Androgen Deprivation as a Strategy for Prostate Cancer Chemoprevention

Androgens are required for the normal development and function of the prostate gland. Prostate cancer and benign prostatic hyperplasia are common in men and develop in an environment of continuous androgen exposure. The utility of androgen deprivation as a treatment for advanced prostate cancer was first demonstrated in 1941 (1), and many new classes of drugs that interfere with androgen production and function have been introduced in recent years. These agents are effective in treating prostate cancer and benign prostatic hyperplasia and may have an important role in chemoprevention. Since 1991, prostate cancer has been the most common noncutaneous cancer and the second leading cause of cancer-related death in men. The major difficulty in treating prostate cancer is the limitation of current tools to accurately predict the behavior of a given tumor. Although many prostate cancers are clinically insignificant and will have an indolent course, others will result in considerable morbidity and mortality. The major dilemma with regard to prostate cancer is determining what men to treat and who would be better served by expectant management (2). Previous efforts (3) have concentrated on treating invasive cancer, often in advanced stages. Other work (4,5) suggests that an evolving multistep molecular and cellular process, known as carcinogenesis, begins years before invasive cancer is detected. Chemoprevention refers to prevention of cancer or reduction of risk in susceptible individuals by administration of natural or synthetic drugs with little or no toxicity that suppress, delay, or reverse carcinogenesis (6,7). Chemoprevention is most effective in the early stages of cancer formation when reversibility may be feasible. In prostatic carcinoma, the time from tumor initiation and progression to invasive carcinoma often begins in men in the fourth and fifth decades of life and extends across decades, allowing sufficient time to halt or to reverse carcinogenesis (8,9). The utility of prevention of disease as a means of improving quality of life and outcome has been amply demonstrated in recent years; one of the first examples was the identification of atherogenesis as the causative mechanism for cardiovascular disease (10). Improved understanding of atherogenesis resulted in important health initiatives, including modifications of lifestyle, diet, and tobacco usage and introduction of cholesterollowering, antihypertensive, and antiplatelet agents that have favorably affected life expectancy by reduction of cardiovascular risk (4,10). Similar preventive strategies are being developed to suppress the process of carcinogenesis. In chemoprevention of prostate cancer, as with prevention of cardiovascular disease, the potential public health implications are substantial. Antioxidant vitamins often have beneficial effects in prevention and treatment of cancer, according to results from chemoprevention trials. Daily treatment with high doses of isotretinoin, a vitamin A derivative (retinoid), is effective in preventing second primary tumors in patients who have been treated for squamous cell carcinoma of the head and neck (11). Administration of vitamin E and several retinoid agents results in clinical and histologic regression of premalignant oral lesions (12). Sulindac, a nonsteroidal anti-inflammatory drug, significantly decreased mean colonic polyp number and diameter in patients with Gardner's syndrome and familial polyposis (13–17). Tamoxifen, an antiestrogen, exhibits potent activity against hormone- sensitive forms of breast cancer and is currently being evaluated in a phase III clinical trial in women at high risk for breast cancer (18). For prostatic carcinoma, the interval between tumor initiation and progression is measured in decades. Autopsy studies indicated that prostatic intraepithelial neoplasia, the precursor of most prostate cancers, precedes carcinoma by 10 years or more, first emerging in men in the third decade of life (8,19,20). There may be sufficient time to halt or reverse carcinogenesis. Absolute eradication of precursor lesions or early cancer may not be required for the strategy to have a beneficial effect, because a delay in the development or slowing the progression of prostatic carcinogenesis may prolong a patient's life. The prostate gland requires androgens for proper growth, maintenance, and function. Men with 5α-reductase deficiency (21,22) and men castrated when young (eunuchs) do not develop prostate cancer (23). These populations provide indirect evidence that a strategy directed toward interrupting or decreasing exposure to androgens can significantly decrease prostate cancer incidence. Androgen-deprivation therapy causes marked and characteristic changes in the normal prostate and in prostate cancer (24). Ferguson et al. (25) reported a marked decrease in the prevalence and extent of high-grade intraepithelial neoplasia in prostates after androgen-deprivation therapy compared with untreated prostates. This decrease was accompanied by epithelial hyperplasia, cytoplasmic clearing, and prominent acinar atrophy, with a decreased ratio of acini to stroma. These findings indicate that the dysplastic prostatic epithelium is hormone dependent. In the normal prostatic epithelium, luminal secretory cells are more sensitive to the absence of androgens than are basal cells, and these results show that the cells of high-grade prostatic intraepithelial neoplasia share this androgen sensitivity. The loss of normal, hyperplastic, and dysplastic epithelial cells with androgen deprivation is probably from acceleration of programmed single-cell death (apoptosis) with subsequent exfoliation intoacinar lumina (26,27). Prostatic epithelium from benign hyperplasia and from most carcinomas responds to hormonal manipulation. This finding suggests that the process of prostatic carcinogenesis can be favorably modulated if an appropriate agent and treatment regimen can be determined. Intervention early in carcinogenesis may be more successful in improving survival because the burden of affected cells is low. Once stromal invasion occurs, it may be too late to prevent further disease progression (10). Agents capable of producing effective and reversible androgendeprivation therapy are now available, and chemoprevention of prostate cancer is a realistic possibility (28). Surgical castration and most chemical agents capable of effective androgen-deprivation therapy are associated with some degree of significant adverse effects. These effects are often acceptable to patients with cancer, but not to a healthy population of men in whom the risk of developing cancer is relatively low. Patients with high-grade prostatic intraepithelial neoplasia or other high-risk target populations are more likely to accept minor side effects if the chemopreventive effort is short term, efficacious, and inexpensive and has the potential for significant improvement in long-term outcome (18,29). Chemoprevention must match potential side effects to degree of cancer risk (30). Estrogen therapy with diethylstilbestrol lowers luteinizing hormone concentration and thereby depresses serum testosterone level (31). Diethylstilbestrol in doses of 3 mg/day or more results in significant cardiovascular toxicity (32). Diethylstilbestrol at 1 mg/day has fewer side effects but does not reliably reduce testosterone concentration to castrate levels (33). Other than cardiovascular complications, estrogens have been associated with fluid retention, gynecomastia (32), and loss of libido and potency (34)—side effects that probably outweigh any chemopreventive benefit of these agents. Continuous administration of high doses of luteinizing hormone-releasing hormone analogues inhibits the release of luteinizing hormone and follicle-stimulating hormone and subsequently suppresses testosterone, reaching castrate levels after 3–4 weeks. Luteinizing hormone-releasing hormone analogues (leuprolide acetate and goserelin acetate), however, initially produce a significant increase in testicular androgen production, known as the flare reaction, which can be inhibited with antiandrogens. The major side effects of these agents include impotence, loss of libido, atrophy of muscles and the reproductive organs, and hot flashes. These potent agents are not suitable for chemoprevention in the general population; however, in highrisk target populations such as those with high-grade prostatic intraepithelial neoplasia, perhaps 3–6 months of therapy could significantly improve outcome and outweigh the side effects incurred during the treatment interval. The 5α-reductase enzyme is a membrane-bound protein dependent on the reduced form of nicotinamide-adenine dinucleotide phosphate and is responsible for conversion of testosterone to the more potent dihydrotestosterone in androgen-dependent target cells (35). Dihydrotestosterone has a greater affinity than testosterone for the androgen receptor, and it is thought to actively modulate prostatic growth. Finasteride is a steroid analogue that acts as a competitive and specific inhibitor of 5areductase, resulting in suppression of serum and intraprostatic dihydrotestosterone concentrations to castrate levels, with subsequent reduction in prostatic size. 5α-Reductase inhibition results in a significant increase in intraprostatic testosterone levels, allowing testosterone-mediated functions including libido, potency, and male musculature to remain largely unaffected (35). Finasteride is approved for treatment of symptomatic benign prostatic hyperplasia (36). It is associated with a 50% reduction in prostate-specific antigen levels (37,38), which is significantly lower than the reduction associated with medical or surgical castration (39). The role of finasteride in the treatment of human prostate cancer or its precursor lesions is unknown. Currently, other 5α-reductase inhibitors are in development (e.g., episteride). Tumor growth has been suppressed by 5a-reductase inhibition in animal and human tumor cell lines (31,40,41). An attractive feature of finasteride is its excellent safety profile (42), making it a reasonable candidate for chemoprevention trials in the general and high-risk target populations. Nonsteroidal antiandrogens competitively bind to androgen receptors at the target cell level. This class of agents blocks testicular and adrenal androgens. Antiandrogens are currently approved (e.g., flutamide and bicalutamide) for use in combination therapy with a leutinizing hormone-releasing hormone analogue for complete androgen blockade (43,44) in the management of prostate cancer. Antiandrogens and 5a-reductase inhibitors, if used alone or in combination for chemoprevention, may offer a quality-of-life advantage over other androgenablation methods because they do not reduce serum testosterone levels and therefore do not have a marked inhibitory effect on libido and potency (43,45). Flutamide and bicalutamide are nonsteroidal antiandrogens that selectively inhibit androgen receptors, including those in the prostatic cytoplasm (46,47). These agents have no androgenic or progestational activity and are able to effectively counteract testicular and adrenal androgens. When flutamide or bicalutamide has been used as monotherapy for prostate cancer, mild increases in serum testosterone levels have been noted (47,48). The increase in serum testosterone levels with flutamide administration appears to be clinically insignificant, showing no association between testosterone levels and patient response to treatment (48). Flutamide is approved at a dose of 250 mg given three times a day at 8-hour intervals for combination therapy with a luteinizing hormone-releasing hormone analogue for treatment of prostate cancer. Flutamide has been evaluated at this dose to determine its long-term efficacy as monotherapy for prostate cancer (46). Adverse effects in this population included diarrhea, gynecomastia, nausea or vomiting, hepatitis, vertigo, and hot flashes (46). A reduction in dose (e.g., flutamide at 250 mg given once daily or 125 mg given two or three times daily) may offer chemopreventive efficacy and minimize or eliminate many adverse effects. Combining androgen-ablation agents may decrease prostate cancer risk in target populations such as those with high-grade prostatic intraepithelial neoplasia. Combination therapy may decrease overall side effects by lowering effective doses and shortening treatment durations. A nonsteroidal antiandrogen agent could be used in combination with a 5α-reductase inhibitor in men who want potency preserved (34,45). Combination therapy (49) with a luteinizing hormone-releasing hormone analogue and an antiandrogen would be an option for men in whom potency is not a concern. Recently, the National Cancer Institute sponsored workshops to design clinical trial strategies for chemoprevention and introduced the concept of surrogate end-point biomarkers as intermediate trial end points. Intermediate trial end points include histologic and biochemical alterations, proliferation, differentiation, and genetic biomarkers (Table 1) (50). Each of these endpoint biomarkers is measured in serum or tissue specimens, such as formalin-fixed, paraffin-embedded, needle biopsy specimens, and may be modifiable by intervention. The problem of the measurement of these biomarkers during clinical trials in humans is the amount of tissue available for analysis. Recent advances in the development of immunohistochemistry, in situ hybridization (51), and other microassay techniques make such measurements feasible on serial sections of needle biopsy specimens of the prostate. Chemoprevention trials may apply these techniques to study the multistep process of carcinogenesis and to determine the cellular effects of potential chemopreventive agents on surrogate end-point biomarkers. A significant limitation of monitoring surrogate end-point biomarkers from sequential prostate needle biopsies is the inability of current techniques to reliably resample the same microscopic focus after chemopreventive therapy to determine if the intervention was efficacious. In a solid organ such as the prostate, this limitation is unresolvable. Studies using cancer as an end point must treat large groups of subjects, recognizing that cancer will develop in only a small number early in the course of study. The number of subjects required in studies using surrogate end-point biomarkers is much fewer than in studies in which cancer incidence is the primary end point (52). Rather than having to wait many years to accumulate sufficient numbers of subjects with invasive cancer to determine if the intervention was successful, surrogate end points shorten the duration of chemoprevention trials (53). The smaller number of subjects and shorter duration of study are associated with a significant reduction in the cost of conducting chemoprevention studies (52). Agents that favorably modulate surrogate end-point biomarkers are reasonably likely to or reverse carcinogenesis, but clinical utility must be by a reduction in cancer incidence There are several target populations who can be for prostate cancer chemoprevention (Table of these trial design strategies include patients with cancer and can be trials These studies may provide the of agents on surrogate end-point premalignant and cancer. using agents that prostate-specific antigen may to these results from the patient and treating during the study to Chemoprevention clinical trials for the general population may a relatively time for cancer to develop in sufficient numbers of Prostate cancer is the of study may be significantly and the number of subjects required than those in studies on high-risk target populations. The major advantage of studies for the general population is that the results are to this A prostate cancer chemoprevention trial a years The Prostate Cancer is a phase III trial of finasteride in the prevention of prostate cancer. This study has men to mg of finasteride or daily for with a during which before treatment The trial on and patients in subjects are men years or with normal results and prostate-specific antigen concentration than 3 these subjects are according to and of with prostate cancer The minimize the of men with prostate cancer. The primary end point of the Prostate Cancer is the prevalence of prostate cancer at years. Prostate cancer in will be by biopsy of subjects at years or by the of prostate cancer any time before that (36). from a study of this determine if finasteride significantly inhibits prostate cancer and what its role is as a chemopreventive however, such a study that on cancer incidence as an end-point has including the of a large number of duration of and high The relatively small risk of prostate cancer death in the general population the of chemopreventive agents capable of being because men at risk would be to accept any degree of adverse effects for a reduction in risk of prostate cancer death (31). Prostate cancer chemoprevention trials directed at men or men with a familial for prostate cancer could include target populations. men have the of prostate cancer in the for prostate cancer probably for of and risk increases as the number of affected increases A risk is if a such as a or has prostate cancer. The risk is to if the has with prostate cancer appears to be an important risk for prostate cancer prostatic intraepithelial neoplasia is a finding to be the precursor of most prostate cancers This the end of the that in high-grade intraepithelial neoplasia and early invasive cancer Prostatic intraepithelial neoplasia is by basal cell loss of secretory and and in It is and high to the is and 3 are high The incidence of prostatic intraepithelial neoplasia to increase with patient In an study of young male patients et al. the of disease in men in third and fourth decades of life and which the of carcinoma by more than 10 years. The of high-grade first in the fifth In the by et al. high-grade in prostates The clinical of recognizing prostatic intraepithelial neoplasia is on its association with prostate cancer. is the only for this premalignant and early invasive cancer. prostatic intraepithelial neoplasia is in to of needle biopsies It has a high as a for et al. carcinoma in of subsequent biopsy specimens from men with prostatic intraepithelial neoplasia compared with of those from a intraepithelial neoplasia, patient and levels of specific antigen in significant of cancer, but prostatic intraepithelial neoplasia the risk ratio Other have reported a high of prostatic intraepithelial neoplasia for cancer, from to The high risk of cancer on subsequent biopsy and the of treatment than and for patients with this premalignant provide a reasonable for chemoprevention efforts in this target from normal prostatic epithelium of prostatic intraepithelial neoplasia to early invasive carcinoma, according to the of prostatic intraepithelial neoplasia 1) to mild to mild and to to and carcinoma in The precursor when cells the this invasion the basal cell is that the dysplastic changes in the secretory cell perhaps in response to luminal of the basal cell the and of high-grade prostatic intraepithelial neoplasia and appears to be a for stromal The is with high-grade prostatic intraepithelial neoplasia and early invasive Prostatic neoplasia and early invasion in prostate cancer. Cancer Cancer of a of treatment of men with prostate cancer by use of potential chemopreventive agents before may be the most and feasible clinical to of high-grade prostatic intraepithelial neoplasia and other potential surrogate end-point biomarkers The delay before can be as as several time for androgen-deprivation agents and other to and appropriate and tissue The would be available for study of the effect of the on and genetic intermediate In a study of deprivation Ferguson et al. (25) that therapy a significant decrease in the prevalence and extent of high-grade prostatic intraepithelial neoplasia in patients compared with untreated Similar studies could be as trials with chemopreventive agents. patients are known to have cancer, this of study could be a agent could be the of the candidate chemopreventive agents on potential surrogate end-point biomarkers and cancer The clinical utility of a potential chemopreventive agent in a be with because its may on the intervention is early or late in the stages of tumor development from trials be in chemoprevention is appropriate management for some patients with prostate cancer, the for such patients remain a of men who have prostate cancer will with cancer than of cancer. are to for patients with clinically significant cancer of tumor is and too for patients to of therapy on this in of prostate cancer extent and potential will some patients to be the level of serum prostate-specific antigen has as a surrogate for tumor growth and agents could be given to patients and biopsies and serum could changes in surrogate end-point biomarkers. This is as or than as chemopreventive because the patient has cancer. This strategy is by of to the prostate gland the patient or is and could be by patient and These make such trials in patients with early cancer treated by feasible and the amount of that could be lesions such as high-grade prostatic intraepithelial neoplasia patients at high risk of developing invasive cancer, and these patients are target populations for chemoprevention trials. The most strategy for developing a chemoprevention may be to two clinical trials on the of high-grade prostatic intraepithelial neoplasia but in target populations. In patients with high-grade intraepithelial neoplasia associated with prostate cancer, a study to the response of a potential chemopreventive agent in the before could be Androgen-deprivation therapy is used in this population to the prostate before This study would determine the response of prostatic intraepithelial neoplasia to the agent in for the effects of chemopreventive agents on high-grade prostatic intraepithelial neoplasia and cancer before for chemoprevention of prostate cancer in men with Chemoprevention of prostate cancer in men with high-grade prostatic intraepithelial neoplasia of the A phase chemoprevention trial with cancer incidence as an end point could be in patients with prostatic intraepithelial neoplasia cancer Chemoprevention trials to reverse high-grade may be by the of but prostate cancer. This problem is by a second biopsy with findings for cancer before the study biopsies with being given to of on or and by including subjects in the study and groups to the risk of cancer between the two groups Prostatic intraepithelial neoplasia is by biopsy in would be including and surrogate intermediate end-point biomarkers. that sequential prostate needle biopsy specimens will of clinical or but they may not of of treatment effect because of and inability to the biopsy in the same microscopic for subjects in a study are treated for 1 with a chemopreventive prostate needle biopsies at and months will effects and will to be evaluated subsequent biopsy prostate cancer, these patients with prostatic intraepithelial neoplasia or no cancer and it is likely that cancer at and 10 years may be the most end point for these chemoprevention This chemoprevention trials and would provide would be to long-term treatment for patients with high-grade prostatic intraepithelial neoplasia. These study methods are with current and would other chemopreventive agents to be evaluated in a The could provide the for clinical trials directed toward the prevention and of prostate cancer in other high-risk target populations and in the general Prostate cancer is an androgen-dependent with a before the of invasive These and recent advances in to the hormonal environment provide an for early intervention with the of decreasing morbidity and from prostate cancer. Chemoprevention trials use agents at doses that have if adverse effects. Chemoprevention however, may have acceptable toxicity levels according to the agent and the risk of cancer in the study Chemoprevention strategies that use highrisk target those with premalignant have potential to shorten the duration of study and fewer resulting in a significant in cost and The results from these however, be in trials the general in understanding of the process of carcinogenesis and the of new chemopreventive including those producing reversible androgen have the potential to have a on the morbidity and from prostate cancer in the