How Do You Treat Prostate Cancer That Has Progressed On Primary Androgen Deprivation Therapy? : Part 1

Quick Search


Part 1 of 2:
A Review of Current Diagnostic
and Treatment Approaches

Revised October 1998



Table of Contents

How does AIPC develop?

Issues in AIPC management

Treatment principles for AIPC patients

Other factors to consider when choosing a treatment





Refer to Part 2 for information about
secondary hormone maneuver for AIPC:
The antiandrogen withdrawal response (AAWR)
High-dose ketoconazole plus hydrocortisone
Aminoglutethimide (AG) plus hydrocortisone
Estrogen treatment



Prostate cancer growth is dependent upon adequate circulating levels of androgens. In man, predominant androgens include testosterone, dihydrotestosterone and the adrenal androgen, androstenedione. At the cellular level, androgens stimulate growth and are necessary for the survival and function of cells that possess an androgen receptor.

The androgen receptor belongs to the steroid-receptor “superfamily” that includes receptors for thyroid hormone, retinoic acid (vitamin A), estrogen, progesterone, glucocorticoids (e.g., cortisol and hydrocortisone) and other steroid hormones.1,2 Therefore, the androgen receptor shares some homologous (similar) features to other receptors in its family. The structure of the androgen receptor consists of a series of exons (nucleotide sequences) which when bound to androgen regulate the expression of androgen-responsive genes and interact with other factors that involve DNA transcription. A notable gene regulated by androgen in prostate cells encodes prostate-specific antigen (PSA).3

Androgen deprivation therapy (ADT) causes dramatic regression of prostate cancer and normal prostate tissue. At the cellular level, programmed cell death or apoptosis appears to be mediated directly or indirectly by androgen receptors that are no longer bound to circulating androgens. Progression of disease on ADT indicates that androgen-independent prostate cancer (AIPC) cells escaped apoptosis by some mechanism.

Whether the tumor can change its character in the course of the disease is unknown. Prostate cancer is a multifocal, and even in a single biopsy specimen there can be cellular heterogeneity. The aggressiveness of these tumors appears to directly correlate with the proportion of higher Gleason grade cells. Therefore, the disease process can involve a mixture of discrete clones of androgen-sensitive, androgen-insensitive, and possibly androgen-altered cells.4,5 This heterogeneity appears related to tumor size or burden. As the tumor burden increases by cell division, chances for gene mutations occurring increases, which in turn may lead to androgen or drug-resistant tumors. Such mutations likely resulted from the activation of oncogenes.

For example, in laboratory experiments, expression of an activated H­ras oncogene in androgen-sensitive LNCaP prostate cancer cells allows these cells to grow independent of the presence of androgens. H­ras oncogene has also been shown to stimulate MDR­1, the Multi-Drug Resistance Gene. Therefore, androgen independence and MDR­1 expression may go hand­in­hand. The sequence of events may be depicted as follows:

Increasing tumor burden arrow gene mutation of an oncogene
Oncogene stimulation (e.g. H-ras) arrow stimulation of MDR gene
MDR expression arrow hormone independence

It is now clear that more than one genetic aberration can lead to androgen independence. Mechanisms that have been identified include expression of the protoncogene bcl-26 and mutation of the genetic biomarker p53, a tumor suppressor gene.7 In essence, androgen-independence is more likely to be present at the time patients are diagnosed with extensive disease such as in the lymph nodes or bone. Since androgen deprivation therapy only kills androgen-sensitive cells, androgen-independent tumor cell populations will continue to grow and eventually emerge as the primary disease entity.

A mutation of the androgen receptor gene was hypothesized as one possible survival mechanism for AIPC during ADT. In an attempt to confirm the hypothesis that AIPC cell growth is mediated by gene mutation, Taplin, et al, examined the androgen receptor genes in 10 patients with AIPC.8 The authors noted high levels of androgen receptor gene expression in all of the patient samples, supporting the hypothesis that tumor progression requires a functional androgen receptor gene. In 5 patients, point mutations in the androgen receptor gene were found and all located on the androgen-binding domain. When functional studies were done, progesterone and estrogen were capable of activating mutant androgen receptors in 2 patients.

The authors concluded ADT selects AIPC cells whose mutated androgen receptors stimulate growth without the presence of usual androgen levels. One of the point mutations found by Taplin, et al, had been previously reported by Veldscholte, et al, in LNCaP, a cell line used as an experimental model of androgen-sensitive human prostate cancer. This androgen receptor gene mutation resulted in an androgen receptor that could be activated by estrogen, progesterone and the antiandrogen flutamide.9


Androgen blood levels

The definition of AIPC is disease progression evidenced by a rising PSA or an increase in tumor mass on bone scan, X-ray, CT scan or MRI despite a castrate level of testosterone. Does it make sense to check the serum levels of androgens to verify the patient is truly castrate? We believe so.

For example, if a patient’s PSA stops falling or begins to rise on ADT using a luteinizing-receptor hormone-receptor agonist (LHRH-A) such as leuprolide (Lupron®) or goserelin (Zoladex®), and the testosterone level is castrate (i.e., < 30 ng/dl), then AIPC is present until proven otherwise. If the testosterone is > 30 ng/dL) the patient’s serum LH level should be checked. If LH is not completely suppressed, we believe it is reasonable to increase the dosage of the LHRH-A. If the LH level is suppressed, we would measure the levels of the adrenal androgens dehydroepiandrosterone (DHEA) and androstenedione. These hormones can be converted to testosterone and may account for the elevated level. If these levels were elevated, we would consider adding drugs that suppress adrenal androgen production such as high-dose ketoconazole (HDK) and hydrocortisone. If the levels of adrenal androgen are not increased, a mutation in the androgen receptor is likely.

Antiandrogen Withdrawal

For more information about AAWR, please refer to Part 2 of 2 in this series of booklets

We now know that an androgen receptor mutation can result in the antiandrogen paradoxically stimulating tumor growth. Antiandrogen withdrawal in such patients has been shown to result in tumor regression in approximately 20% of patients. This phenomenon is referred to as an AntiAndrogen Withdrawal Response (AAWR).

If androgen blockade included an antiandrogen, e.g., flutamide (Eulexin®), bicalutamide (Casodex®) or nilutamide (Nilandron®), these agents must be stopped in order to monitor for a possible AAWR. Failure to recognize an AAWR is one of the problems encountered when we attempt to interpret results of studies of AIPC treatments published in the past. If a patient stopped antiandrogen therapy at the same time a different therapy was started, an AAWR, if it occurred, could affect the assessment of response to the second therapy. AIPC treatment studies should require withdrawal of antiandrogens for at least one month to assess whether or not the PSA continues to rise or declines due to an AAWR.

Another reason supporting measurement of adrenal androgen levels was reported in a 1994 abstract. In that study, Herrada, et al meaured serum levels of dehydroepiandrosterone (DHEA) in 10 patients with PSA progression on ADT.10 After antiandrogen therapy was stopped, patients were observed for the possible development of an AAWR. The data from this report is summarized in the table below:

Serum DHEA level (ng/ml) at PSA progression


# Pts. (%)




< 75 ng/ml


5 (50%)




0/5 ( 0%)


5 (50%)




3/5 (60%)

None of the patients with DHEA levels > 75 ng/ml attained an AAWR, while 3 of 5 patients (60%) with DHEA levels < 75 ng/ml achieved an AAWR. Therefore, DHEA levels at the time of PSA progression may be used to identify patients who may or may not benefit from antiandrogen withdrawal.

Problems with published studies using chemotherapy for AIPC

Many past studies that evaluated the efficacy of various secondary treatments predated the days of PSA testing. In these studies, responses were evaluated by improvement in symptoms such as bone pain, or by reduction in tumor size on bone scans or CT scans. Based upon the limited sensitivity of scans to assess tumor response, older studies may have missed patient response that might have been noted if PSA were available.

Past studies may also have underestimated the importance of drug absorption, proper drug dosing based on elimination half-life, dose intensity, and altered drug metabolism. Treatments that were labeled as ineffective in the past may conceivably turn out to be more effective when given to patients with less tumor volume and under better pharmacological conditions. In a thorough review of the literature, long-lasting responses to secondary therapies have been documented. What patient or treatment-related variables were present in such responding patients?

Dose Intensity

Regimen A:
Drug dose: 1000 mg
Frequency: every two weeks
Average: 2,000 mg/month
Regimen B:
Drug dose: 3000 mg
Frequency: every three weeks
Average: 4,000 mg/month

Dose intensity is a term used to compare relative amounts of a drug administered in a given unit of time. For example, compare the relative dose intensities of regimen A and B. Regimen A delivers a dose intensity that averages 2,000 mg/month. Regimen B delivers a dose intensity that averages 4,000 mg/month. When examined in this manner, it becomes clear that the relative dose intensity of Regimen A is 2 times greater than that of Regimen A.

Exposure time

Most chemotherapy agents kill cancer cells that are active multiplying. Prostate cancer generally grow slowly which mandates that they receive a longer exposure time to the chemotherapy or other anticancer agent.Examples of ways to increase exposure time include daily oral therapy, a more frequent schedule of intravenous administration or use of low-dose continuous intravenous infusions.

Bone marrow support

One of the key factors for the successful management of the cancer patient is adequate supportive care. This may involve multiple aspects in the medical and surgical management of the patient, and often includes psychological support as well. With the advent of agents that can stimulate the bone marrow, we now have a mean to give chemotherapy at higher doses by supporting and/or preventing toxicity such as low white blood cell counts, anemia and low platelet counts (see table that follows).

Marrow cell stimulated

Generic name

Trade name
Granulocytes Filgrastim Neupogen®
Granulocytes & macrophages Sargramostim Leukine®
Erythrocytes Erythropoietin alpha Procrit®, Epogen®
Platelets Oprelvekin Numega®

A low white blood cell count (also called granulocytopenia or neutropenia) is a major dose-limiting factor with chemotherapy and is the cause for the most serious side effect of chemotherapy, infection. AIPC patients who receive agents that stimulate the bone marrow to produce white blood cells tolerate this chemotherapy side effect remarkably better. Filgrastim or sargramostim support reduces the number of hospitalizations for infection associated with chemotherapy and reduces other problems such as mouth and throat sores.

For more information about Anemia Associated with Androgen Deprivation (AAAD), please refer to our booklet that is specific to that topic.

A low red blood cell count, or anemia, can also be a significant source of concern for AIPC patients receiving chemotherapy. Anemia is usually already present to some degree in AIPC patients due to their ADT. Anemia, left untreated, can cause severe weakness, shortness of breath, dizziness, mental status changes and chest pain. The availability of erythropoietin to stimulate bone marrow red blood cell production can help minimize the adverse effect severe anemia can have upon the AIPC patient. The use of erythropoietin has largely replaced the need for blood transfusions.

A low platelet count, also called thrombocytopenia, is another dose-limiting factor with chemotherapy and is the cause for a serious side effect of chemotherapy, bleeding. Until recently, thrombocytopenia could delay chemotherapy, cause dosage reductions or even changes in drug therapy. Oprelvekin has recently become available as a marrow stimulant specific for platelet production and its use may treat patients for low platelet counts.

Other supportive care

A medical oncologist should offer the most effective medications or other approaches to maximize the level of supportive care for the AIPC patient receiving chemotherapy. Other chemotherapy side effects include:

Potential side effect Supportive care options
Loss of appetite Megestrol acetate (Megace®)
Nausea and/or vomiting Ondansetron (Zofran®), Granisetron (Kytril®), Dolasetron (Anzemet®), metoclopramide (Reglan®), dexamethasone (Decadron®)
Diarrhea Loperamide (Imodium-AD®), Lomotil®


Potential side effect Supportive care options
Constipation Docusate sodium (Colace®), milk of magnesia
Dry skin, hair loss Emollients, vitamin E, zinc supplements
Heart injury Dexrazoxane (Zinecard®)
Bladder injury Mesna (Mesnex®)
Nerve injury Amifostine (Ethyol®)
Kidney injury Sodium thiosulfate injection

Unfortunately, there are no medications or approaches available that will prevent loss of hair from chemotherapy. However, hair will grow back in a few weeks after therapy is stopped, and may actually begin to grow back during continued chemotherapy treatments.

Certain intravenous chemotherapy drugs, if they accidentally leak out of the vein and into surrounding tissues, can cause significant damage called an extravasation injury. Drugs that can cause extravasation injuries are known as vesicant chemotherapy agents. To prevent potential extravasation injuries, vesicant chemotherapy should be given with caution to patients with poor quality veins, or are to receive drug as a protracted infusion over several days. In most cases, it may be preferable in such patients for them to have a central venous catheter or access device placed before therapy is started. This not only avoids a potential extravasation injury, but also preserves access to a patient’s veins to draw blood.

It is very important that a patient promptly report any unusual symptoms or side effects during chemotherapy treatment to his physician to be sure that it is not, or does not become a major problem. Patients receiving vesicant chemotherapy through a peripheral (hand, arm or leg) vein should inspect the chemotherapy injection site for several days after each treatment.


Due to our concern for the emergence of androgen independence in prostate cancer, the following principles are relevant until we have a better understanding of hormone sensitivity and independence:

1) Don’t wait for symptoms to develop before starting treatment

There is an inverse correlation with diminished survival in patients who are more symptomatic from their prostate cancer than those are with fewer symptoms. The symptom complex is a manifestation of tumor burden. It is also expressed in the stage of disease and may explain why patients with 1-5 bone metastases do so much better than those with greater numbers of bone lesions. Therefore, consider initiating treatment if definite increases in PSA occur. This can be confirmed by 3 consecutive increases of the PSA obtained in the same medical center or office using the same PSA methodology.

2) Use the PSA or another tumor marker to follow response

This is your barometer that reflects the success or failure of therapy. A definite upward trend in the PSA level should dictate a treatment change whereas a flat PSA graph or downward trend would suggest that the treatment remain unchanged. Other tumor markers, such as prostatic acid phosphatase (PAP), alkaline phosphatase or carcino-embryonic antigen (CEA), if initially abnormal, should be followed as well.


Steideck, et al, showed the value of monitoring response to AIPC treatment by using more than one tumor marker. In a retrospective study, he demonstrated that the overall survival of AIPC patients was longer if both PSA and PAP levels declined on therapy than if only PSA or PAP declined. The shortest survival was seen in patients in whom neither marker declined.11

Even if other tumor markers are not abnormal when a particular therapy is started, it is reasonable to monitor their levels periodically on treatment. It is also better to follow trends in PSA, other markers or in other parameters to measure disease activity than to use the results from a single test or the interpretation of an X-ray study or nuclear scan.

3) Understand the importance of drug absorption, dosing and toxicity

Many of the drugs currently in use do not have long half-life in the body and are commonly given every 8 or 12 hours. Patient compliance to these dosing intervals is important to the success of such treatment. Ketoconazole (Nizoral®) and estramustine phosphate (Emcyt®), for example, require an empty stomach for complete absorption. Nizoral also requires a sufficient amount of stomach acid to improve absorption.

4) Use synergistic (more than additive) drug combinations

Treatments employing synergistic combinations of more than one chemotherapy agent or chemotherapy combined with second-line hormonal therapy result in higher rates of anticancer response. It has bee demonstrated that the duration of response and overall survival are significantly longer in patients who have > 50% decrease in PSA with these therapies and even longer in patients who have > 80% PSA fall. We refer to these combination treatments as “high-response regimens.”


The physician’s approach to the patient that has progressive prostate cancer after ADT is complicated. A number of important variables in each patient history and previous pattern of response must be addressed. These include:

  • Age and general health of the patient

Patients with progressive disease after CHB who are elderly, frail or have other significant medical problems do not tolerate many of the therapies described herein as well as do younger patients or patients in otherwise good health. This is not an absolute statement but an overall viewpoint.

  • Amount of disease as reflected by PSA level

Patients that have extensive disease with large tumor burdens have a lower chance of a complete response. In addition, the duration of response, in general, is not as long in these patients. This is true for primary hormonal blockade and also secondary therapies.

  • Potential response to “secondary” hormonal treatments

The term “secondary” hormone therapy includes non-chemotherapy treatments that may be effective in patients with AIPC. The patient’s hormonal status at the time of disease progression is the most important single factor influencing his potential to respond to such therapies.

For more information about second-line hormonal treatments, please refer to Part 2 of this booklet series.

Secondary hormonal treatments include antiandrogen withdrawal alone or coupled with high-dose ketoconazole or aminoglutethimide plus hydrocortisone, estrogens or progestins. We are learning that the effectiveness of such therapies may relate to non-hormonal effects of drugs such as ketoconazole and estrogens.


AIPC has traditionally been considered refractory to all forms of therapy other than analgesics for symptomatic relief. In 1998, there is a growing list of treatment options available for men with this disease that can result in significant palliation and ultimately may improve survival.

Urologists have traditionally not considered secondary hormone treatment or cytotoxic agents as a viable treatment options for AIPC. Their rationale is largely based upon a history of limited efficacy in older studies and risks for side effects when these drugs are given to men who are debilitated by extensive cancer. This is no longer considered reasonable today.

Under the direction of a medical oncologist, there are many active regimens for AIPC today that have good toxicity profiles such that any side effects can be minimized or avoided. The use of high-response regimens is justified for men with AIPC, both to palliate disease-related symptoms and to improve quantity as well as quality of their lives.


  1. Evans RM: The steroid and thyroid hormone receptor superfamily. Science 240:889-95, 1988.
  2. Beato M: Gene regulation by steroid hormones. Cell 56:335-44, 1989.
  3. Riegman PHJ, Vliestra RJ, van der Korput JAGM, Brinkmann AO and Trapman J: The promoter of the prostate-specific antigen gene contains a functional androgen responsive element. Mol Endocrinol 5:1921-30, 1991.
  4. Aihara M, Wheeler TM, Ohori M and Scardino PT: Heterogeneity of prostate cancer in radical prostatectomy samples. Urology 43:60-4, 1994.
  5. Brawn PN: The dedifferentiation of prostate cancer. Cancer 52:246-51, 1983.
  6. McDonnell TJ, Troncoso P, Brisbay SM, et al: Expression of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res 52:6940-4, 1992.
  7. Bookstein R, MacGrogan D, Hilsenbeck SG, Sharkey F, and Allred DC: p53 is mutated in a subset of advanced-stage prostate cancers. Cancer Res 53:3369-73, 1993.
  8. Taplin ME, Bubley GJ, Shuster TD, Frantz ME, Spooner AE, et al: Mutation of the androgen receptor gene in metastatic androgen-independent prostate cancer. N Engl J Med 332:1393-8, 1995.
  9. Veldscholte J, Ris-Stalpers C, Kuiper GGJM, et al: A mutation in the ligand binding domain of the androgen receptor of human LNCaP cells affects steroid binding characteristics and response to anti-androgens. Biochem Biophys Res Commun 173:534-40, 1990.
  10. Herrada J, Hossan B, Amato R, et al: Adrenal androgens predict for early progression to flutamide withdrawal in patients with androgen-independent prostate carcinoma. Proc Am Soc Clin Oncol 13:237, 1994.
  11. Steideck, et al: Urology 47:719, 1996.

© Prostate Cancer Research Institute (PCRI) 1998. All rights reserved.