Of the many novel new cancer therapeutic concepts under development, chemoprevention recognizes that malignancies derive from a long, complex interaction of environmental stress modulated by individual genetic phenotypic expression. As described in depth by Drs. Singh and Lippman in this two-part article, published in last and this month’s issues of oncology, substances with potential chemopreventive activity have been identified from multiple sources. These include: (1) human cancer epidemiology, with an emphasis on dietary assessment, geographic dietary and environmental variation, and differences in cancer incidence among similar regional populations; (2) from mechanistic hypotheses; and (3) clinical observations after treatment of cancer (eg, tamoxifen [Nolvadex] for breast cancer). Drs. Singh and Lippman ably demon-strate the wide variety of sources of potential chemopreventive agents and describe current research studies and outcomes.
Of the many novel new cancer therapeutic concepts under development, chemoprevention recognizes that malignancies derive from a long, complex interaction of environmental stress modulated by individual genetic phenotypic expression. As described in depth by Drs. Singh and Lippman in this two-part article, published in last and this months issues of oncology, substances with potential chemopreventive activity have been identified from multiple sources. These include: (1) human cancer epidemiology, with an emphasis on dietary assessment, geographic dietary and environmental variation, and differences in cancer incidence among similar regional populations; (2) from mechanistic hypotheses; and (3) clinical observations after treatment of cancer (eg, tamoxifen [Nolvadex] for breast cancer). Drs. Singh and Lippman ably demon-strate the wide variety of sources of potential chemopreventive agents and describe current research studies and outcomes.
A Mixed Bag
To say the least, the outcomes of large, risk-reduction chemoprevention trials are a mixed bag of success and failure. For example, tamoxifen prevents breast cancer development in high-risk women. However, beta-carotene does not prevent the development of lung cancer in smokers and, at the doses used in the CARET trial, enhances the development of lung cancer. These data and those from other trials of chemopreventive agents highlight the need for careful, systematic research that starts with in vitro models of carcinogenesis, proceeds to studies in animal carcinogenesis models, and ends with clinical trials in humans.
The agents reviewed in this second part of the review by Singh and Lippmanhormones, nonclassic antioxidant natural agents, nonsteroidal anti-inflammatory agents (NSAIDs), difluoromethylornithine (DFMO), and phase II metabolic enzyme inducers (oltipraz, N-acetylcysteine)represent an excellent cross- section of the chemopreventive agents that are currently the subjects of intense research at the bench and the bedside. Among the agents reviewed in part 2 of this series, the hormonal agents and NSAIDs deserve special mention.
The recent publication of the large National Surgical Adjuvant Breast and Bowel Project (NSABP) risk-reduction trial, sponsored by the National Cancer Institute (NCI), represents an early success of a chemopreventive clinical intervention. However, the decision to intervene in healthy women whose risk of breast cancer development is related to age is a difficult one for clinicians.
Tamoxifen, as with any drug, has important, uncommon toxicities. Does the small, but real risk of development of endometrial cancer or thrombotic events represent an unacceptable risk for healthy people? Should all women over 60 years of age take tamoxifen to reduce their risk of breast cancer? These types of questions are likely to be a recurrent theme as more chemopreventive agents are found to reduce cancer risk, albeit at a price of occasional toxicity.
Nonsteroidal anti-inflammatory agents may prevent or delay the onset of colorectal cancer in high-risk individuals. As Singh and Lippman point out, sulindac reduces polyp numbers in some, but not all, genetically high-risk individuals. Reports of adenocarcinomas arising from flat epithelium in sulindac-treated, genetically high-risk patients is a disturbing and potentially confounding piece of evidence regarding the use of polyps as surrogate end points for colorectal cancer. The reduction in polyp numbers and size in genetically high-risk individuals or persons with sporadic adenomatous polyps and no familial risk does not represent conclusive evidence of preventive efficacy.
Yet, the problem of selecting the most promising NSAID for a colorectal risk-reduction trial requiring tens of thousands of subjects and at least a decade of observation is sobering. Using an adenomatous polyp as a pathologic surrogate for chemopreventive efficacy is a rational, if not completely satisfying, compromise that enables the collection of chemopreventive human efficacy data at much lower cost and time compared to a risk-reduction trial.
Singh and Lippman also mention the lack of a clear-cut chemopreventive mechanism of action of NSAIDs. Although both sulindac (a cyclooxygenase inhibitor) and sulindac sulfone (a noncyclooxygenase inhibitor) induce colonic epithelial apoptosis in vitro and in vivo, their mechanisms of action may differ but may culminate in a similar outcome: increased apoptosis.
Cyclooxygenase-2 (COX-2), the inducible isoenzyme of cyclooxygenase, is expressed early in the carcinogenesis process of the colonic epithelium, is associated with a reduction in apoptosis, and is induced by cellular mitogens.[2,3] When COX-2 is selectively inhibited, apoptosis is induced, with associated reductions in adenoma and carcinoma formation in chemical and genetic rodent carcinogenesis models.[4,5] Cyclooxygenase-2 probably plays an important role in the control of apoptosis and, ultimately, in colorectal epithelial carcinogenesis and, potentially, in the formation of tumors in other organs (eg, the stomach, esophagus, breast, and lung).[6-9]
Although COX-2 represents an important chemopreventive target, other related proteins and signal transduction targets, such as NF-kappa-B, H-ras, and AP-1, may also be targets for the chemoprevention effects of NSAIDs.[10,11] The coming development of selective COX-2 inhibitors as chemopreventives for epithelial tumors will also permit researchers to probe the relationship between cyclooxygenase and noncyclo-oxygenase signal transduction pathways in epithelial cellular carcinogenesis.
Potential New Chemopreventive Approaches
The current excitement about selective COX-2 inhibition, coupled with confusion over the roles and chemopreventive mechanisms of NSAIDs, suggests new approaches for preclinical discovery and testing of preventive agents. New concepts of molecular epidemiology, signal transduction, molecular and biochemical regulation of cellular growth, differentiation, and apoptosis present targets for chemopreventive agents. Genetic knock-out and transgenic animals provide an added dimension to preclinical in vivo tests of chemopreventive agents, and may allow specific testing of potential chemopreventives derived from natural and nutritional sources or synthesized on the basis of important molecular carcinogenesis targets.
Identifying the crucial molecular, biochemical, genetic, and environmental events prior to transformation, and then developing, testing, and validating interventions that reverse or limit the genetic damage caused by these events, is a major scientific goal of cancer chemoprevention. Success should result in a later onset of common, spor-adic epithelial malignancies in the population, and, ultimately, a reduction in cancer-associated mortality. Perhaps the most important promise of cancer chemoprevention is the prolongation of useful, functional life from a disease that, on many occasions, appears at the peak of human economic and social productivity.
1. Fisher B, Costantino J, Wickerham D, et al: Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 study. J Natl Cancer Inst 90:1371-1388, 1998.
2. DuBois R, Radhika A, Reddy B, et al: Increased cyclooxygenase-2 levels in carcinogen-induced rat colonic tumors. Gastroenterology 110:1259-1262, 1996.
3. Coffey R, Hawkey C, Damstrup L, et al: Epidermal growth factor receptor activation induces nuclear targeting of cyclooxygenase-2, basolateral release of prostaglandins, and mitogenetis in polarizing colon cancer cells. Proc Natl Acad Sci USA 94:657-662, 1997.
4. Kawamori T, Rao C, Seibert K, et al: Chemopreventive effect of celecoxib, a specific cyclooxygenase-2 inhibitor on colon carcinogenesis. Cancer Res 58:409-412, 1998.
5. Oshima M, Dinchuk J, Kargman S, et al: Suppression of intestinal polyposis in ApcD716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell 87:803-809, 1996.
6. Parrett M, Harris R, Joarder F, Et al: Cyclooxygenase-2 gene expression in human breast cancer. Int J Oncol 10:503-507, 1997.
7. Ristimaki A, Honkanen N, Jankala H, et al: Expression of cyclooxygenase-2 in human gastric carcinoma. Cancer Res 57:1276-1280, 1997.
8. Hida T, Yatabe Y, Achiwa H, et al: Increase expression of cyclooxygenase 2 occurs frequently in human lung cancers, specifically adenocarcinomas. Cancer Res 58:3761-3764, 1998.
9. Wilson K, Fu S, Ramanujam K, et al: Increased expression of inducible nitric oxide synthase and cyclooxygenase-2 in Barretts esophagus and associated adenocarcinomas. Cancer Res 58:2929-2934, 1998.
10. Kopp E, Ghosh S: Inhibition of NF-kB by sodium salicylate and aspirin. Science 265:956-958, 1994.
11. Dong Z, Huang C, Brown R, et al: Inhibition of activator protein 1 activity and neoplastic transformation by aspirin. J Biol Chem 272:9962-9970, 1997.