distant disease (OR = 1.79, CI = 1.3-2.4). Further evaluationshowed that the increased risk a ssociated with calories was restricted tocases who reported having been treated with chemotherapy (OR = 1.66, CI =1.3-2.1). A gradient of increasing risk with time interval from diagnosis tointerview suggested the chemotherapy regimen itself and not necessarilycharacteristics of tumors requiring this treatment was responsible for theobserved increased risk. These results indicate that epidemiologic studies ofdiet and breast cancer, particularly among young women, should evaluatepossible bias related to post-diagnosis influences.     

Spontaneous abortion and risk of fatal breast cancer in a prospective cohort of United States women

Controversy exists over the possible relationship between induced and spontaneous abortion and risk of breast cancer. Thus, the association of fatal breast cancer and spontaneous abortion was examined in a large prospective study of United States adult women. After seven years of follow-up, 1,247 cases of fatal breast cancer were observed among 579,274 women who were cancer-free at interview in 1982 and who provided complete reproductive histories. Results from Cox proportional hazards models, adjusted for other risk factors, showed no association between a history of spontaneous abortion and risk of fatal breast cancer (rate ratio [RR]=0.89, 95 percent confidence interval [CI]=0.78–1.02). The RR did not increase with increasing numbers of abortions. Parous women who had a spontaneous abortion before their first term birth were not at increased risk compared with parous women with no history of spontaneous abortion (RR=0.76, CI=0.54–1.05). Women whose only pregnancy ended in a spontaneous abortion were not at increased risk compared with women who were never pregnant (RR=0.61, CI=0.27–1.38) or whose only pregnancy ended in a livebirth (RR=0.72, CI=0.32–1.65). These findings do not support an association between spontaneous abortion and fatal breast cancer.     

Tobacco use and prostate cancer in Blacks and Whites in the United States

Prostate cancer occurs more frequently in Blacks than Whites in the United States. A population-based case-control study which investigated the association between tobacco use and prostate cancer risk was carried out among 981 pathologically confirmed cases (479 Blacks, 502 Whites) of prostate cancer, diagnosed between 1 August 1986 and 30 April 1989, and 1,315 controls (594 Blacks, 721 Whites). Study subjects, aged 40 to 79 years, resided in Atlanta (GA), Detroit (MI), and 10 counties in New Jersey, geographic areas covered by three, population-based, cancer registries. No excesses in risk for prostate cancer were seen for former cigarette smokers, in Blacks (odds ratio [OR]=1.1, 95 percent confidence interval [CI]=0.7–1.5) and in Whites (OR=1.2, CI=0.9–1.6), or for current cigarette smokers, in Blacks (OR=1.0, CI=0.7–1.4) and in Whites (OR=1.2, CI=0.8–1.7). Increases in risk were noted for smokers of 40 or more cigarettes per day, among former (OR=1.4, CI=1.0–1.5) and current (OR=1.5, CI=1.0–2.4) smokers. Duration of cigarette use and cumulative amount of cigarette use (pack-years) were not associated with prostate cancer risk for Blacks or Whites. By age, only the youngest subjects, aged 40 to 59 years, showed excess risk associated with current (OR=1.5, CI=1.0–2.3) and former (OR=1.7, CI=1.1–2.6) use of cigarettes, but there were no consistent patterns in this group according to amount or duration of smoking. Risks also were not elevated for former or current users of pipes, cigars, or chewing tobacco, but the risk associated with current snuff use was OR=5.5 (CI=1.2–26.2). This subgroup finding may have been due to chance. The results of the present study may be consistent with a small excess risk for prostate cancer associated with tobacco use, but the lack of consistent findings in population subgroups and the lack of a clear dose-response relationship argue more strongly that no causal association exists. The data do not indicate that the Black-White difference in prostate cancer risk is related to tobacco use.This research was performed under contracts: NO1-CP-51090, NO1-CN-0522, NO1-CP-51089, NO1-CN-31022, NO1-CP-51092, and NO1-CN-5227.