Lactation and breast cancer. Evidence for a negative association in premenopausal women.

The authors report a case-control study of breast cancer based on an analysis of data collected by interview between the years 1957-1965 from women residing in the communities of Buffalo and Kenmore, New York. Prior reproductive factors, including a detailed lactation history, were examined for 453 white females with breast cancer and 1,365 white females without breast cancer who were selected randomly from the population of Buffalo and Kenmore. There is evidence of a negative association between length of nursing and breast cancer risk in premenopausal women which is not seen in the postmenopausal women. This apparent "protective" effect of lactation persists after statistical control for the potential confounding factors of age, parity, age at first pregnancy, age at menarche, and education. Cases are more likely than controls to have reported unsuccessful lactations due to "insufficient milk." The findings of this study, in conjunction with the authors' review of previously published studies that have examined prior lactation as related to breast cancer risk, suggest that there may be an independent negative association between nursing and subsequent breast cancer risk in premenopausal women. Whether this means that breast feeding is protective or that some women who are unsuccessful at lactation are at increased risk for subsequent breast cancer is not clear. A detailed reanalysis of existing data and more careful attention to detailed prior breast function/dysfunction in future studies are recommended.

Predilection of lung cancer for the upper lobes: an epidemiologic inquiry.

Data from 24,798 cases of primary neoplasm of the lung registered by the Surveillance, Epidemiology, and End Results Program between 1973 and 1977 were examined for the demographic determinants of lobe of origin. Lung cancer occurred predominantly in the upper lobes in both sexes, in both blacks and whites, and at all ages. This pattern was observed for squamous cell and small cell carcinoma as well as adenocarcinoma. The predominance of lung cancer in the upper lobes was much stronger in the young and somewhat stronger in blacks. These findings are discussed in light of current theories of pulmonary carcinogenesis. Current theories may not explain entirely these observed patterns, and further inquiries into this question, therefore, may be useful.

Diet in the epidemiology of carcinoma of the prostate gland.

In vivo, in vitro, prospective, and retrospective epidemiologic inquiries have suggested that retinoids inhibit cancer, and fats have been hypothesized to enhance and ascorbic acid to reduce cancer risk. Comparison of 260 patients from Buffalo with cancer of the prostate gland was made with two different control series of similar size and age distribution. Regardless of the control group, risk of prostate cancer gained with increases in ingestion of retinoids, animal fats, and vitamin C. These anomalous findings may be due to peculiarities in methodology. From the possible specificity of effect of the nutrients studied, as shown in experimental animals and in vitro, a hypothesis could be made that a substance like vitamin A or C, which may inhibit certain cancers, also may enhance risk of other cancer types or have neither effect.

Dietary history from the distant past: a methodological study.

One of the major concerns regarding case-control studies of diet and cancer is that dietary measures based on current habits may not accurately reflect dietary intake from the time period of cancer initiation and early promotion in the more distant past. Thus, the extent to which current diet correlates with past diet and the reliability of retrospective estimates of past diet are important questions for researchers investigating dietary factors in cancer causation. For these reasons, we conducted this study of the reliability of dietary history as recalled from the distant past. Individuals (N = 175) who completed dietary interviews between the years 1957 and 1965 were reinterviewed in 1982. Subjects were asked to report the usual frequency of intake of selected food items, both at the time of their original interview in the 1957-1965 era as well as at the current time. Dietary histories as recalled from the distant past more closely agreed with those originally recorded than did current diets. The diets as recalled from the distant past appeared to be biased, however, by current dietary habits. The implications of these findings for diet and cancer research are discussed.

Diet in the epidemiology of breast cancer.

A variety of studies have shown that diets high in fat, particularly polyunsaturated, have enhanced the production of tumors in animals challenged with chemical carcinogens. Other studies have found an apparent contradiction of no difference in the incidence of breast cancer among women with varying levels of serum cholesterol as measured decades earlier. The present study concerns 2024 breast cancer cases and 1463 control patients without neoplasms or pathology of the reproductive and digestive organs, seen at Roswell Park Memorial Institute from 1958 to 1965. Based upon the assessments of their varying ingestion of fats from their own reports of diets, no difference in risk was found. Similarly, there was no difference in risk of breast cancer associated with ingesting diets containing various levels of either vitamin C or the cruciferous vegetables. Risk for breast cancer in women 55 years of age and older increased somewhat with decreases in ingestion of foods containing vitamin A.

Dietary factors in the epidemiology of cancer of the larynx.

Interviews were carried out with 374 male patients at Roswell Park Memorial Institute with cancer of the larynx and 381 controls with diseases other than of the digestive or respiratory system and other than neoplasms. As have other investigators, the authors found a high risk associated with smoking as well as a risk associated with alcohol ingestion. In addition, the study found that males ingesting low amounts of vitamin A in their diet had approximately twice the risk of those ingesting large amounts. The same was true for vitamin C. All of these relationships were significant, and risk increased in dose-response fashion with decreases in ingestion of the above-mentioned dietary constituents. When the risks associated with vitamins A and C were examined controlling for smoking and drinking, the risk enhancement associated with low ingestion persisted. The findings on nutrients are consistent with results of studies on animals and tissue cultures. Nevertheless, the sparse epidemiologic replication, the difficulties with the methods, and the complexity of the field demand substantial further inquiry on these relationships.

Spouse-subject interviews and the reliability of diet studies.

Recently published data suggest relationships between ingestion of a number of food items and risk of cancer of the stomach, bowel and mouth. This engenders concern over the accuracy of such data. To study this, dietary interviews of 158 males in the Western New York Study of Cancer Epidemiology are compared to their spouses' estimates of the husband's dietary histories as taken in separate interviews. Respondents were asked to estimate the frequency with which the males consumed each of several foods. For analysis, the frequency estimates were assembled into categories similar to those used in recent epidemiologic analysis: 5-7 times per week, 1-4 times per week, 1-3 times per month and less than once per month. Generally, 60-80% of respondents pairs agree exactly on the frequency of consumption for individual food items. Over 90% of spouse and respondent food frequency estimates are within one frequency category of each other. Radical disagreements between spouses are rare; generally, in less than 2% of spouse pairs does one member estimate the maximum frequency category, and the other minimum frequency category. The authors suggest therefore that, although diet histories are not precisely replicable, they must be adequate to reveal gross differences between cases and controls.

Preliminary report on radiation and heart disease.

Using data from the Tri-State and a mathematical model, the hypothesis that low levels of diagnostic radiation can produce both leukemia and heart disease has been tested. The range of radiation considered is within a factor of 10 of 5 rads (0.5-50 rads). Even at the lowest levels, an estimated 5% of the persons exposed suffer damage to the DNA which is subsequently expressed as heart disease or leukemia. The risk of heart disease in an affected group is roughly tripled and there is a 10-fold increase in the risk of leukemia according to preliminary estimates.

The changing histopathology of lung cancer: a review of 1682 cases.

We have reviewed the histopathology of lung cancer patients seen over the past 13 years at RPMI. Assessment of this data indicates that adenocarcinoma is becoming progressively more prevalent as related to the other forms of lung cancer. Factors which in part may account for this increased prevalence are: 1) changes in criteria for reading histopathology of lung cancer, particularly since 1967; 2) the increased incidence of lung cancer among the female population who have a propensity for adenocarcinoma; and 3) occupational and environmental factors. In 1974 adenocarcinoma for the first time became the most prevalent type of lung cancer at RPMI. Whatever the reason, if our data are truly representative of a national trend, adenocarcinoma will soon become the most prevalent type of lung cancer in the United States. This fact may result in an increasing death rate since the present 18-month survival rate for adenocarcinoma is substantially less than for squamous cell carcinoma, which has in the past been the prevalent form of the disease. As the smoking habits of women more closely approximate those of men, we expect that the incidence and mortality of lung cancer will prove to be quite similar in both sexes.