There has been substantial activity in the Adult Cancer Core in the past year related to environmental factors and genetic influences. Several potentially important observations have been made; as in the past some of these based so far on preliminary analyses or incomplete data sets.

Dietary biomarkers are being developed for several possible chemopreventive nutrients, including isothiocyanates, isoflavonoids and tea polyphenols. The Core has been extremely active in the past year in validating these biomarkers and beginning to utilize them in large prospective studies on the relationship between these and cancer risk.

Analysis has continued of the largest single institutional studies ever conducted on renal cell carcinoma and bladder cancer, focusing on long-term use of various formulations of analgesics. A moderately strong but internally highly consistent increase in risk across multiple formulation categories for renal cell carcinoma has been demonstrated. In contrast, no increase in risk whatsoever was apparent for bladder cancer; in fact non-steroidal anti-inflammatory drugs were associated with a reduction in risk, in a duration-related manner.

A number of Core members participated in writing a “Perspectives in Cancer Research Paper” which summarized our work on the molecular genetics of prostate cancer and proposed a polygenic model of prostate cancer etiology, focused around androgen metabolism genes, including their possible interaction with environmental (dietary) risk factors.

The Adult Cancer Core continues to be very active in studies of environmental carcinogenesis. Selected activities of note on a cancer site-specific basis are summarized below followed by a listing of some of the pertinent publications by the Core during the past year.

Analyses and manuscript preparations have continued from the large population-based case-control study of renal cell carcinoma. A manuscript is in review showing a moderately strong and very internally consistent association between regular analgesic use and renal cell carcinoma risk. Risk increased with increasing dose measured in a variety of ways and was somewhat stronger for acetaminophen formulations than for aspirin or other NSAIDs. Importantly there was no increase in risk observed with low daily doses of aspirin (as might be taken, for example, for cardiovascular chemoprevention) even after long periods of use. A parallel study of bladder cancer using the same instrument and same interviewers has been completed. Not only has no increase in bladder cancer risk associated with particular analgesic formulation been observed, there is some evidence of a reduction in risk with increasing duration of NSAIDs. The group is exploring whether this might be due to inhibition of cyclooxygenase-II (COX-II) in bladder mucosa.

Results were recently published from the renal cell carcinoma case-control study showing that previous suggestions by the team and others that diuretics are associated with renal cell carcinoma is best explained by confounding by hypertension, a strong and highly consistent risk factor in our study. We have also confirmed an unexplained observation made by several groups that hysterectomy is associated with increased risk of renal cell carcinoma. It is presumed this effect is due to damage to renal blood supply during surgery to remove the ovaries in conjunction with the hysterectomy.

The molecular epidemiologic study of bladder cancer, being concurrently conducted in Los Angeles and Shanghai, China, is near completion. Between the two sites there are 1,200 case-control pairs with completed interviews, and blood and urine samples collected. Preliminary analyses of this large data set are being conducted. GSTM1 null genotypes but not GSTT1 or GSTN1 are at increased risk of bladder cancer, and as anticipated, this effect is exclusively in smokers. Similarly NAT1 slow phenotype is associated with increased risk, again exclusively in smokers, but these preliminary analyses demonstrate no association with NAT2.

Substantial progress has been made in identifying risk and protective factors for adenomatous polyps, precursor lesions for colon adenocarcinomas. A series of papers were published in the past year (1) demonstrating the complex interplay among folate intake, alcohol consumption and smoking on risk of polyps; (2) the interrelationship between NAT1 "fast" genotypes and GSTM1 null genotypes in reducing risk of polyps, according to specific dietary habits; and (3) a clinical trial showing that calcium supplementation can greatly reduce risk of polyps.

An authoritative Perspectives paper in Cancer Research on the prospects of developing a polygenic model of prostate cancer focused around androgen metabolizing genes has been published, and described the progress to date in building such a model. This model includes the prospect for expanding it to incorporate environmental (especially) dietary risk factors. New funding has been received in several areas of prostate cancer, including utilizing the model we developed for evaluating short term efficacy of chemopreventive agents to test selenium, and a multiracial ethnic study to look at the interaction of vitamin D receptor genotype and vitamin D biomarkers in relationship to advanced prostate cancer risk.

The Core continues to evaluate the possible role of estrogen metabolism genes in breast cancer development. One of the important observations in the past year is that a common polymorphism variant of the CYP17 gene which controls rate limiting steps in estradiol biosynthesis and which we had previously shown to be related to both circulating estrogen levels and age at menarche is also strongly correlated with use of hormone replacement therapy (HRT). A manuscript is in preparation on these observations and there are plans to explore the risk of this gene as a confounder or risk modifier on the relationship between HRT and both breast and endometrial cancers.

Detailed studies of physical activity, ovulatory and hormonal patterns, and breast cancer risk are being continued. In particular we have (in press) the first detailed study of the relationship between physical exercise and breast cancer risk in postmenopausal women. This study demonstrates than to maintain the benefit on breast cancer due to physical exercise at younger ages, one must not only continue to exercise but also maintain body weight at a reasonable level.

The team has been active participants in collaborative studies investigating such wide-ranging environmental health problems as parental occupation, farm and farm animal exposure, electromagnetic fields, and pesticide exposure on risk of childhood brain tumors; medical radiation and prescription drug use on risk of acute myelogenous leukemia, and has produced several detailed publications on possible protection from childhood brain tumors by prenatal vitamin supplementation.

Programmatically, the team has been very involved in the last five years in developing and validating various types of biomarkers as accurate predictors of certain types of environmental exposures. In the past year there has been particular involvement in three categories of potential dietary chemopreventive agents—isothiocyanates, soy, and tea polyphenols. Several papers in the past year were published focused around the two Chinese cohorts, demonstrating the relationship between urinary levels of thiol conjugates of isothiocyanates and detailed dietary intake and how these levels might be modifiable by certain phase I and phase II enzymes. We have demonstrated how spot urinary sample levels of isoflavonoids relate to dietary soy intake. Finally, preliminary results in the Shanghai cohort study suggest that low levels of urinary isothiocyanates might modestly prevent lung cancer, but that urinary polyphenols do not predict stomach cancer risk whatsoever.