Antiandrogen Withdrawal Syndrome Associated with Prostate Cancer Therapies
- 70 Downloads
The antiandrogen withdrawal syndrome is a well established phenomenon in prostate cancer. It is widely accepted that a subset of patients will benefit from the withdrawal of antiandrogen or steroidal hormone from hormonal therapy, exhibiting decreasing prostate-specific antigen (PSA) values and clinical improvement. The pathophysiology of antiandrogen withdrawal syndrome is not completely understood, although androgen receptor gene mutations seem to be the likely explanation. Currently, it is not possible to identify the subset of patients whose tumours will respond to antiandrogen or steroid withdrawal. Tumours that will respond may be classified as androgen-independent and hormone-sensitive tumours as opposed to androgen-independent and hormone-insensitive tumours that do not respond. Patients who respond to antiandrogen withdrawal experience approximately 6 months with improved quality of life; however, it is unknown if this translates into prolonged survival. At the very least, antiandrogen withdrawal offers a therapeutic modality that is not associated with adverse effects and improves quality of life even if only for a very limited time.
Recent reports suggest that adding a secondary hormonal therapy such as aminoglutethimide, ketoconazole or steroidal hormones may enhance the response rate and prolong response time to the antiandrogen withdrawal syndrome. However, unless there is proof that this secondary hormonal manipulation prolongs survival, maintenance of quality of life is mandatory because of the possible adverse effects from these potent drugs.
The fact that about 30% of patients will respond to antiandrogen or steroid withdrawal in hormone refractory prostate cancer must be taken into account in clinical trials of new cytotoxic agents which have been and will be conducted. Cessation of flutamide for at least 4 weeks and, in the case of bicalutamide, even 8 weeks, is mandatory before antiandrogen withdrawal syndrome can be excluded as the cause of decreasing PSA values.
The antiandrogen withdrawal syndrome offers another piece of the puzzle of prostatic carcinoma, but at the same time it demonstrates how different advanced prostate cancer cells may react to therapeutic strategies and, therefore, hormone refractory prostate cancer remains a difficult challenge which must be solved in the future.
Hormonal therapy and antiandrogens are widely used for the treatment of prostate cancer. Several therapeutic strategies, such as antiandrogen monotherapy or total androgen blockade using a gonadotropin releasing hormone (GnRH) agonist or bilateral orchiectomy in combination with an antiandrogen, are used in the noncurative treatment of prostate cancer.
Hormonal therapy of prostate cancer was first described by Huggins and Hodges in 1941. First results at that time suggested that hormonal therapy might be curative, but it turned out that after an initial response in approximately 80% of patients with prostate cancer to hormonal therapy, about 40% of patients will develop tumour progression within 3 years of therapy. 50% of these patients will die within 12 months.[3, 4, 5] Treatment of hormone-resistant prostate cancer is a great challenge for urologists and medical oncologists because currently there is no therapy available which has been shown to prolong survival in these patients.
Prostate specific antigen (PSA), since its isolation and description in seminal fluid by Wang and et al., is the most clinically useful tumour marker for monitoring response to therapy. The first apparent sign of tumour progression under therapy is increasing PSA values. Clinical progression of prostate cancer will follow rising PSA values in most of patients who receive treatment. On the other hand, it is widely accepted that declining PSA values under therapy are a surrogate marker of tumour response to therapy. The widespread use of PSA determinations for monitoring prostate cancer disease status lead to the discovery of a new syndrome — the antiandrogen withdrawal syndrome.
New reports of secondary hormonal manipulation demonstrated decreasing PSA values after withdrawal of the antiandrogen in a significant subset of patients with hormone refractory disease. This observation, which will be discussed, is called ‘antiandrogen withdrawal syndrome’.
‘Antiandrogen withdrawal syndrome’ is defined as a subjective and/or objective improvement following discontinuation of the antiandrogen in patients with rising PSA values under hormonal therapy, which includes the administration of an antiandrogen (also called maximum androgen blockade).
The first evidence of the molecular basis of the antiandrogen withdrawal syndrome was demonstrated by Wilding et al. in 1989. They proved that flutamide in vitro had a paradoxical growth stimulatory effect on the prostate cancer cell line LNCaP. In addition, flutamide was unable to inhibit the stimulatory effect of dihydrotestosterone in this hormone-sensitive prostate cancer cell line. In 1990, a mutation in the ligand binding domain of the androgen receptor in the LNCaP cell line was identified as the reason for the growth stimulatory effect of flutamide. This mutation is not responsible for the observed growth stimulation of bicalutamide or nilutamide, but since then several other mutations have been discovered, not only in prostate cancer cell lines but also in tumour specimens.[21, 22, 23]
The cornerstone to understanding hormone resistance and learning how to overcome hormonal resistance in prostate cancer seems to be the androgen receptor (AR). This receptor belongs to the steroid receptor superfamily and is a potent intracellular signal transducer. The AR regulates prostatic growth and function. The AR gene is located on the X-chromosome and is comprised of 3 domains: the steroid binding domain at the C terminus; the central DNA binding domain; and the transactivation domain at the N terminus. The presence of AR protein as shown by immunohisto-chemistry does not correlate with response to treatment and disease outcome. Also, the AR protein can still be detected in hormone refractory prostatic cancer. Therefore, changes in AR receptor function may be involved in the differing responses of patients to hormonal therapy by modulating the signalling pathway starting from the ligand binding affinity of androgens up to the activation of the transcription factor functions of the AR. One of the first observations of altered AR function was the description of the point mutation of codon 877 in the AR gene in the cell line LNCaP by Veldschole et al. This point mutations leads to an exchange of threonine to alanine. This exchange of a single amino acid leads to stimulation of the AR receptor by flutamide but not by bicalutamide or nilutamide. Transfection studies of the mutated AR gene demonstrate increased AR signalling activity explaining growth stimulatory effects in LNCaP cells by flutamide. Since then several mutations of the AR gene in prostate cancer have been reported. Mostly point mutations have been found and interestingly most mutations are located in the ligand binding domain. Suzuki et al. were the first to report the codon 877 mutation in relation to the antiandrogen withdrawal syndrome. The LNCaP mutation of codon 877 was found in several patients, whereas all other reported mutations were detected only once. The human AR gene mutation database is available on the internet and lists all reports of AR gene mutations currently detected. This database shows the increasing number entries for prostate cancer (currently 60), as well as the LNCaP mutation. The results of mutation analysis show that point mutations may be a possible explanation for the androgen withdrawal syndrome. However, there is no single mutation responsible in all patients, and there is a significant subset of patients demonstrating androgen withdrawal syndrome but only a minor subset of patients with point mutations of the AR receptor.
Therefore, an alternative aetiology could be the amplification of the AR gene. In support of this idea, Visakorpi et al. demonstrated that although no amplification was found in primary prostate cancer, up to 30% of investigated samples revealed AR gene amplification in hormone-resistant prostate cancer.
There is evidence from molecular biology analyses as to how the antiandrogen withdrawal syndrome can be explained, but currently, the data available do not sufficiently explain this syndrome in all patients.
3. Results of Antiandrogen Withdrawal in Hormone-Resistant Prostate Cancer
Androgen-independent prostate cancer develops under androgen ablation therapy. The median time to hormone resistance is usually 2 to 3 years. An increasing PSA value under endocrine therapy is the first sign of hormone refractory prostate cancer (HRPC). In this situation Scher et al. proposed a classification of 2 distinct different types of HRPC. First, androgen-independent and hormone-sensitive tumours. These tumours proliferate despite castrate levels of androgens. As a result of secondary hormonal manipulations these tumours will undergo apoptosis (programmed cell death), demonstrating their hormone-sensitive stage. The second type of HRPC is called hormone-independent, or androgen-independent and hormone-insensitive tumours. These tumours due not respond to secondary hormone manipulation or progress after second-line hormonal therapy. As the antiandrogen withdrawal syndrome belongs to the secondary hormonal therapies, it is obvious that not all patients will improve after cessation of antiandrogen therapy because a subset of patients will belong to the hormone-independent group. But a significant subset of patients, who belong to the androgen-independent but hormone-sensitive group, will experience falling PSA values after discontinuation of antiandrogen therapy.
3.1 Rate of Response Following Antiandrogen Withdrawal
3.2 Time to Response Following Antiandrogen Withdrawal
An important point to assess the effect of antiandrogen withdrawal syndrome is the time to response. Therefore, patients with increasing PSA values after discontinuation of the antiandrogen can be treated with alternative therapies if they have not shown a biochemical response within the expected time frame.
After cessation of flutamide falling PSA values occur within 4 to 6 weeks or earlier after discontinuation. If bicalutamide was used as antiandrogen the time to response seems to be prolonged compared with flutamide because of the longer halflife of bicalutamide. An average of 4 to 8 weeks is needed before declining PSA values are noted.
Because only case reports exist of other antiandrogens and steroidal hormones the expected time to response can not be determined, but the published data suggest that an effect can be expected within 2 months after discontinuation.
On the other hand, if no decline in PSA value is noted after 8 weeks following antiandrogen withdrawal and in that time 2 consecutive increasing PSA values were obtained, these patients are hormone-independent and can be treated by alternative treatment modalities, i.e. chemotherapy.
3.3 Duration of Response to Antiandrogen Withdrawal
Response to the antiandrogen withdrawal syndrome is of limited duration. Patients who demonstrate falling PSA values seem to benefit from simple discontinuation of the antiandrogen for about 3.3 to 14.5 months for flutamide treated patients and for about 6 months for bicalutamide treated patients (table II). Keeping in mind that currently there is no alternative therapy which prolongs survival in patients with hormone-resistant prostate cancer, a therapy option that is devoid of adverse effects and that offers patients 6 months of life with improved quality of life, seems to be a significant advantage for patients with prostate cancer. Avery important factor in the improved quality of life is the psychological advantage for patients of the declining PSA values. Since PSA was introduced as a parameter to control for disease status, the quality of life in patients with prostate cancer depends significantly on the rate of decline of in PSA values. Even clinically asymptomatic patients demonstrate impaired quality of life if their PSA values are increasing.
3.4 Survival Following Antiandrogen Withdrawal
The question is, will antiandrogen withdrawal syndrome translate into prolonged survival in these patients? Mean survival in patients with hormone-resistant prostate cancer is about 12 months, therefore, prolonged survival, even if it will be only a few months, seems to be a significant gain for these patients. Small and Srinivas demonstrated a nonsignificant increase in total survival time from starting hormonal therapy for responders to antiandrogen withdrawal of 44.5 months compared with 35 months for nonresponders. Comparing survival time after initiation of the antiandrogen withdrawal syndrome they noted 13 months for responders and 12 months for nonresponders. Whether responders and nonresponders to the antiandrogen withdrawal show identical survival times needs to be proven in further studies, which are ongoing. The study of Small and Srinivas suggests that patients who respond to antiandrogen withdrawal belong to a group of patients with a better prognosis, but this fact does not translate into prolonged survival after tumours become androgen-resistant but hormone-sensitive.
3.5 Defining Prognostic Criteria of Response to Antiandrogen Withdrawal
Because only a subset of patients will benefit from antiandrogen withdrawal it would be helpful to be able to stratify patients with hormone refractory cancer in order to establish prognostic parameters to identify patients who will respond to antiandrogen withdrawal, as well as nonresponders who could receive alternative therapy earlier on. There have been several attempts to identify which type of hormonal therapy may predict response to antiandrogen withdrawal. Scher et al. and Herrada et al. suggest that patients who have low androgen levels (following orchiectomy or GnRH agonist therapy) combined with an antiandrogen are more likely to respond than patients who received antiandrogen monotherapy. Contrary to the results obtained by Small and Srinivias, Sartor et al. could not find a difference in the response rate with regard to which type of hormonal therapy was used.
Several other parameters have been investigated, such as PSA value, alkaline phosphatase level, haemoglobin level, lactate dehydrogenase level, hepatic transaminase levels, histological grade, extent of bone disease, soft tissue extent, age and performance status. None of these parameters was found commonly to correlate with the response rate to antiandrogen withdrawal in these studies.[17,39,43] Furuya et al. defined a subgroup of patients who experienced normalisation of their PSA values within 3 months after the start of initial hormonal therapy who showed a higher response rate to the antiandrogen withdrawal. Several groups[32,35,39,42] have shown that the duration of antiandrogen exposure is related to response after discontinuation of the antiandrogen. It seems that patients who respond to hormonal therapy quickly and over a prolonged period represent the subgroup of patients with hormone-sensitive tumours, who will benefit from secondary hormonal manipulation. On the other hand patients whose PSA values do not normalise within 3 months and experience early rising PSA values probably will not respond well to secondary hormonal therapy strategies.
4. Impact of Secondary Hormone Treatment Strategies on the Antiandrogen Withdrawal Syndrome
Knowing that a subset of patients may be classified as having androgen-independent but hormone-sensitive tumours, these patients may benefit not only from cessation of the antiandrogen but also from secondary hormonal manipulation. Secondary hormonal therapy is not a well described treatment modality. Some authors use this term to describe simple antiandrogen or steroid withdrawal, while others use this term to describe second-line antiandrogen treatment or inhibition of adrenal steroido-genesis and treatment with alternative steroids, i.e. glucocorticoids or progestins. Here we will focus on the use of inhibition of adrenal steroidogenesis and treatment with alternative steroids in addition to the withdrawal of the first line antiandrogen to produce maximum androgen blockade.
Several trials have been conducted using different treatment modalities (table II).
Blockade of adrenal androgen production to reduce peripheral testosterone has been investigated by several authors.[38,39] Aminoglutethimide induces a functional adrenalectomy by blocking several steps of hydroxylation in the steroid biosynthesis. Because this drug is not selective for sex-hormones, replacement of the glucocorticoids is necessary to prevent adrenal insufficiency in patients treated with this agent. Figg et al. demonstrated that in the 65% of patients who responded, the mean duration of response was 11 months. Sartor et al. found a mean duration of response of 8 months in the 48% of treated patients who responded. Dupont reported an 80% response rate in patients treated with aminogluthetimide and hydrocortisone and the mean duration of response after antiandrogen withdrawal was 14.5 months. These results are improved in terms of the rate of response as well as the duration of response compared with antiandrogen withdrawal alone.
An agent which is also used in this context is ketoconazole, which is widely used as antifungal drug. Ketoconazole blocks the cytochrome P450 enzyme system and therefore also acts as a potent inhibitor of adrenal and testicular androgen production. In hormone refractory prostate cancer ketoconazole produced a response rate of 62.5%, but a mean response duration time of only 3.5 months if it was administered when tumour progression was noted after the antiandrogen was withdrawn. By combining antiandrogen withdrawal with ketoconazole and hydrocortisone the duration of response was prolonged to 8.5 months. Newer agents for adrenal inhibition are under investigation, such as liarozole, but to date no studies combining liarozole with antiandrogen withdrawal have been conducted.
Glucocorticoids were used for a long time as palliative treatment in prostate cancer. Dawson et al. treated patients with hydrocortisone (30 mg/day) together with flutamide withdrawal and demonstrated a response rate of 29%. Figg et al. presented a case report in 1997 of a patient with PSA progression under maximum androgen blockade who received hydrocortisone and simultaneous antiandrogen withdrawal. This patient’s PSA value dropped from 64 μg/L to below the level of detection. Especially remarkable in this particular case is the fact that this complete response was noted over 46 months. In a large cohort of 230 patients who were randomised to receive placebo plus hydrocortisone (40 mg/day) versus suramin plus hydrocortisone, overall 16% of patients responded with a greater than 50% decline in PSA values.
Besides hydrocortisone other corticosteroids have been studied. Nishiyma and Terunuma treated patients following progression after antiandrogen withdrawal with dexamethasone starting with 1.5 mg/day initially, decreasing it to 0.5 mg/day later on. Of the 7 treated patients, 4 patients (57%) experienced a decrease in PSA value of over 90% and a response duration of 3 to 11 months. This confirms a previous study of Storlie et al., in which they reported a response rate of 61% for PSA value decline and 79% for symptomatic improvement in a total of 38 patients treated with dexamethasone (1.5 mg/day).
It remains open to question as to whether patients with hormone refractory prostate cancer may benefit from secondary hormonal therapy at the time of antiandrogen withdrawal, although treatment with aminoglutethimide is especially intriguing. Since there is no proof that the decline of PSA values at that stage of the disease is associated with prolonged survival it is mandatory to look for quality of life in these patients. Particularly the adverse effects of any drugs used must be taken into account.
5. Impact of the Antiandrogen Withdrawal Syndrome on Study Protocol of Clinical Trials in Hormone Refractory Prostate Cancer
Before the first reports of antiandrogen withdrawal syndrome appeared in 1993 it was unknown that in a subset of approximately 30% of patients discontinuation of antiandrogen would result in decreasing PSA values. In particular, phase II treatment trials investigating new cytotoxic drugs must be looked at if patients may have responded because of antiandrogen withdrawal rather than because of the investigated substance. Since the antiandrogen withdrawal syndrome became known and widely accepted, patients entering clinical trials have to discontinue the antiandrogen before entry into a study (at least 4 weeks before for flutamide and 8 weeks before for bicalutamide). Also at least 2 consecutive rising PSA values should be obtained before patients can be included in clinical trials to control for the effect of antiandrogen hormone withdrawal syndrome.
6. Antiandrogen Withdrawal or Discontinuation of Complete Hormonal Therapy
The antiandrogen withdrawal syndrome should not lead to discontinuation of GnRH agonist therapy. It is known that prostate cancer always remains sensitive to androgens. In HRPC, both androgen-independent/hormone-sensitive tumours and hormone-independent tumours will proliferate under endogenous or exogenous testosterone. Discontinuation of GnRH agonists will result in decreased survival, even if in the group with continued hormonal therapy the prolongation of life was only of short duration as reported from the Eastern Cooperative Oncology Group. Adding exogenous testosterone was found to result in progressive disease in 87% of treated patients with increased risk of serious complications such as cord compression, pain progression or death. These results were also reported by Manni et al., including increased toxicity, worse outcome and failing to demonstrate androgen priming prior to chemotherapy. Although a suppressed castrate level of testosterone may persist for up to 1 year after GnRH agonist treatment for more than 2 years, persistence is very variable and not predictable.
These results seem to be controversial in the literature because of a review of a total of 205 patients of the Southwest Oncology Group. 32 patients in this study discontinued hormonal therapy and no survival benefit was seen in either group. But in this study, overall survival in both groups was very short with a median value of only 6 months, suggesting that the patient population had far advanced disease, in whom a survival difference would be difficult to detect.
The existing data indicate that most prostatic carcinoma in the hormone refractory stage would show increased progression in the presence of androgens. Therefore, continued gonadal androgen suppression should be considered standard care for patients with hormone-independent prostatic cancer.
The antiandrogen withdrawal syndrome offers another piece in the puzzle of prostatic carcinoma, in that we are able to understand how the prostate cancer cell is able to escape hormonal therapy, but at the same time, it demonstrates how different advanced prostate cancer cells may react to therapeutic strategies. The antiandrogen withdrawal syndrome is not a simple solution for hormone refractory prostate cancer. Therefore, hormone refractory prostate cancer will remain a difficult challenge that has to be solved in the future.
- 2.Huggins C, Hodges CV. Studies on prostatic cancer, effect of castration of estrogen and of androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1941; 1: 293–7Google Scholar
- 22.Gaddipati GP, McLeod DG, Heidenberg HB, et al. Frequent detection of codon 877 mutation in the androgen receptor gene in advanced prostate cancer. Cancer Res 54: 2862–4Google Scholar
- 25.Kleinerman DI, Troncoso P, Pisters LL, et al. Expression and structure of the androgen receptor in bone metastases of hormone refractory prostate cancer. J Urol 1996; 155: 1254AGoogle Scholar
- 26.Klocker H, Culig Z, Hobisch A, et al. Androgen receptor mutations in prostate cancer. In: Wirth M, Altwein JE, Schmitz-Drager B, et al., editors. Molecular biology of prostate cancer. Berlin: Walter de Gruyter, 1998: 81–8Google Scholar
- 28.The androgen receptor gene mutations database. Available online at URL: http://www.mcgill.ca/androgendb/
- 32.Scher HI, Zhang ZF, Cohen L, et al. Hormonally relapsed prostatic cancer: Lessons from the flutamide withdrawal syndrome. Adv Urol 1995; 8: 61–95Google Scholar
- 44.Small EJ, Baron AD, Fippin L, et al. Ketoconazole retains activity in advanced prostate cancer patients with progression despite flutamide withdrawal. J Urol 1997; 4: 1204–7Google Scholar
- 48.Small EJ, Marshall ME, Reyno L, et al. Superiority of suramin plus hydrocortisone (S+H) over placebo plus hydrocortisone (P+H): Results of a multicenter double-blind phase III study in patients with hormone refractory prostate cancer. Proc Am Soc Clin Oncol 1998; 17: 308–12Google Scholar