Sequencing of Agents for Castration-Resistant Prostate Cancer
Sequencing of Agents for Castration-Resistant Prostate Cancer
Ten years ago, the clinician treating metastatic castration-resistant prostate cancer (CRPC) had palliative options for treatment of symptomatic patients, such as the combination of mitoxantrone combined with prednisone, as well as isotope therapy. In 2004, docetaxel-based chemotherapy regimens were shown to provide an overall survival benefit for patients with CRPC.[1,2] Today, the prostate cancer oncologist is in the enviable position of having six US Food and Drug Administration–approved agents to choose from: immunotherapy (sipuleucel-T), hormonal therapies (abiraterone, enzalutamide), radiopharmaceuticals (radium-223), and chemotherapy (docetaxel, cabazitaxel), in addition to agents being administered in clinical trials. In general, the sequencing of these drugs is based upon the entry criteria from the phase III trials that led to their approval. Selection of treatment is based on symptoms, sites of disease (bone vs visceral) and types of prior treatment (docetaxel-ineligible vs pre-docetaxel vs post-docetaxel setting). Unfortunately, there is a lack of useful correlative biomarkers in prostate cancer to help oncologists select treatment. This problem is best illustrated in the post-docetaxel castration-resistant setting, for which there are indications to use all five other approved agents. In this review we will outline an approach to sequencing these new therapies, with particular attention paid to the biology of CRPC.
The Definition of CRPC
Castration-resistant prostate cancer (CRPC), also referred to as hormone-refractory prostate cancer, has been defined generally as prostate cancer that continues to grow after either bilateral orchiectomy or treatment with a luteinizing hormone–releasing hormone (LHRH) agonist and concurrent treatment with a nonsteroidal anti-androgen (NSAA). This type of disease used to be referred to as androgen-independent prostate cancer, but that term is used less commonly now, given our understanding that tumors remain at least partially androgen-sensitive until much later in the disease process. Furthermore, as will be discussed, androgens are present in cancer cells both in patients undergoing treatment with LHRH agonists and in those receiving treatment with NSAAs. Thus, most CRPC as currently defined depends on activity of androgens or the androgen receptor (AR). Since the term CRPC was coined, we have developed much more powerful agents to decrease testosterone levels. With the likely use of these drugs earlier in the course of disease, the term CRPC will either become outdated or eventually will refer to an entirely different set of clinical circumstances.
Androgen Signaling Basics
Our understanding of androgen signaling in prostate cancer has changed dramatically over the last decade, leading to the development of much more effective treatments. The hypothalamus produces LHRH (also called gonadotropin-releasing hormone) in pulses throughout the day. LHRH promotes production of luteinizing hormone (LH) in the pituitary. LH, in turn, acts in the Leydig cells of the testes to produce testosterone. This pathway (Figure 1) produces the majority of testosterone in the body (50% to 90%, depending on age). However, testosterone precursor compounds (eg, dihydroepiandrostenedione [DHEA], DHEA-sulfate [DHEA-S], and androstenedione) are produced by the adrenal glands. These precursors are taken up by the prostate, other end organs, and prostate cancer cells, where they are converted into testosterone and its more potent metabolite, dihydrotestosterone (DHT).[5-8]
The AR is a transcription factor found in the cytoplasm when it is not bound by ligand. DHT binding to the AR causes release of heat shock proteins, AR conformational change, and translocation of the AR to the nucleus. In the nucleus, the AR initiates transcription of genes that promote survival and proliferation, and of markers such as kallikrein-related peptidase 3 (KLK3), which encodes prostate-specific antigen (PSA).[9,10] Transcriptional activation by the AR is the central driver of early prostate cancers and probably of most advanced prostate cancers.
Decreasing circulating testosterone has been the major strategy for treatment of metastatic prostate cancer for over half a century. Bilateral orchiectomy was used until LHRH agonists became available (leuprolide, goserelin, triptorelin, buserelin). An LHRH antagonist (degarelix) is also approved in this setting.[11,12] Both surgical and medical castration dramatically reduce the amount of testosterone and almost always cause shrinkage of tumors, as well as decreases in the amount of circulating PSA. As noted, because of peripheral production of testosterone precursors that are metabolized to testosterone in target tissues, orchiectomy and LHRH agonist/antagonist treatment do not eliminate testosterone completely.
Given the availability of testosterone to cancer cells in patients on LHRH agonist monotherapy, LHRH agonists are combined with an NSAA AR inhibitor such as bicalutamide, flutamide, or nilutamide at first to prevent a flare reaction. A meta-analysis concluded that continuation of antiandrogens, termed combined androgen blockade (CAB), results in a 5% overall survival (OS) advantage over LHRH monotherapy, albeit with some decrement in quality of life.
Initial Treatments: Inhibition of Androgen Signaling
When PSA rises or imaging studies demonstrate disease progression in patients on an LHRH agonist /antagonist or CAB, the first step in disease management is to assess whether the patient has castrate levels of circulating testosterone. Normal testosterone levels are > 300 ng/dL. Castrate levels are defined as being ≤ 50 ng/dL, although this definition is based on testing using outdated technology. Furthermore, data indicate that even levels of 20 ng/dL are high enough to drive prostate cancer growth.[15-18] Testing by mass spectrometry is the gold standard, although radioimmunoassay (RIA) and chemiluminescent (CL) assays are used most commonly. Because RIA and CL assays have a high degree of variability, their results should be interpreted with caution.
Most patients will achieve very low levels of testosterone (< 20 ng/dL). For patients who do not reach castrate levels of testosterone initially, or for those with castrate levels initially who then have rising testosterone, our practice is to switch from an LHRH agonist to the LHRH antagonist degarelix. Responses to degarelix have been reported in patients with noncastrate testosterone levels during treatment with an LHRH agonist.
Prior to the approval of abiraterone acetate in pre-docetaxel patients, men with disease progression on LHRH monotherapy were treated most commonly by continuing the LHRH agonist and adding an NSAA. The addition of an NSAA has been shown to produce objective responses, reduce PSA level, and decrease pain,[20,21] but no OS increase has ever been demonstrated. Unfortunately, responses to NSAAs are limited; response rates are approximately 30% and usually last 3 to 5 months, although occasional durable responses lasting several years are seen.
When disease progression occurs in patients on CAB, the NSAA can be withdrawn or changed to a different NSAA. Similar to adding NSAAs, withdrawal results in responses in 20% to 40% of patients, and these usually last for several months.[22-29] When NSAAs bind to the AR, they cause it to translocate to the nucleus and bind to DNA as androgens do, but initially they repress transcription.[30-32] Studies in prostate cancer cell lines show that when the levels of AR are elevated, NSAAs become agonists and promote transcription of AR-responsive genes.[31,32]
Thus, NSAAs can become AR agonists; withdrawal probably results in exit of most of the ARs from the nucleus and decreased AR signaling. Alternatively, switching to a different NSAA results in response rates and median times to progression similar to those associated with NSAA withdrawal.[33-36] It is possible that this occurs because differences in AR-binding affinities among the three available NSAAs alters AR activity.[37,38]
One other hormonal manipulation, oral ketoconazole, has been used for more than 30 years. Ketoconazole inhibits CYP17, which catalyzes two reactions required for testosterone production. Thus, ketoconazole can lower the level of testosterone still remaining in a patient receiving LHRH agonists. The response rate (approximately 30%) and time to progression (about 3 months) are similar to results seen with NSAAs. Soon after ketoconazole is used, patients’ testosterone levels rise dramatically, resulting in loss of efficacy. In a randomized trial comparing treatment with ketoconazole vs antiandrogen withdrawal, no survival benefit was noted. The prechemotherapy use of ketoconazole has declined significantly with the approval of abiraterone acetate in the pre-docetaxel setting. Ketoconazole is still an appropriate choice for patients who cannot afford abiraterone acetate or who have nonmetastatic CRPC.
Sipuleucel-T is the first cell-based immunotherapy to improve OS in any carcinoma in a phase III study. In this treatment, the patient’s own mononuclear cells (including dendritic cells, the major antigen presenting cells of the immune system) are harvested by leukapheresis, incubated with a prostatic acid phosphatase (PAP)-granulocyte macrophage colony-stimulating factor (GM-CSF) fusion protein, and then re-infused into the patient. The treatment is repeated three times every 2 weeks. The rationale for the treatment is to expose dendritic cells to PAP in the presence of growth factor, causing them to activate the immune system to attack cells expressing PAP (ie, prostate and prostate cancer cells) upon reinfusion. US Food and Drug Administration (FDA) approval was granted after the phase III IMPACT study (Identification of Men with a genetic predisposition to ProstAte Cancer: Targeted screening in men at higher genetic risk and controls) demonstrated a 4.1-month OS advantage of treatment with sipuleucel-T over placebo. Among the 512 men on the study, 84.5% had not received docetaxel prior to treatment with sipuleucel-T. In this study, 50.7% of patients had only bony disease and 41.9% had both bony and soft-tissue disease. Patients were either asymptomatic or minimally symptomatic (with the latter defined as Eastern Cooperative Oncology Group performance status [ECOG PS] ≤ 1 with none of the following: pathologic bone fractures, visceral metastases, recently initiated treatment with bisphosphonates, and spinal cord compression). Use of glucocorticoids was also excluded, based on their potential immunosuppressive effects. Despite the reported survival advantage, no progression-free survival (PFS) advantage, PSA response, or measurable disease response was seen with sipuleucel-T. Nevertheless, PSA responses were seen in earlier studies, as was a single documented complete response. It is postulated that this disconnect between PFS and OS may be due in part to delayed or prolonged activation of the immune system. Given this concept, sipuleucel-T is most appropriate for patients with slowly rising PSA levels and disease that is not seen to be increasing rapidly on CT or bone scanning. Trials are being designed to broaden indications for its use to patients with castration-resistant, biochemically relapsed prostate cancer.
Abiraterone acetate provided a major breakthrough in prostate cancer, both clinically and conceptually. As previously noted, abiraterone inhibits both enzymatic activities of the CYP17 enzyme: the 17α-hydroxylase and 17,20-lyase functions. These reactions are required to produce DHEA, DHEA-S, and androstenedione in the adrenal glands, which are converted to testosterone in the prostate or in prostate cancer cells. Thus, abiraterone acetate dramatically reduces the amount of testosterone in cancer cells, even in patients already producing no testicular androgen (ie, orchiectomized patients or those taking LHRH agonists).
Multiple studies,[45-47] including a large randomized placebo-controlled trial, have shown impressive responses to abiraterone in patients with disease progression on LHRH agonists. The latter study was stopped early because a planned interim analysis demonstrated significant PFS and OS advantages in the abiraterone acetate arm (Table 1). Regarding results of treatment with abiraterone vs placebo, the median OS was not reached with abiraterone vs 27.2 months with placebo, and the PFS was 16.5 months vs 8.3 months, respectively. Notably, for OS, a prespecified significance boundary was not reached, however the abiraterone treatment effect was favorable across all prespecified subgroups. In multiple other measures tested, treatment with abiraterone and prednisone was superior to treatment with prednisone alone (in terms of outcomes as assessed by Response Evaluation Criteria In Solid Tumors [RECIST], time to increase in pain, time to initiation of cytotoxic chemotherapy, PSA response, time to PSA progression, time to decline in functional status, etc).
A previous randomized controlled trial in patients who had already received docetaxel also demonstrated an OS advantage and will be discussed later in this article.
Radiopharmaceuticals have been used for pain control in prostate cancer for many years. Samarium 153 ethylenediamine tetramethylene phosphonate (153Sm-EDTMP), rhenium-186 hydroxyethylidenediphosphonate (186Re-HEDP), and strontium-89 chloride (89SrCl), are compounds that contain beta-emitting radioisotopes of samarium, rhenium, and strontium, respectively. These compounds home to areas of metabolic bone turnover, causing DNA and oxidative damage to metastatic cells at those sites. All have been shown to decrease pain from bony metastases, but evidence for an OS advantage is limited to one small phase II study. These drugs are limited in efficacy because of dose-limiting bone marrow toxicity. Beta particles penetrate between 0.5 mm and 2.5 mm into soft tissue. Radium-223 dichloride (Ra-223) is an alpha-emitting isotope that also targets and attacks bony disease in the same way. However, the tissue penetration of Ra-223 is only between 2 and 10 cell diameters and causes very little bone marrow toxicity. A small phase II study of 64 docetaxel-pretreated patients randomized to Ra-223 or placebo not only demonstrated a benefit in bone pain, but also demonstrated an OS advantage.[50,51] In the large phase III study comparing Ra-223 to placebo in patients with bony metastases but no visceral metastases (the Alpharadin in Symptomatic Prostate Cancer Patients [ALSYMPCA] trial), OS and time to first skeletal-related event were significantly longer in the Ra-223 arm. This patient population included patients who had received docetaxel (57%) as well as those who were ineligible to receive docetaxel or declined to be treated with it (43%).
The first therapies with proven survival advantages in prostate cancer contained the taxane docetaxel. Taxanes inhibit microtubule breakdown, resulting in apoptosis, although recent evidence suggests that in prostate cancer cells they can inhibit nuclear translocation and expression of the AR.[53,54] In 2004, two studies demonstrated OS advantages in patients treated with docetaxel-containing regimens.[1,2] In the TAX 327 study, 1,006 men with hormone-refractory prostate cancer were randomly assigned to one of three treatment arms: mitoxantrone, docetaxel given every 3 weeks, or docetaxel given weekly. In previous studies mitoxantrone was shown to have no effect on survival (although it did palliate symptoms), so it was used as a comparator arm.[55-59] OS in both docetaxel containing arms was improved compared with the mitoxantrone arm, although only the difference between 3-week docetaxel dosing and mitoxantrone reached statistical significance. Additionally, pain, quality of life, and PSA responses were significantly improved. In the Southwestern Oncology Group (SWOG) 99-16 study, 770 men with hormone-refractory prostate cancer were randomly assigned to treatment with docetaxel and estramustine or mitoxantrone and prednisone every 3 weeks. Similar to the TAX 327 results, OS, time to progression, and PSA responses were significantly better in the docetaxel-containing arm. Owing to the similarity between the results of the two studies and the comparative ease of giving prednisone (as opposed to estramustine), docetaxel/prednisone with every-3-week dosing of docetaxel has become the standard of care.