ARTICLES

CANCER CHEMOPREVENTION - PRESENT STATUS

Lt. Col., MANOMOY GANGULY
MBBS, MS, DNB, MCH (Surg. 0nco.), FICS
CI, Specialist ( Surg. & Surg. Oncology)
Army Hospital, Research & Referral
Delhi Cantt,. New Delhi - 110 010

CHEMOPREVENTION is the use of pharmacologic intervention for disease prevention. The intent is to intervene in the pathophysiologic pathways that lead to clinical disease, but before clinical disease occurs. Although cancer chemoprevention is relatively new, there is a long history of chemoprevention in cardiovascular diseases, where interventions to counter hypertension, hypercho-lesterolemia and thrombosis have been used for years to prevent clinical atherosclerotic disease. As in cardiovascular diseases, chemopreventive agents may be identified through epidemiologic observations, through experiences with agents that have been used to treat frank cancer and through laboratory experiments. All of these modalities have identified successful cancer chemopreventive agents and also suggested some agents that have not been clinically successful.

ANTIOXIDANT SUPPLEMENTATION AND LUNG CANCER

In the early 19805, the hypothesis was advanced that beta-carotene, a vitamin A precursor, might have important cancer chemopreventive effect. The carotenoids, including beta-carotene, are plant pigments that are widespread in nature; they are responsible for many of the red, orange and yellow colors of plant and the animals that eat them. In plants carotenoids function to protect the cells from damage caused by excited molecular species generated by photosynthesis. These properties provided the rationale for the proposed chemo- preventive nature of beta-carotene, under the assumption that this compound would function as an antioxidant and quench excited genotoxic molecular species. A variety of other possibilities to explain anticarcinogenic effects of carotenoids were also advanced, including induction immune effects. In humans, beta-carotene is among the carotenoids found in high tissue concentrations, and epidemiologic finding suggested that high intake of beta- carotene was inversely related to the risk of several malignancies, particularly lung cancer. Vitamin E (a term that refers to eight natural compounds, including the tocopherols) is another antioxidant that has been studied with regard to lung cancer prevention. Selenium, though not in itself an antioxidant is often considered to be in this group because of its important role in glutathione peroxidase, an antioxidant enzyme system. For several reasons the study of the relationship between nutrients such as beta-carotene or vitamin E is difficult. The characterization of diet over long periods of time presents serious measurement problems, and the resulting measurement error tends to obscure associations of diet with disease. Also nutrients are not found in isolation, they are found in foods in combination with hundreds of other compounds, some of which may be biologically active.

For these reasons clinical trials are needed to clarify the effects of nutrients on risk, despite the large amount of observational data available. The increased lung cancer risks (and increased overall mortality) from beta-carotene in the Alpha Tocopherol, Beta-cartone Trial and Beta-carotene and Retinol Efficacy Trial created much scientific and public concern. The explanation for the findings is not clear, but several possibilities have been advanced. Since the Clinical trials began, it has emerged that beta-carotene may not be as effective an antioxidant in vivo as previously thought, a pro-oxidant effect of beta- carotene at high concentration has also been reported. The difference between the observational studies and the clinical trial findings could have been due to several factors. There may well be important differences between a lifetime of eating foods containing a complex mixture of nutrients and a few years of high dose supplementation with a single compound. Although the clinical trials involved treatment periods as along as 12 years, most of the lung cancer cases that emerged during these studies were probably the result of neoplastic progression that was already well underway at the beginning of the studies. Beta- carotene supplementation might have a preventive effect in individuals without neoplastic but a harmful effect on tumors already initiated. Meanwhile, the apparent protective effect of beta-carotene in the observational studies could have been due to other constituents of food rich in beta- carotene or to other characteristic of individuals who eat those foods. The protective effect of selenium on lung cancer in the Skin Cancer Prevention Trial (relative risk [RR) = 0.5, 95% confidence interval [CI), 0.2 to 0.9) was unexpected, but this provocative positive finding cannot be considered conclusive. It is not clear whether the result reflects real protective efficacy or whether the observation is the result of the play of chance. Because there were many cancer sites studied at the end of the trial, it is possible that the risk of some of these would be low in the selenium group simply by chance.

ANTIOXIDANT CHEMOPREVENTION OF PROSTATE CANCER ?

Chance finding could also underlie the lower risk of prostate cancer observed for alpha tocopherol or selenium supplementation in trials that were intended to study other cancer sites. In the Alpha Tocopherol, Beta- Carotene lung cancer study vitamin E conferred an RR of 0.71 (95% CL, 0.5 to 0.9) for prostate cancer. For clinical tumors (stages II-IV) the RR was 0.6 (95% CL. 0.5 to 0.8) and there was a similar reduction in prostate cancer death. Selenium also .led to a reduction in the risk of prostate cancer incidence and mortality in the Skin Cancer Prevention Trial (RR=0.4, 95% CL, 0.2 to 0.7) (Numbers of prostate cancer deaths were too few for meaningful analysis). Because of the possibility of chance effects, neither of these agents can be considered established preventive interventions for this cancer.

RETINOID CHEMOPREVENTION OF HEAD AND NECK CANCER

Retinoids are natural and synthetic analogs of Vitamin A, many of which act through specific receptors to modulate differentiation, the drugs can prevent -or in some case reverse -neoplastic phenotypes. Based on the premise, 13-cisretinoic acid(isotretinoin) was tested in patients with oral leukoplakia. Short -term (3 months) use of relatively high doses ( 1 to 2 mg/kg/d) was found to decrease the size of the lesions, reverse dysplasia, and inhibit progression to carcinoma. After this treatment, lower dose (0.5 mg/kg/d) maintenance therapy suppressed progression in contrast to beta- carotene, which was ineffective. Unfortunately the preventive effect is not sustained after cessation of treatment. Other retinoids also seem to be effective in preventing progression of leukoplakia to oral cancer (Vitamin A 200,000 IU/wk, 4-hydroxycarbophenyl retinamide 40 mg/d or 4N- hydroxyphenyl retinamide 200 mg/d), at least over the short term. In trials of patients treated with surgery and/ or radiation therapy for head and neck cancer isotretinoir, has also been effective in reducing the risk of second primary cancers (mostly smoking related cancer of the aerodigestive tract). After 12 months of adjuvant of treatment (1.5 mg/kg/d) suppression of second primaries lasted for approximately 3 years. Similar findings have been reported for adjuvant treatment of stage I lung cancer with retinol palmitate (300,000 I U daily for 1 year). However, etretinate, a synthetic retinoid, was not effective in preventing second primary cancer is one large trial among patients with oral cancer. A dose -dependent toxicity has been a feature of isotretinoin treatment, mucocutaneous effects, hepatotoxicity, and elevation in serum triglycerides. Unfortunately in some circumstances higher doses maybe important for efficacy of this drug. In nonmelanoma skin cancer for example, 10 mg is ineffective in preventing recurrence, but 2 mg/kg/d reversibly reduced the number of cancers by more than 60% in xeroderma pigmentosum and 0.4 mg/kg/d may be effective for basal cell nevus syndrome. For patients with xeroderma pigmentosum .or head and neck cancer with a very risk of cancers, some toxicity may be tolerable in view of the benefit. For lower-risk individuals, however (eg. in primary prevention) such toxicity is unacceptable, particularly because effective prevention may require prolonged administration of the drug.

ANTIOXIDANT SUPPLEMENTATION AND COLORECTAL NEOPLASIA

Because of the size and complexity of trials that are adequate for studying colorectal cancer itself as an end point in prevention studies much prevention research on colorectal carcinogenesis has focused on adenomas, which are precursors to most colorectal cancers. Patients with sporadic adenomas are routinely followed with endoscopy, and patients with familial adenomatous polyposis (FAP) receive even more intense surveillane. Assuming that effects on adenomas reflect those on cancer, these end points provide a convenient end point for study of the prevention of colorectal cancer itself.

Antioxidants have been studied in many trials, beta-carotene most intensively. It is clear that beta-carotene is ineffective when given for up to 4 years in middle-aged subjects who have already had at least one adenoma. Vitamin E also seems to be ineffective. Several of the antioxidant lung cancer trials were large enough to accrue sufficient numbers of colorectal cancer cases for meaningful analysis; the findings confirm that several years of beta-carotene or vitamin E will not alter colorectal cancer risk. For vitamin G, an early study suggested some benefit but subsequent studies failed to confirm this. Thus as in lung cancer, beta-carotene and Vitamin E have not demonstrated beneficial effects for large bowel neoplasia. However the interpretation of these trials is also not straightforward. The end points in the adenoma trials were small adenomas {typically < 0.5 mm), the earliest visible signs of neoplasia in .the bowel. Thus these investigations studied relatively early phases of carcinogenesis, and if beta-carotene affected later developments, the effect would be over- looked. Nonetheless, more prolonged supplementation might be required and this caveat applies particularly to the large trails that studied clinical cancer.

The Skin Cancer Prevention Trials suggested that selenium might inhibit colorectal neoplasia, but as for the lung cancer and prostate cancer findings, it is not clear whether this is a chance finding or a reflection of genuine efficacy.

CHEMOPREVENTION OF COLORECTAL NEOPLASIA: OTHER PROSPECTS

The suggestion in the early 1980s that calcium intake might lower the risk of colorectal neoplasia generated consider- able investigation. Although some epidemiologic studies have supported this hypothesis, these investigations have been mixed, perhaps reflecting the difficulties of dietary epidemiology. A recent clinical trial has confirmed a modest beneficial effect of calcium carbonate supplementation on adenoma occurrence. Over a 4 year treatment period, the risk of an individual having a recurrent adenoma was decreased by approximately 15% and the number of adenoms reduced by approxi- mately 25%. Findings that nonsteroidal anti-inflammatory drugs (NSAIDs) de- creased colorectal neoplasia in animals also motivated numerous human epidemiologic studies. With few exceptions, these have pointed to a chemoprotective effect, suggesting that risk can be reduced by approximately 50% among individuals who take aspirin or other NSAIDs regularly. Indeed, clinical trials in FAP patients have unif()rmly found that sulindac can lead to polyp regression and prevention of new polyps. ~n patients with sporadic adenomas there have been some hints of efficacy, but the effect seems to be less striking than in FAP. Thus anti-inflammatory agents have considerable promise for chemoprevention of colorectal neoplasia. Tempering this enthusiasm, however is the difficulty of extrapolating findings from FAP to the population of patients with sporadic adenomas: to the extent that the biology of FAP differs from the sporadic situation, there could well be difference in efficacy. In addition, it is clear from the observational studies (of sporadic colorectal cancer and adenomas) as well as the clinical trials (of FAP) that long term continued use of the NSAIDs is necessary for prevention. In individuals without inherited colorectal cancer syndromes,15 or more years will probably be required before there is a reduction in the risk of clinical colorectal cancer.

BREAST CANCER

The efficacy of tamoxifen as adjuvant treatment of breast cancer is well established. With 5 years of treatment, the drug confers approximately a 50% reduction in the risk of recurrence in women with early breast cancer. The benefit is concentrated in patients who have estrogen receptor (EA)- positive tumors, although there may be a small reduction in risk among women with EA-negative disease as well. However both younger and older women benefit. More relevant to cancer prevention is the effect of the drug to preventing contralateral breast cancer: again, approximately a 50% reduction with 5 year of treatment. In this case, women with both EA -positive and EA -negative primary cancers have about the same relative benefit, and the reduction in risk was similar in younger and older age group.

Building on these data, the Breast Cancer Prevention Trials studied 20 mg/d of tamoxifen in more than 13,000 women with- out breast cancer but deemed to be at higher than average risk. The trials was stopped after a median follow-up of approximately 4-5 years; tamoxifen reduced the risk of breast cancer by almost 50%. Only ER-positive tumors were affected; there was no reduction in risk for ER- negative breast cancer. The benefit entailed a cost: increase in the risk of endometrial cancer, venous thromoboembolism, and possibly stroke. Preliminary results from two European trials have not confirmed the finding but the small size (and consequent variability) of these studies and difference in the patient population leave the .Breast Cancer Prevention Trials as the dominant result.

In conclusion chemo prevention research to date has provided many surprises and illustrated many of the difficulties that may accompany studies in the future. The adverse finding of the beta-carotene trials has forced a re-evaluation of the assumptions underlying a whole body of research. Negative trials introduce their own questions: whether the dose tested was too low and whether the administration of the agent was too late or too limited in duration. Positive findings for the main hypothesis of a trial are relatively straightfoward. But in secondary analyses. Chance findings can be difficult to interpret.

Nonetheless, cancer chemoprevention is a reality. Tamoxifen can be considered to be an established preventive drug for breast cancer, and isotretinoin seems to be effective is in preventing second primary cancers in patients with cancers of the head and neck. However understanding these effective agents is not straightforward. Although on average both drugs provide a net benefit, it is likely that for some low-risk patients, their adverse effects will counter- balance their benefits. On the other hand, a very nontoxic agent, calcium carbonate, has been shown to have chemopreventive efficacy against large bowel carcinogenesia in one clinical trials but its modest effect may not lead to change in follow-up recommendations. Balancing efficacy against toxicity may well be a recurring difficulty for follow-up of successful chemoprevention research. This balance is made more difficult because of the apparent need for continuous use of chemopreventive interventions; to date, no agent has been found to confer a "permanent" protective effect.

REFERENCES

1.Sporn MB: Chemoprevention of cancer. Lincet. 342:1211-1213, 1993.

2.International Agency for Research on Cancer, WHO: IARC Handbooks of Cancer Prevention. Lyon, France, 1998.

3.The Alpha -Tocopherol, Beta-Carotene Cancer Prevention Study Group: N Engl J Med 330;1029-1035,1994.

4.Benner SE, Pajak TF, Lippman SM et al: Prevention of second primary tumors with isotreinon in patients with squamous cell carcinoma of the head and neck: Long- term follow-up. J Natl Cancer Inst 86; 140- 141, 1994.

5.Baron J, Beach M, Mandel J, et al: Calcium supplements for the prevention of colorectal adenomas. N Engl J' Med 340;101-107, 1999.

6.Pritchard KI; Is tamoxifen effective in prevention of breast cancer ? Lancet 352;80-81, 1998.