First published online on May 2, 2005
Endocrine Reviews, doi:10.1210/er.2004-0017
Endocrine Reviews 26 (3): 361-379
Copyright © 2005 by The Endocrine Society
Gonadotropin-Releasing Hormone Agonists in the Treatment of Prostate Cancer
Fernand Labrie,
Alain Bélanger,
Van Luu-The,
Claude Labrie,
Jacques Simard,
Leonello Cusan,
José Gomez and
Bernard Candas
Oncology and Molecular Endocrinology Research Center, Laval University Medical Center (Centre Hospitalier de l’Université Laval) and Laval University, Quebec City, Quebec, Canada G1V 4G2
Correspondence: Address all correspondence and requests for reprints to: Professor Fernand Labrie, Oncology and Molecular Endocrinology Research Center, Laval University Medical Center (Centre Hospitalier de l’Université Laval), 2705 Laurier Boulevard, Quebec City, Quebec, Canada G1V 4G2. E-mail: fernand.labrie{at}crchul.ulaval.ca
 |
Abstract
|
|---|
In 1979, the first prostate cancer patient was treated with a GnRH agonist at the Laval University Medical Center in Québec City, Canada, thus rapidly leading to the worldwide replacement of surgical castration and high doses of estrogens. The discovery of medical castration with GnRH agonists was soon followed by fundamental changes in the endocrine therapy of prostate cancer. Most importantly, the excellent tolerance accompanying the treatment with GnRH agonists has been a key factor that permitted a series of studies demonstrating a major reduction in the death rate from prostate cancer ranging from 31 to 87% at 5 yr of follow-up in patients with localized or locally advanced prostate cancer. In fact, a one third reduction in prostate cancer deaths has been calculated in the metaanalysis of all available studies. The general acceptance of this discovery by patients and physicians is illustrated by world sales above 3.0 billion U.S. dollars in 2003.
Although extremely efficient in achieving complete medical castration and well tolerated, with no other side effects than the expected hypoandrogenicity, GnRH agonists should not be administered alone. In fact, shortly after discovery of the castration effects of GnRH agonists, we observed that approximately 50% of androgens remain in the prostate after castration, thus leading to the recognition of the role of adrenal dehydroepiandrosterone as an important source of the androgens synthesized locally in the prostate and in many peripheral target tissues. We therefore developed combined androgen blockade (CAB), whereby the androgens of both testicular and adrenal origins are blocked simultaneously at start of treatment with the combination of a GnRH agonist to block the testis and a pure antiandrogen to block the action of the androgens produced locally. CAB, first used in advanced metastatic disease, has been the first treatment shown to prolong life in prostate cancer. Most interestingly, in 2002, we made the observation that CAB alone given continuously for 6.5 yr or more leads to cure of the disease in at least 90% of cases, thus suggesting that androgen blockade combining a GnRH agonist and a pure antiandrogen could well be the most efficient treatment of localized prostate cancer, and thus offering the possibility of practically eliminating death from prostate cancer.
- I. Introduction
- II. Limitations of Surgical Castration and High Doses of Estrogens
- III. Discovery of the Castration Effect of GnRH Agonists Led to a Remarkable Therapeutical Success
- A. GnRH superagonists were developed to treat infertility although their medical use turned out to be the opposite
- B. Paradoxical inhibitory effects of GnRH agonists on the testis, a unique observation in biology and medicine
- IV. First Prostate Cancer Patient Treated with a GnRH Agonist
- V. Medical Castration with GnRH Agonists Is Equivalent to Surgical Castration: Very Rapid Medical Use Worldwide
- VI. Mechanisms of Medical Castration by GnRH Agonists
- VII. Transient Stimulation of Serum Testosterone Completely Neutralized by a Pure Antiandrogen: the Potential Risk of Flare with a GnRH Agonist Is Thus Completely Removed
- VIII. Monotherapy with a GnRH Agonist Decreases Cancer Death by at Least One Third in Localized Prostate Cancer
- IX. Major Importance of Adding a Pure Antiandrogen to the GnRH Agonist (or Surgical Castration)
- A. Two equally important sources of androgens are present in men
- B. Combined androgen blockade in advanced disease
- X. High Probability of Cure of Localized Prostate Cancer by Treatment with Combined Androgen Blockade
- XI. Neoadjuvant Therapy
- XII. Secondary Effects of Androgen Blockade
- XIII. Substitution of Castration or CAB by an Antiandrogen Alone
- XIV. High Risk of Watchful Waiting
- XV. Conclusions
|
I. Introduction
|
|---|
PROSTATE CANCER IS the most frequently diagnosed cancer and the second cause of cancer death in men in North America (1). In fact, one in eight men will be diagnosed with prostate cancer during his lifetime. At the present rate, of the male population living in the United States, prostate cancer will kill more than 3 million men. Prostate cancer is thus a major medicosocial problem comparable to that of breast cancer in women. In fact, it was predicted that 20,900 men would die from prostate cancer in the United States in 2004.
|
II. Limitations of Surgical Castration and High Doses of Estrogens
|
|---|
In 1941, Huggins et al. (2) observed some dramatic responses in metastatic prostate cancer patients after surgical castration or treatment with high doses of estrogens. During the 50 yr that followed the introduction of the concept of androgen dependency of prostate cancer, orchiectomy and high doses of estrogens have been the gold standard for the treatment of advanced prostate cancer (Fig. 1
). Although this treatment achieves only a partial inhibition of androgens because it is limited to blockade of the androgens of testicular origin, reports from many groups have shown that such treatment achieves a positive response in 60 to 70% of patients, although for a limited period of time (3, 4, 5, 6, 7). As indicated by such a high proportion of positive responses achieved after only partial blockade of androgens, prostate cancer is the most sensitive of all hormone-sensitive cancers to endocrine therapy. This uniquely high sensitivity of prostate cancer to androgens should be exploited optimally to best succeed in the fight against this disease.
The serious and frequently lethal cardio- and cerebrovascular complications of estrogens (5, 8, 9) on one hand, and the psychological (10, 11) as well as physical limitations of orchiectomy on the other hand, have generally delayed endocrine treatment until late stages of the disease when pain and debility had developed. Typically, at such a late stage, the large and disseminated tumors show poor and short-lived responses, thus limiting the success of endocrine therapy. In fact, similar to treatments for all other types of cancers, androgen blockade loses its effectiveness with increasing size of the tumors (12).
|
III. Discovery of the Castration Effect of GnRH Agonists Led to a Remarkable Therapeutical Success
|
|---|
A. GnRH superagonists were developed to treat infertility although their medical use turned out to be the opposite
The discovery by Schally and Guillemin and their colleagues (13, 14) of the structure of GnRH, the hypothalamic hormone that controls the secretion of LH and FSH by the anterior pituitary gland, was a major breakthrough. In fact, the availability of synthetic GnRH has allowed detailed studies of the mechanism of action of GnRH on LH and FSH secretion in the anterior pituitary gland (15) while providing essential information on the interactions of GnRH with androgens, estrogens, and progestins at the pituitary level (16, 17, 18). Such studies were fundamental for a proper understanding of the delicate feedback mechanisms controlling GnRH secretion by the hypothalamus as well as the action of GnRH on LH and FSH secretion in the anterior pituitary gland with their resulting impact on gonadal functions and fertility.
In addition to providing the possibility of performing extensive biological studies with GnRH, knowledge of the structure of this neurohormone offered the opportunity of designing and synthesizing peptides much more potent than GnRH itself. The objective of that research program was to obtain highly potent GnRH agonists for the treatment of infertility (19, 20). In fact, within 4 yr after discovery of the structure of GnRH, superagonists of GnRH having 100 to 200 times the in vivo biological activity of GnRH were already available (21, 22, 23). The original objective determined in 1971 to obtain highly active GnRH agonists that would require small quantities of compound for administration to the human with infertility problems had been achieved within 4 yr. Since 1975, no further significant improvement of the molecules of GnRH agonists in terms of potency or otherwise was expected or needed, and such improvement has not occurred. In fact, all of the GnRH agonists that later became commercially available possess equivalent potency, efficacy, and mechanism of action (Fig. 2).
All research groups working in this field, including ours, were then attempting to find the dose and vehicle as well as the dosing schedule and route of administration that would permit the use of GnRH agonists to stimulate fertility in experimental animals and humans.
B. Paradoxical inhibitory effects of GnRH agonists on the testis, a unique observation in biology and medicine
In the course of our attempts to find an explanation for the lack of stimulatory effect of chronic administration of GnRH agonists on gonadal functions, we then made the unexpected observation that treatment for a few days of adult male rats led to variable degrees of inhibition of serum testosterone levels accompanied by a relatively small but usually significant inhibition of ventral prostate, seminal vesicle, and testis weight (24, 25). It should be mentioned that when we were treating rats with a GnRH agonist some 28 yr ago, we were expecting to observe larger seminal vesicles and a prostate of increased volume. Most unexpectedly, the opposite observation was made: the prostate, the seminal vesicles, and the testicles became smaller instead of larger after a few days of treatment with a GnRH superagonist (Fig. 3) (24, 25). It should be mentioned, however, that marked sensitivity differences exist between animal species. Accordingly, male mice and monkeys (26, 27, 28, 29) are relatively insensitive to GnRH agonists, whereas rats are moderately sensitive.
This inhibitory effect of GnRH agonists was in complete contradiction with the well-established pharmacological principle that a synthetic molecule having a biological activity higher than the natural hormone should exert a more potent but otherwise similar activity. In fact, superactive molecules are usually synthesized to permit the use of a lower dose and thus possibly minimize the risk of side effects associated with high quantities of administered compounds. Because of its paradoxical nature and its lack of conformity with classical biology, the understanding of the mechanism of action of these partially inhibitory effects exerted on testicular functions by GnRH agonists in the rat remained an unsolved problem for some time.
|
IV. First Prostate Cancer Patient Treated with a GnRH Agonist
|
|---|
Although experiments performed in the rat were simply suggestive of an inhibitory effect of GnRH agonists on testicular functions, we discovered in 1979 at our clinic at the Laval University Medical Center that medical castration is achieved in men after chronic administration of GnRH agonists.
We then studied the effect of administering the GnRH agonist buserelin to a patient suffering from stage B prostate cancer. Thus, in the first prostate cancer patient treated with a GnRH agonist, the 500-µg dose of buserelin administered intranasally caused 70% and 85% inhibitions of the serum levels of testosterone and dihydrotestosterone (DHT), respectively, as early as 2 wk after the start of therapy (Fig. 4) (30). This marked inhibition of the serum concentration of both testosterone and DHT followed an initial period of stimulation that lasted approximately 1 wk. Most importantly, it could be seen that the serum DHT concentration was decreased even further than that of serum testosterone, thus clearly indicating that treatment of adult men with a GnRH agonist was not accompanied, contrary to the situation in the rat (30), by a simultaneous increase in the concentration of DHT that would compensate for the inhibition of serum testosterone. Medical castration induced by a GnRH agonist had thus become a clear possibility in men.
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 4. Effect of twice daily intranasal administration of the GnRH agonist buserelin on the serum levels of testosterone (A) and DHT (B) in a patient suffering from stage B prostate cancer (30 ).
|
|
Soon after our observation (30) that administration of the GnRH agonist buserelin led to an almost complete inhibition of serum testosterone and DHT levels within 2 wk after administration by the intranasal route, a less than optimal route of administration (30), a detailed comparison of the effect of various doses of the same GnRH agonist was performed after administration by the intranasal and sc routes (31). The effect of chronic treatment with the GnRH agonist buserelin administered by nasal spray (200 or 500 µg twice daily) or sc (50 µg daily) for periods up to 8 months was thus studied on serum sex steroid and LH levels in patients with stage A or B prostate cancer who had been treated surgically by prostatectomy and had no sign of active disease (32). Basal serum testosterone concentration decreased to 71.1 ± 18.3% (not significant) and 28.6 ± 9.3% (P < 0.01) of control in patients receiving the 200- and 500-µg doses by nasal spray, respectively (Fig. 5). In patients treated sc with the 50-µg daily dose, a more rapid inhibition of serum testosterone levels to 19.6 ± 6.4% of control (P < 0.01) was observed (31, 32). The most complete and rapid inhibition of serum testosterone levels to 5–8% of control (castrated levels) was achieved with the daily 200- and 500-µg doses administered sc (Fig. 5).
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 5. Effect of increasing doses of the GnRH agonist [D-Ser(TBU)6, des-Gly-NH210]-GnRH ethylamide (buserelin) administered for 1 month intranasally (200 or 500 µg) twice a day (B.I.D.) or sc [50, 200, or 500 µg once daily (I.D.)]. Measurements were performed at 0800 h and are compared with the values observed after surgical castration. Data are presented as means ± SEM (31 32 ). The number of patients ranged between five and eight per group.
|
|
|
V. Medical Castration with GnRH Agonists Is Equivalent to Surgical Castration: Very Rapid Medical Use Worldwide
|
|---|
Medical castration with a GnRH agonist is equivalent to orchiectomy for prostate cancer therapy (33). In a comparison of 11 trials in which a GnRH agonist was used and in 17 trials in which orchiectomy was used, no difference was seen in the response or survival rate (33).
The availability of a safe and highly efficient method of medical castration has generated renewed interest in the treatment of prostate cancer and has stimulated an unprecedented number of clinical studies, which rapidly led to the worldwide commercialization of a series of GnRH agonists having equivalent characteristics, mechanisms of action, and efficacy (Fig. 2
). This marked the end of the requirement for surgical castration, a procedure that is psychologically difficult to accept by the majority of men. Most importantly, this was the end of the need to administer high doses of estrogens to achieve medical castration at the expense of serious cardiovascular effects (5, 8, 9).
As can be seen in Fig. 6, the number of publications on GnRH agonists has increased since 1971 to 524 per year in 1989. In 2001, GnRH agonists held nearly half of the market of hormonal cancer drugs with annual sales of the order of 3 billion U.S. dollars. Such numbers clearly indicate the popularity of the GnRH agonists in medicine. In fact, GnRH agonists are now the treatment of choice for men needing testicular androgen blockade for the treatment of prostate cancer. GnRH agonists have thus been used by millions of prostate cancer patients for more than 20 yr with no adverse effect other than those associated with androgen deprivation (34).
The importance of such results obtained with GnRH agonists was well recognized by Jacobi and Wenderoth (35) who stated: "What medical developments have urologists witnessed since orchiectomy and estrogen treatment by Huggins 40 yr ago? Gestagens, antiandrogens, adrenal inhibitors, antiprolactins, antiestrogens, cytotoxic agents? In principle, the gain in terms of efficacy and the loss as a result of toxicity have never been balanced to a degree which could establish one of the aforementioned drugs as the generally accepted standard treatment to replace estrogens. GnRH analogs may prove to be the first nontoxic medical castration measure applicable for general use in the future."
|
VI. Mechanisms of Medical Castration by GnRH Agonists
|
|---|
It was only 3 yr after our observation of the castration effect of GnRH agonists in men (30) that we discovered that the biological activity of LH was progressively lost during long-term treatment of prostate cancer patients with GnRH agonists (36, 37), thus explaining the castration effect of GnRH agonists in men. In fact, in the presence of a greater than 95% inhibition of serum testosterone and DHT levels, serum LH measured by RIA can remain normal or be only slightly decreased (31). Because we had previously found a discrepancy between serum LH measured by RIA and by bioassay in rhesus monkeys treated with a high dose of a GnRH agonist (27), we performed a similar study in men. We then observed that although the values of serum LH measured by RIA and bioassay (mouse Leydig cell assay) varied in a parallel manner during the first 2 wk of treatment, a progressive and marked loss of LH bioactivity was measured at later time intervals, whereas LH measured by RIA was much less reduced or sometimes unchanged. Thus, after 3 months of treatment, LH bioactivity was reduced to about 5% of control, whereas the radioimmunoassayable LH was reduced by only 40–50% (Fig. 7) (36). These data indicated that the loss of LH bioactivity, rather than testicular desensitization, is the major factor responsible for the complete inhibition of testicular steroidogenesis that occurs after 2–3 wk of treatment with GnRH agonists in men. In fact, in men, GnRH agonists achieve a medical hypophysectomy selective for gonadotrophs.
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 7. Effect of 1-month treatment with the GnRH agonist buserelin (500 µg/d, sc) and the pure antiandrogen RU-23908 (Anandron; 100 mg, three times daily, orally) on serum LH measured by RIA and by the mouse Leydig cell bioassay. Also shown is the effect on serum testosterone concentration in patients with advanced cancer of the prostate (116 ). TESTO, Testosterone.
|
|
|
VII. Transient Stimulation of Serum Testosterone Completely Neutralized by a Pure Antiandrogen: the Potential Risk of Flare with a GnRH Agonist Is Thus Completely Removed
|
|---|
A significant but temporary difference between medical castration achieved with a GnRH agonist and surgical castration or orchiectomy is the elevation of the blood levels of testosterone and DHT that lasts for 7 to 10 d at the start of treatment with a GnRH agonist (Fig. 8). Such a transient increase in serum testosterone lasting for a few days had already been seen in the first patient treated with a GnRH agonist (30). This transient elevation of serum testosterone is followed by a gradual decrease to castration levels, which are reached at 2 to 3 wk. On the other hand, it is well known that serum androgens decrease to castration levels within a few hours after orchiectomy. Such a difference should be taken into account when treating such a highly androgen-sensitive disease as prostate cancer.
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 8. Changes in serum PAP (dotted line) and testosterone (solid line) levels during the first month of treatment in four previously untreated advanced prostate cancer patients (A–D) receiving the combined administration of the GnRH agonist buserelin and the pure antiandrogen Anandron. Note the rapid and marked decrease of serum PAP in the presence of elevated serum testosterone levels, thus showing the efficacy of the antiandrogen to neutralize the effect of the transient rise in serum androgens (39 )
|
|
We therefore concluded that patients with advanced disease should not be treated with a GnRH agonist alone but always in combination with a pure antiandrogen administered concomitantly from the start of treatment. We reasoned that although the GnRH agonist-induced secretion of androgens observed during the first days of treatment with a GnRH agonist alone carries the unacceptable risk of disease flare, the addition of a pure antiandrogen should completely eliminate this risk. In fact, in our first patients who received a pure antiandrogen (flutamide or nilutamide) in association with a GnRH agonist (32, 38, 39, 40, 41, 42), serum prostatic acid phosphatase (PAP) decreased rapidly at a time when serum androgens were maximally increased (32, 38, 39, 40, 41, 42) (Fig. 8). These early data clearly indicated that the addition of a pure antiandrogen completely eliminated the risk of disease flare associated with the use of the otherwise exceptionally efficient and well-tolerated GnRH agonists in prostate cancer patients.
|
VIII. Monotherapy with a GnRH Agonist Decreases Cancer Death by at Least One Third in Localized Prostate Cancer
|
|---|
As mentioned above, the exceptionally well-tolerated medical castration achieved with GnRH agonists (30) rapidly opened the way to a much more acceptable hormonal therapy of prostate cancer, especially for localized disease where a well-tolerated treatment is essential to be accepted by patients who need long-term administration. In fact, only GnRH agonists could permit studies in localized prostate cancer, because orchiectomy is very difficult to accept by patients having no symptom or sign of cancer.
Most importantly, prospective randomized trials could then be performed much more easily in patients with localized or locally advanced disease. All these trials have demonstrated that an important prolongation of life or a reduction of death from prostate cancer is achieved in localized and locally advanced prostate cancer patients treated with androgen blockade (Table 1). When considering deaths from prostate cancer at 4.5 to 9.3 yr of follow-up, decreases in deaths from prostate cancer ranging from 37.5 to 81% were observed in six studies whereas, in the seventh study (43), a 45% decrease in overall deaths was observed and no data are available on cancer-specific deaths.
In the first published study, namely the European Organization for Research and Treatment of Cancer (EORTC) trial performed in 415 stage T3 patients, overall survival at 5 yr was increased from 62% in the group of patients who received radiation therapy alone to 79% (45% difference) in the group of patients who received androgen blockade using a GnRH agonist for 3 yr and an antiandrogen for 1 month in addition to radiotherapy (44). In that study, deaths from prostate cancer at 5 yr were decreased by 77% (26 deaths in the radiotherapy alone group and six deaths in the combined treatment group) by androgen blockade (Table 1).
In Radiation Therapy Oncology Group (RTOG) trial 85-31 (45), a 37.5% decrease in cancer-specific deaths was observed at 4.5 yr of median follow-up in the subgroup of 276 patients with Gleason score 8 to 10 who received androgen blockade (GnRH agonist indefinitely or until progression) in association with radiotherapy (66%) vs. radiotherapy alone (55%). In fact, when considering prostate cancer deaths, deaths decreased from 40 in the group who received hormone therapy late to 25 deaths in the group of patients who received androgen blockade with an LHRH agonist continuously starting at time of radiotherapy for a difference of 37.5% in favor of early androgen blockade.
The results of another recent study are particularly interesting. In that study, of 98 men who had stage T2 prostate cancer at diagnosis but who were found to have pelvic lymph node metastases at radical prostatectomy, 47 had immediate hormonal therapy, whereas 51 were followed until progression (46). After a median follow-up of 7.1 yr, 16 have died from prostate cancer in the deferred treatment group compared with only three in the immediate androgen blockade group, for an 81% decrease in deaths from prostate cancer (P = 0.001) for men who had immediate hormonal therapy. On the other hand, the 67% decrease in the incidence of death from prostate cancer observed during the first 8 yr of our randomized and prospective study on prostate cancer screening can only be due to the treatments used as well as the early stage of the disease at time of treatment initiation (47).
In another study, a 59% decrease in cancer-specific death has been found in patients with a 8–10 Gleason score who had long-term androgen blockade (48). In fact, in Gleason 8–10 tumors, 29 patients died of prostate cancer in the short-term androgen deprivation group, whereas 12 died from prostate cancer in the long-term androgen deprivation group. All patients received Zoladex and flutamide 2 months before and 2 months during radiation therapy and were randomized to no further therapy (short-term androgen deprivation) or 24 months of additional Zoladex alone (long-term androgen deprivation). In a study of 91 patients with clinically localized prostate cancer, after a median follow-up of 9.3 yr, cancer-specific mortality was decreased by 39% from 44% in patients who had external beam radiotherapy alone to 27% in the group of patients who had combined orchiectomy and radiotherapy (P = 0.06) while total mortality was decreased by 38% from 61% to 38% (P = –0.02). More recently, in a study of 201 patients with localized prostate cancer, the 5-yr overall survival was improved from 78% in the radiotherapy alone group to 88% in the patients who received radiotherapy plus combined androgen blockade (CAB) with flutamide as adjuvant for only 6 months (45% improvement; P = 0.04) (43).
It seems appropriate to indicate the conclusion of Drs. Peto and Dalesio (49) after their recent metaanalysis of androgen blockade in localized prostate cancer: "Prostate cancer is usually treated with surgery or radiation, but a few cancer cells may remain and cause an often-fatal recurrence. Since the mid-1980s, oncologists have increasingly followed up with either surgical removal of the testes, or with newer antihormone drugs. Seventy-four percent of men who received hormone therapy were still alive 10 yr later, compared with 62% of those who did not" (112). Thus, in the metaanalysis of several studies (5000 men), hormonal treatment given immediately vs. waiting until disease progressed, it was concluded that the risk of dying from prostate cancer within 10 yr dropped by one third (49). It should be mentioned that the one third decrease in the risk of dying from prostate cancer is not the comparison between androgen blockade and placebo but between early vs. late androgen blockade, thus indicating that the risk of dying from prostate cancer is in fact reduced by more than one third by administering androgen blockade. These results led Dr. Peto to the following conclusion: "Hormone treatment as a whole works much better than previously thought." It should be mentioned here that the remarkable results mentioned above have usually been obtained using only partial blockade of androgens or monotherapy.
|
IX. Major Importance of Adding a Pure Antiandrogen to the GnRH Agonist (or Surgical Castration)
|
|---|
A. Two equally important sources of androgens are present in men
An important advance in our understanding of the biology and endocrinology of prostate cancer and its major impact on cancer treatment is the observation that humans and some other primates are unique among animal species in having adrenals that secrete large amounts of the inactive precursor steroids dehydroepiandrosterone (DHEA), its sulfate DHEA-S, and some androstenedione (4-dione), which are converted into potent androgens in a large series of peripheral tissues, including the prostate (Fig. 9). In fact, the plasma concentration of DHEA-S secreted by the adrenals in adult men is 100–500 times higher than that of testosterone (32), the main secretory product of the testicles. Such high circulating levels of DHEA-S (and also DHEA) provide high amounts of the prehormones or precursors required for conversion into active androgens in the prostate as well as in other peripheral intracrine tissues.
View larger version (31K):
[in this window]
[in a new window]
|
FIG. 9. Intracrine activity of the human prostate or biosynthetic steps involved in the formation of the active androgen DHT from testicular testosterone as well as from the adrenal precursors DHEA, DHEA-S, and 4-dione in human prostatic tissue. 3ß-HSD, 3ß-hydroxysteroid dehydrogenase/ 5-4-isomerase; CRF, corticotropin-releasing factor. The widths of the arrows indicate the relative importance of the sources of DHT in the human prostate: 60% originating from the testes and 40% from the adrenals in 65-yr-old men. The testis secretes testosterone (T), which is transformed into the more potent androgen DHT by 5 -reductase in the prostate. Instead of secreting T or DHT directly, the adrenal secretes very large amounts of DHEA and DHEA-S, with some 4-dione, which are transported in the blood to the prostate and other peripheral tissues. These inactive precursors are then transformed locally into the active androgens T and DHT. The enzymes DHEA sulfatase, 3ß-HSD, 17ß-HSD, and 5-reductase are all present in the prostatic cells, thus providing 40% of total DHT in this tissue (132 ). A (LHRH agonist) and B (antiandrogen flutamide) are the two components of CAB.
|
|
The local synthesis of active steroids in peripheral target tissues has been named intracrinology (50, 51, 52). The active androgens made locally in the prostate exert their action by interacting with the androgen receptor in the same cells where their synthesis takes place without being released in significant amounts in the extracellular environment or the general circulation. Contrary to the previous belief that the testes are responsible for 90–95% of total androgen production in men (as could be inferred from the 90–95% decrease in serum testosterone observed after castration), it is now well demonstrated that the prostatic tissue efficiently transforms the inactive steroid precursors DHEA-S, DHEA, and 4-dione into the active androgens testosterone and DHT locally in peripheral tissues, including the prostate, without significant release of the active androgens in the circulation. In fact, the prostate makes its own androgens at a level comparable to the androgens of testicular origin.
The human steroidogenic pathway is composed of 15 main steps transforming cholesterol into the five classes of active hormonal steroids, namely androgens (testosterone and DHT), estrogens (estradiol and 5-androstenediol), progesterone, glucocorticoids (cortisol or corticosterone), and mineralocorticoids (aldosterone), as well as their inactive sulfated and glucuronosylated derivatives (Fig. 10). So far, 30 human genes have been found to encode enzymes of the steroidogenic pathway, a significant proportion of these genes have been isolated, cloned, and characterized in our laboratory (53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66).
The formation and inactivation of androgens and estrogens is under fine control by a series of enzymes that reduce (activate) or oxidize (inactivate) the C17 position of C19 (androgens) or C18 (estrogens) steroids. At the present time, 13 human 17ß-hydroxysteroid dehydrogenases (17ß-HSDs) are known and can be divided into two functional groups. Those of the first group are represented by types 1, 3, 5, 7, 12, and 13 17ß-HSDs that prefer reduced nicotinamide adenine dinucleotide phosphate as cofactor and catalyze the reduction of 17-ketosteroids (inactive form) into 17ß-hydroxysteroids (active form) (Fig. 10). On the other hand, the members of the second group represented by types 2, 4, 6, 8, 9, 10, and 11 17ß-HSDs function as dehydrogenases or inactivating enzymes. These enzymes prefer nicotinamide adenine dinucleotide as cofactor and catalyze the oxidation of 17ß-hydroxysteroids.
Such a large number of steroidogenic enzyme isoforms (Fig. 10) allows tissue-specific expression and thus local control of steroid formation and inactivation according to the local needs dictated by intracellular and extracellular signals. These control mechanisms permit local regulation of steroid action independently from the circulating levels of these steroids.
B. Combined androgen blockade in advanced disease
The first treatment shown to prolong life in prostate cancer is the combination of a GnRH agonist to block androgen secretion by the testes in association with an effective dose of a pure antiandrogen such as flutamide, nilutamide, or bicalutamide (32, 38). These antiandrogens (sometimes called nonsteroidal antiandrogens) block the action of the androgens produced locally in the prostate by interfering at the level of the androgen receptor.
An interesting observation is that the first demonstration of the benefits of CAB on survival (32, 38) has been achieved in the most difficult group of patients to treat, namely those suffering from metastatic or advanced disease. These data have been obtained with flutamide and nilutamide. Although, in principle, the clinical results should be similar for bicalutamide, the two antiandrogens flutamide and nilutamide were demonstrated in prospective and randomized studies to prolong life, to increase the number of complete and partial responses, to delay progression, and to provide better pain control (thus improving quality of life) in metastatic prostate cancer when added to surgical or medical castration compared with castration alone (33, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78). In the first large-scale randomized study, patients who were treated with flutamide and the GnRH agonist Lupron lived on average 7.3 months longer than those who received Lupron plus placebo (67).
Analysis of all the studies performed with flutamide and nilutamide associated with medical or surgical castration compared with castration plus placebo shows that overall survival (deaths from all causes) is increased by an average of 3–6 months after the addition of a pure antiandrogen (33, 67, 68, 69, 70, 71, 72, 73, 74) (Fig. 11). Because about 50% of patients in that age group die from causes other than prostate cancer, this 3- to 6-month difference in overall survival corresponds to an average of 6–12 months of life gained when cancer-specific survival is calculated. These additional months, or sometimes, years of life are obtained by simply adding a pure antiandrogen (flutamide, nilutamide, or bicalutamide at a proper dose) to castration. Considering that such statistically significant benefits on survival are obtained, even at the very advanced stage of metastatic disease, these data demonstrate, as mentioned earlier, the particularly high level of sensitivity of prostate cancer to androgen deprivation.
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 11. Summary of metaanalyses comparing CAB (combination of medical or surgical castration alone) vs. medical or surgical castration associated with a pure antiandrogen or nonsteroidal antiandrogen (NSAA), namely flutamide or nilutamide. [Adapted from Klotz (75 ).]
|
|
As illustrated in Fig. 11, all the metaanalyses of all the data have shown significant (2p < 0.05) or highly significant (2p < 0.01) advantages of CAB vs. castration alone in advanced prostate cancer (33, 70, 73, 74, 75, 78). However, when the studies providing the most rigorous data are analyzed (70), a 20% advantage in overall survival is observed.
It is of interest to mention the first results of a Japanese study (79) showing improved prostate-specific antigen (PSA) normalization (79.4 vs. 38.6%) at 12 wk and time to treatment failure (96.1 vs. 67.7 wk) in advanced prostate cancer patients who received the combination of a GnRH agonist and 80 mg/d bicalutamide vs. the GnRH agonist and placebo. The risk of progression during follow-up was thus reduced by 54% in the CAB group compared with chemotherapy. The study is not sufficiently mature to calculate the effect on survival, but the early effects observed are in line with previous studies.
Concerning the costs of treatment, as recently published by Aprikian et al. (76), the cost per month of prolonged survival in prostate cancer achieved with the simple addition of a nonsteroidal antiandrogen to castration (GnRH agonist or orchiectomy) is 50% that of vinorelbine for lung cancer, 10% of the cost of irinotecan for colon cancer, and 10% of the cost of trastuzumab for breast cancer. Moreover, the nonsteroidal antiandrogens have minimal toxicity, whereas vinorelbine and irinotecan are associated with severe grade 3 and 4 clinical toxicities, and trastuzumab has cardiac side effects when associated with anthracyclines. As Klotz (77) said, "We should embrace the modest survival benefit of combined androgen blockade in advanced prostate cancer and offer it to the appropriate patients."
In addition to the prolongation of survival, all the studies have shown that the decrease in bone pain is more rapid and more complete and that progression of the cancer is delayed, thus improving quality of life, when CAB is used compared with monotherapy. Moreover, this is the only treatment shown to prolong life in advanced disease. There is no other choice. It should also be realized that there is no treatment of similarly advanced cancers that provides 3 to 6 months of prolongation of life or 6 to 12 months of cancer-specific survival with such a good quality of life. To the living population of males in the United States, where 3 million are expected to die from prostate cancer, six additional months of life correspond to the addition of 1.5 million years of life, whereas 12 additional months correspond to 3.0 million years of life.
|
X. High Probability of Cure of Localized Prostate Cancer by Treatment with Combined Androgen Blockade
|
|---|
Despite the important advance observed with monotherapy (GnRH agonists) in localized prostate cancer, namely at least a one third reduction in deaths from prostate cancer (49), can we achieve better results? Based upon the observation that 50% of androgens are left in the prostate after castration alone, it is reasonable to suggest that superior results can be achieved with the combination of a GnRH agonist and a pure antiandrogen. There were already data indicating that patients with minimal metastatic disease derive greater benefits than those with extensive metastatic disease (67, 72, 80).
Using CAB in localized and locally advanced disease, the evidence obtained even indicates that long-term control or cure of the disease can be obtained in 90% of patients (81). In fact, although almost all studies performed so far in localized prostate cancer have used monotherapy (medical or surgical castration) (43, 44, 45, 46, 48, 82), there are good scientific reasons to believe that much better results can be expected with CAB (32, 33, 70, 73, 83, 84). The direct effect of CAB on the volume of prostate cancer is well illustrated in Fig. 12, which shows the results of two detailed anatomopathological studies performed in patients who had radical prostatectomy after 3 and 6 months of CAB, respectively (85, 86).
View larger version (25K):
[in this window]
[in a new window]
|
FIG. 12. Inhibitory effect of 3 and 6 months of treatment with the combination of a GnRH agonist and flutamide (250 mg, three times a day) on the volume of the cancer measured on the radical prostatectomy specimens (85 86 ). A, A first study performed in 94 patients has shown a 44% decrease in tumor volume in patients treated for 3 months with an LHRH agonist and flutamide compared with radical prostatectomy alone (85 ). B, In a second study performed in 40 patients, the same treatment for 6 months caused an additional 60% decrease in tumor volume compared with 3 months of treatment (86 ).
|
|
Because we already had obtained evidence for the high efficacy of long-term and continuous CAB in localized prostate cancer (87), it was felt important to examine the long-term outcome of these patients as assessed by biochemical failure (PSA progression) after cessation of continuous CAB previously administered for periods up to 11.3 yr. The effect of CAB on long-term control or possible cure of prostate cancer was thus evaluated by the absence of biochemical failure or the absence of PSA rise for at least 5 yr after cessation of continuous treatment. A total of 57 patients with initial localized or locally advanced disease thus received CAB for periods ranging from 1 to 11 yr. CAB was then discontinued, and the patients were followed for a minimum of 5 yr. Among the 20 patients with stage T2-T3 cancer initially who stopped treatment after continuous CAB for more than 6.5 yr, only two PSA rises occurred for a nonfailure rate of 90% (Fig. 13). For the 11 patients who had received CAB for 3.5 to 6.5 yr, the nonfailure rate was only 36%. It is of major interest that serum PSA increased within 1 yr after cessation of CAB in all 11 patients with stage B2/T2 cancer initially treated with CAB for only 1 yr, thus showing that active cancer remained present after short-term androgen blockade limited to 1 yr despite undetectable PSA levels. Most importantly, in all patients who had biochemical failure after stopping CAB, serum PSA rapidly decreased again to undetectable levels soon after CAB was restarted, and PSA has remained at such low levels afterward. Of these 57 patients, only one patient had died of prostate cancer at last follow-up (81).
View larger version (23K):
[in this window]
[in a new window]
|
FIG. 13. Effect of treatment duration of localized prostate cancer with continuous CAB on the probability of long-term control or cure of the disease as determined by no recurrence of PSA rise for at least 5 yr after CAB cessation. Point at 4.75 yr of treatment (33%) refers to three patients treated with CAB for 3.5 to 5.0 yr and followed up for at least 5 yr; point at 5.75 yr refers to eight patients treated continuously with CAB for 5.0 to 6.5 yr before treatment cessation. Point at 8.25 yr refers to eight patients treated continuously for 6.5 to 9.0 yr, and point at 11 yr refers to 12 patients treated for 10 to 11.7 yr with continuous CAB before stopping treatment. All patients were followed up for at least 5 yr after continuous CAB or until PSA rise. Only one patient died of prostate cancer, and 18 have died of other causes.
|
|
These are remarkable results observed in patients with localized prostate cancer. Treatment, however, must be continuous, without interruption, and should last for many years. It is important to mention that the major survival benefits observed after androgen blockade, even in localized or locally advanced disease, are always associated with long term (many years of noninterrupted treatment) (44, 46, 47, 81). In fact, an important observation made is that when PSA increased after cessation of treatment, administration of CAB was successful in all cases in decreasing PSA to undetectable levels again, thus showing that even after a long duration of treatment, resistance to CAB had not developed. In fact, resistance to CAB is the common finding in prostate cancer metastasized to the bone, whereas it does not occur for the cancer localized in the prostate or in the prostatic area.
The present results obtained in prostate cancer patients diagnosed with localized disease and treated continuously for many years with CAB are not too different from the results that we have recently obtained with human breast tumor xenografts in nude mice where complete estrogen blockade achieved with a highly potent antiestrogen led to the disappearance or cure of the tumors in 61% of cases within a few months (88). In fact, in both breast and prostate cancer, when the estrogens in breast cancer and the androgens in prostate cancer are blocked efficiently, cure of the disease can be achieved with hormonal therapy.
As mentioned above, the success of therapy, however, requires long-term and continuous treatment before complete apoptosis or total cell death is achieved. Such results clearly indicate that intermittent androgen blockade should remain experimental and should not be used outside clinical trials. Breast and prostate cancers have many characteristics in common, and much can be learned from looking at the results obtained in each of them. In fact, when we consider the biology of these two cancers, there are many common features, especially the high level of sensitivity to hormones.
With the knowledge gained by the analysis of the data summarized above, it seems reasonable to suggest that the minimal duration of continuous CAB in localized prostate cancer should be at least 6 yr, thus providing at least a 50% probability of long-term control or probable cure of the cancer. Most importantly, the present data indicate that possible cure of the disease can be obtained in most patients with localized prostate cancer treated continuously with CAB for more than 8 yr, thus raising hopes for the successful treatment of patients who fail after surgery, radiotherapy, or brachytherapy where no or minimally effective alternative therapeutic approach exists.
|
XI. Neoadjuvant Therapy
|
|---|
The major problem with radical prostatectomy as well as with radiotherapy and brachytherapy in clinically localized prostate cancer is that the cancer is not limited to the prostate in approximately 50% of cases, thus eliminating the possibility of a cure. To increase the probability of organ-confined disease at surgery, we have performed the first prospective and randomized study of the effect of CAB (GnRH agonist plus flutamide) given for 3 months before radical prostatectomy in patients found to be candidates for prostatectomy (89, 90, 91). As confirmed by many other studies, neoadjuvant CAB increases the rate of organ-confined disease, and decreases the rate of positive margins and invasion of the seminal vesicles (77, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98).
At the time these studies were performed, we did not know that CAB would truly cause efficient and continuous regression of the cancer and we did not know that resistance to treatment would not develop. For these reasons, we chose a period of only 3 months of neoadjuvant treatment with CAB before surgery (89, 91). In fact, we learned much later (87) that much longer periods of treatment are required to kill all cancer cells by apoptosis, namely 8 to 10 yr of continuous administration of CAB (87). It is thus not too surprising that the benefits of neoadjuvant androgen blockade (89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99), especially monotherapy, could not translate into significant prolongation of survival, although this should be the case for longer durations of neoadjuvant hormone therapy. In fact, with the knowledge gained since the start of that first randomized trial performed in 1988 (90), it has become clear that 3 months of neoadjuvant CAB is much too short to expect significant benefits on survival.
We have also performed a randomized study comparing external beam radiation therapy (EBRT) alone, 3 months of neoadjuvant CAB (LHRH agonist + flutamide), and a third group of stage T2 patients who received CAB for 3 months before, 1.5 months during, and 6 months after EBRT (total of 10.5 months of CAB) (100). Control follow-up transrectal ultrasound-guided biopsies at 24 months have shown that 60% of control patients (EBRT alone) had positive biopsies compared with 21 and 5% in patients who had 3 and 10.5 months of CAB, respectively (100).
In fact, except for the potential advantages of reducing the volume of the prostate, especially the volume of the cancer (Fig. 12) as well as the reduction of the malignancy of the cancer cells that are likely to escape into the blood stream at time of surgery, thus possibly causing distant metastases, CAB could well be reserved as adjuvant therapy to be used immediately at time of rising PSA in cases of nonconfined disease at surgery. The same applies to radiation therapy and brachytherapy. It is always important to remember that long-term and continuous treatment is needed to kill cancer cells.
|
XII. Secondary Effects of Androgen Blockade
|
|---|
Top
Abstract
I. Introduction
