Richard B. Everson

Aspirin use and lung, colon, and breast cancer incidence in a prospective study.

A large body of experimental data and several recent epidemiologic studies indicate that aspirin use may decrease cancer risk. The experimental studies found effects at many anatomic sites, whereas the epidemiologic studies saw the greatest effect on mortality from digestive cancers. To provide further human data, we examined the association between aspirin use and cancer risk using data from the National Health and Nutrition Examination Survey I (NHANES I) and the NHANES I Epidemiologic Follow-up Studies (NHEFS). Characterization of aspirin use was based on questions in the baseline interview asking whether subjects used aspirin during the previous 30 days. Data were available from 12,668 subjects age 25–74, at time of initial examination for NHANES I, who were followed for an average of 12.4 years. Among these subjects, 1,257 were diagnosed with cancer more than 2 years after their NHANES I examination. Incidence of several cancers was lower among persons who reported aspirin use: the incidence rate ratios (and 95% confidence intervals) for all sites combined were 0.83 (0.74–0.93), lung cancer 0.68 (0.49–0.94), breast cancer in women 0.70 (0.50–0.96), and colorectal cancer in younger men 0.35 (0.17–0.73). These findings were not readily explained by potentially confounding factors. The data suggest an association between aspirin consumption and decreased cancer incidence at several cancer sites.

Six families prone to ovarian cancer

Six families are presented with multiple cases of ovarian cancer, mainly serous cystadenocarcinoma. Three families had concomitant aggregation of breast cancer, suggesting genetic determinants common to both tumors. The exceptional cancer risk in these families prompted prophylactic oophorectomy in 14 asymptomatic women from four families. Review of the original microscopic sections from 8 women revealed that 3, representing two families, had abnormalities of ovarian surface epithelium and mesothelial tissue, which may be of etiologic significance and portend neoplastic changes. To enable early detection and prevention of ovarian cancer, new diagnostic techniques and etiologic studies should be applied whenever possible to high‐risk families.

The effect of chemotherapy on the in vivo frequency of glycophorin A ‘null’ variant erythrocytes

A human in vivo somatic cell assay based on the enumeration of variant erythrocytes lacking expression of an allelic form of the cell-surface sialoglycoprotein, glycophorin A, was applied to the study of blood samples from patients obtained prior to, during, and following chemotherapy for malignant disease in order to determine the effect of mutagenic chemical agents on the frequency of variant cells. In 22 patients assayed prior to therapy, the mean variant cell frequency was 11.9 per million, which was not significantly different from that observed in healthy controls. In an initial cross-sectional survey, blood samples were obtained at various times during and after therapy from 30 patients diagnosed with a variety of malignancies who were treated with one or more known mutagenic agents including adriamycin, bleomycin, cis-platinum, cyclophosphamide, dacarbazine, etoposide, lomustine, mechlorethamine, melphalan, mitomycin C, and procarbazine. Significant elevations in the mean frequency of variant cells over pre-therapy and normal levels were observed in samples obtained during and after therapy. In a time-series study, 14 breast cancer patients treated with CAF (cyclophosphamide, adriamycin, 5-fluorouracil), CMF (cyclophosphamide, methotrexate, 5-fluorouracil), or VMF (vinblastine, methotrexate, 5-fluorouracil) adjuvant chemotherapy were sampled repeatedly during and after therapy. For the CAF and CMF patients an increase in the frequency of variant cells was observed with a lag in the appearance of induced variants after initiation of therapy; variant frequencies gradually increased during therapy reaching a maximum at or shortly after the end of therapy, then declined to near pre-therapy levels within 6 months. The maximum level of induced variants ranged from 2- to 7-fold over pre-therapy or normal levels depending on the combination of agents used. The breast cancer patients treated with both adriamycin and cyclophosphamide showed consistent elevations in the frequency of variant cells; patients treated only with cyclophosphamide showed lower and more variable elevations. The data demonstrate that mutagenic chemotherapy agents induce elevated levels of glycophorin A variant erythrocytes consistent with the hypothesis that variant cells result from somatic mutation. The elevations in variant cells were transient, suggesting that these agents primarily affect the rapidly cycling committed erythroid cell population.

CANCER OF THE UTERINE CORPUS AFTER HORMONAL TREATMENT FOR BREAST CANCER

Among 45853 women in whom breast cancer was diagnosed after age forty-nine, from the series of the End Results Program of the National Cancer Institute, cancer of the uterine corpus subsequently developed in 203. The risk was greater among those women receiving hormones than in other treatment groups, and tended to rise with increasing interval from first treatment. One method of estimating an expected value indicated that the excess risk of corpus cancer in breast-cancer patients was restricted to those treated with hormones. Given the time period under study, it may be assumed that the hormones were primarily non-steroidal oestrogens.

Familial glioblastoma with hepatic focal nodular hyperplasia

Two siblings succumbed to glioblastoma multiforme, associated with hepatic focal nodular hyperplasia and cafe‐au‐lait spots. One sibling had syndactyly and multiple colonic polyps, while the other had an angiomatous malformation in the brain. Their mother died of myasthenia gravis. The findings suggest a hamartomatous syndrome of malformations that predisposes to brain tumors.

Increased hair cadmium in newborns of women occupationally exposed to heavy metals

Newborn and maternal hair samples were obtained from subjects occupationally exposed to heavy metals and from matched controls. The geometric means of levels of cadmium and lead in hair from exposed mothers and of cadmium in hair from transplacentally exposed newborns were twice as high as levels present in samples from controls. There was a positive correlation between levels of cadmium in maternal and newborns hair, but no such correlation for lead. Despite statistically significant evidence of increased exposure to cadmium, no adverse health effects were documented in the small group of exposed newborns included in this study. Problems associated with exogenous contamination of hair by heavy metals and potential advantages of hair sampling for measuring fetal exposures to heavy metals are discussed.

Effects of 5-bromo-2-deoxyuridine concentration and α-naphthoflavone on the association between smoking and the frequency of sister-chromatid exchanges in lymphocytes from maternal and cord blood

The frequency of sister-chromatid exchanges was analyzed in maternal and cord blood lymphocytes obtained at delivery from 23 nonsmokers and 21 smokers. Lymphocytes were cultured under 3 conditions: in the presence of 100 microM 5-bromo-2-deoxyuridine (BUdR), 20 microM BUdR and 20 microM BUdR with 40 microM alpha-naphthoflavone (ANF). Under all assay conditions, frequencies of SCEs were consistently higher for maternal lymphocytes than for cord lymphocytes. There was no association between SCE values for cultures of the same blood specimen with 100 microM BUdR and 20 microM BUdR. When cultured with 100 microM BUdR, maternal lymphocytes from smokers had a mean SCE frequency of 13.5, which was significantly higher than the value of 11.1 observed for nonsmokers (p = 0.001 by the Wilcoxon rank sum test). Maternal smoking had no significant effect on overall frequencies of SCEs in maternal blood cultured with 20 microM BUdR either with or without ANF or when the differential between cells cultured with and without ANF was considered. Use of caffeinated beverages was associated with increased SCE values for maternal lymphocytes cultured with 20 microM BUdR (Tau beta = 0.36, p = 0.02 for the Kendalls Rank Correlation), but no such association was seen with 100 microM BUdR. For cord blood lymphocytes, however, neither smoking nor caffeine use were associated with SCE values obtained by any of the assay conditions used. The findings suggest that results of human monitoring studies using SCEs could differ depending on the concentration of BUdR used in cultures.

Analysis of human urine for mutagens associated with carcinoma of the bilharzial bladder by the Ames Salmonella plate assay: Interpretation employing quantitation of viable lawn bacteria

The Ames Salmonella plate assay was employed to test urine samples from bladder cancer patients and controls living in Egypt for the presence of chemical mutagens. Urine from five groups of Egyptian adults were tested, including individuals with (1) neither bilharziasis nor bladder cancer, (2) urinary bilharziasis and normal urinary cytology, (3) urinary bilharziasis and atypical urinary cytology, (4) carcinoma of the bilharzial bladder, and (5) bladder cancer without bilharziasis. Plates treated with histidine dependent bacteria, S‐9 mix, beta‐glucuronidases and 0.3 ml sterile urine from all five groups yielded 50 to 150 percent more colonies than plates treated with saline instead of urine. These differences were highly statistically significant for all groups except subjects with bladder cancer without bilharziasis. Gross and microscopic inspection suggested, however, that plates containing urine had heavier bacterial lawns than plates treated with saline. A procedure for quantitating viable bacteria in the lawn was devised which demonstrated that the increase in colonies on urine treated plates could be attributed to increased numbers of viable bacteria in the bacterial lawn on those plates. There was, therefore, no convincing evidence for the presence of mutagenic substances in these urine samples.

Quantification of aryl hydrocarbon hydroxylase and 7-ethoxycoumarin O-deethylase activity in human placentae: Development of a protocol suitable for studying effects of environmental exposures on human metabolism

Several issues regarding measurement of placental AHH and 7ECD activity were studied, and standardized procedures that appeared more suitable than previous assay procedures for measurement of MFO induction in epidemiological studies were adopted. In the AHH assay, deletion of the rat-liver supernatant eliminated a possible extraneous contribution to measurement of low levels of AHH activity and did not substantially affect measurement of higher levels of activity. Increasing the protein concentration of placentae homogenate from 2 to 6 mg, and the length of the incubation time from 20 to 60 min, allowed for accumulation of more BaP products, potentially maximizing the detection of low levels of AHH activity. Use of tissue homogenates made the procedure more convenient and did not appear to interfere with interindividual comparisons of activity. Assay of homogenates of fresh and frozen tissue from the same placenta gave similar results, so that frozen tissue was adopted for convenience and replicability. Although a potential problem for specimens with high AHH activity, degradation of product(s) was modest in AHH assays of human placenta. The efficiency of extraction of fluorescent products declined with increasing protein concentrations in the reaction mixture of AHH assays, but it was stable for a range of product concentrations, and could be controlled by the use of a constant amount of protein per assay. Recovery of product for the 7ECD assay was more complete and was not affected by protein concentration. Additionally, the 7ECD activity was easily detected in every placenta, regardless of smoking exposure. However, in this study the AHH assay appeared to be better at discriminating between non-smokers and smokers. These observations, and the potential differences in the spectrum of agents causing induction of mixed-function oxidases, suggest that both assays are potentially useful measures of human MFO induction in clinical or epidemiological studies.

Use of phytohemagglutinin response to determine the extent that somatic cell mutation accounts for 6-thioguanine resistance of human blood mononuclear cells

We investigated the likelihood that human peripheral blood mononuclear cells which incorporate tritiated thymidine in the presence of 6-thioguanine (6-TG) are mutants by comparing findings for cells from the same blood specimen cultured in the presence or absence of phytohemagglutinin (PHA). After 42 h in culture, autoradiography revealed that about 0.1% of cells from healthy control donors and 1% of cells from patients receiving chemotherapy for breast cancer labeled in the absence of PHA. About 5% of this number labeled when 6-TG was added to the medium. This high frequency of cells labeling in the presence of 6-TG suggested that many cells were resistant to 6-TG for reasons other than mutation at the HGPRT locus. Despite a labeling index of about 30% in the presence of PHA alone, the addition of PHA to cells cultured with 6-TG did not increase the labeling index over that observed with 6-TG alone. Also, in the presence of 6-TG there was a significant correlation between labeling indices for cells from the same blood specimen cultured with or without PHA. These findings suggest that the two conditions measured the same population of non-mutant cells. These studies reinforce observations made by the developers of this technique and other laboratories which indicate that refinements will be required before the assay can be used to study somatic cell mutation in human populations. Such refinements are underway in several laboratories; although not definitive, the approach and procedures used in this report which investigated the original assay procedure provide methods and criteria to judge whether modifications of assay techniques improve the quantification of in vivo human somatic cell mutation.