Novel Agents in the Management of Castration-Resistant Metastatic Prostate Cancer

The Prostate Cancer Clinical States model was first proposed by Scher and colleagues nearly 10 years ago. As this model has been refined, it provides a framework for thinking about prostate cancer from a clinical management perspective and in the context of conducting investigational studies for new drug development.

Historically, androgen-deprivation therapy (ADT) has been used as a first-line treatment for advanced prostate cancer, with a goal of achieving testosterone levels in the body nearly equivalent to the levels afforded by surgical castration (Figure 2).² Based on Medicare data from the Surveillance, Epidemiology, and End Results (SEER) program, the use of ADT among Medicare beneficiaries was used in 44.8% of all prostate cancer cases diagnosed between 2000 and 2002; in 15% of patients, ADT was used as an adjuvant with radiation, and in 14% of patients, it was given as primary therapy.³

ADT initially induces remission or stabilization in most patients with locally advanced or metastatic prostate cancer, but within 18 to 24 months, nearly all will experience disease progression.⁴ Prior to the advent of prostate-specific antigen (PSA) testing, patients with known metastatic disease who developed disease-related symptoms (eg, pain, progressive fatigue, weight loss) were believed to be experiencing progression, which was verified using bone scanning and/or computed tomography (CT) imaging. Patients progressing on hormonal therapy were typically identified as androgen independent or hormone refractory. Median survival from the onset of progression for androgen-independent patients with metastatic disease was 10 to 12 months.⁴

In today’s PSA era, the identification of “progression” is typically a rising PSA in the setting of castrate levels of testosterone (<50 ng/dL). Recent evidence from hypothesis-generating studies has resulted in proposals to reduce the target castrate level of testosterone to ≤20 ng/dL, which has been associated with improved progression-free survival.^5,6 However, a subset of patients treated with luteinizing hormone releasing hormone (LHRH) agonists, a therapy commonly used since the early 1970s, failed to achieve a target level of 50 ng/dL,⁶ and, of course, even patients who achieve this nadir eventually progress.

Management of patients with disease progression following ADT has typically involved an empiric effort to maintain patients at castrate levels of testosterone, though evidence to support this practice has been limited. Second-line hormonal therapies with antiandrogens or other agents such as ketoconazole, aminoglutethimide, and estrogenic compounds have been used. Although only relative minorities of patients respond and the response tends to be quite brief, patients occasionally will manifest a response that can be very durable.⁷

ANDROGEN RECEPTOR: PROSTATE CANCER’S HOLY GRAIL

Prostate cancer cells depend on androgens to survive and proliferate, which is the rationale behind androgen-ablative therapy.⁸ The most potent androgens are testosterone and dihydrotestosterone (DHT). In the initial development of prostate cancer and in progression of the disease, testosterone and DHT bind to and activate the androgen receptor (AR). DHT has a binding affinity for AR that is nearly 10 times that of testosterone. AR, a phosphoprotein, functions as a transcription factor to mediate the effects of these steroids on the cell cycle.⁹

Intraprostatic levels of testosterone and DHT (in situ and metastatic) remain elevated, despite achieving serum castrate levels with ADT.¹⁰ In addition, PSA—encoded by an androgen-responsive gene—is detectable in nearly all cases of Castration-Resistant Prostate Cancer (CRPC). It is clear that the AR-signaling pathway remains persistently activated despite androgen blockade.

In patients with CRPC, resumed progression is likely due to AR mutations, AR gene amplification, AR overexpression, modulation of AR via signaling pathways, alteration of coactivators, or a combination of molecular mechanisms.^8,10 Some patients appear to have a hypersensitive AR phenotype that renders prostate cancer cells exquisitely sensitive to extremely low levels of testosterone and DHT.¹¹

The development of newer and more effective methods of blocking AR synthesis may have therapeutic promise for patients with CRPC. Abiraterone acetate and MDV3100, two novel AR antagonists, are currently being evaluated in phase III studies. In preclinical and clinical studies, these drugs have shown promising activity against metastatic CRPC.

ABIRATERONE ACETATE

Abiraterone acetate is an important compound in late-stage development for CRPC and has been under investigation for more than 2 decades. It is an oral bioavailable prodrug of abiraterone, which selectively and irreversibly inhibits the activity of 17-hydroxylase/17,20-lyase (CYP17), an enzyme principally responsible for androgen and estrogen biosynthesis.¹² CYP17 inhibition results in significantly lower levels of circulating dehydroepiandrosterone (DHEA) and androstenedione, a metabolite of DHEA that is synthesized into the more potent androgens testosterone and DHT.

Abiraterone is 10 to 30 times more potent than the nonspecific antiandrogen ketoconazole at inhibiting CYP17.¹³ Abiraterone acetate results in decreased levels of adrenal androgens downstream of CYP17. These levels do not increase at progression, demonstrating the irreversibility of abiraterone’s androgen blockade.^12,14 Conversely, ketoconazole is associated with an increase in steroid levels downstream of CYP17 at progression.¹⁴

Continuous CYP17 inhibition leads to elevated levels of adrenocorticotropic hormone, which in turn increases the levels of corticosterone and deoxycorticosterone upstream of CYP17. While this helps to prevent adrenocortical insufficiency, some patients develop mineralocorticoid excess syndrome, evidenced by symptoms of fluid retention, hypertension, and hypokalemia (Table 1). These toxicities are easily and effectively managed using mineralocorticoid antagonists or low-dose glucocorticoids.^13-15

Phase II Trial in Chemotherapy-Naïve Patients

In the first half of 2010, updated data from 2 phase II trials of abiraterone acetate were published.^13,16 Combined with data from a phase II trial published in 2009,¹⁵ they continue to support the activity of abiraterone acetate in metastatic CRPC (Table 2) and provide rationale for phase III studies, which have completed accrual and are maturing.

In a phase I/II study, Attard and colleagues found the drug highly active in men with chemotherapy-naïve CRPC. A preliminary analysis of phase II data published in the Journal of Clinical Oncology in 2009 showed that a 1000 mg dose of abiraterone daily for 12 weeks reduced PSA levels by ≥50% compared with baseline in 28 of 42 patients (67%). Of these patients, 8 had a PSA decline of ≥90% from baseline.¹⁵

The men in this study had progressed after a median of 3 lines of hormonal treatment prior to enrollment, including LHRH analogs and antiandrogens. At baseline, CT scanning showed measurable tumors in 57% of patients, and a bone scan showed bone metastases in 76% of patients.¹⁵

Median follow-up for the 42 phase II patients was 505 days. CT scanning was conducted every 12 weeks and bone scans were obtained every 24 weeks. In addition, counts were taken of circulating tumor cells (CTCs) using CellSearch technology on the first day of every 28-day cycle. On the last day of each cycle, levels of PSA, lactate dehydrogenase, and alkaline phosphatase were assessed. For the 24 patients with measurable tumors at baseline, 9 experienced tumor shrinkage according to RECIST criteria, for an objective response rate of 37.5%.¹⁵ Of the 17 patients with baseline CTC counts ≥5 per 7.5 mL, 10 (59%) saw a decline to <5 CTCs per 7.5 mL.

The median time to PSA progression (TTPP) for patients who received only abiraterone was 225 days (95% confidence interval [CI], 122-383 d).¹⁵ There was a positive correlation between TTPP and PSA decline, with a TTPP of 393 days for patients whose PSA dropped ≥90% (95% CI, 252-533 d).

The study design allowed for concomitant administration of 0.5 mg per day of dexamethasone with abiraterone acetate at progression, and 39 patients received this combination. Some patients had prior exposure to dexamethasone, and one-third of these patients had a secondary PSA decline ≥50% when dexamethasone was added to abiraterone acetate, prompting investigation into combining abiraterone acetate with a corticosteroid.^15,16

Abiraterone Post-Docetaxel

In a multicenter phase II trial, Danila and associates administered abiraterone acetate and prednisone to 58 men with metastatic CRPC who had all failed docetaxel-based chemotherapy. Nearly half (47%) had prior ketoconazole. PSA levels declined ≥50% in 22 patients (36%), and in the 22 patients with RECIST-evaluable lesions, 4 (18%) had partial response, 13 (59%) experienced stable disease >3 months, and 5 (23%) progressed. Overall median TTPP was 169 days (95% CI, 82-200 d), and 16 (28%) patients saw improvements in ECOG PS.¹⁶

In this study, patients pretreated with ketoconazole had worse outcomes than patients who were ketoconazolenaïve. They were less likely to achieve the primary endpoint of a PSA decline ≥50% and had shorter TTPP. However, 30% of patients previously treated with ketoconazole did experience a decline of ≥50% in PSA levels, indicating no cross-resistance between these therapies. Data from ongoing prospective studies investigating abiraterone in ketoconazole treated patients with CRPC are needed to validate this hypothesis.

The addition of prednisone contributed to reduced rates of mineralocorticoid-associated toxicities of hypertension (5%), hypokalemia (<5%), and fluid retention (<10%) and obviated any need for a mineralocorticoid antagonist. One case of grade 3 fatigue was the only adverse event >grade 2.¹⁶

In a similar phase II study of 47 men with CRPC pretreated with docetaxel, once-daily abiraterone acetate alone led to a PSA decline ≥50% in 24 (51%) patients.¹⁴ Another 7 (15%) experienced a PSA decline ≥90%. Median TTPP in this trial was also 169 days (95% CI, 113-281 d). CTC counts declined in 11 of the 27 patients (41%) with baseline counts ≥5 per 7.5 mL. In this study, 55% of patients experienced hypokalemia, 17% developed hypertension, and 15% had fluid retention. This trial has been advanced to a phase III study comparing abiraterone acetate with placebo. The trial rapidly completed accrual, and data may be available as early as the end of 2010.

Based on these studies, abiraterone acetate demonstrates objective antitumor activity in a preponderance of patients, measured biochemically with PSA testing and as antitumor activity in soft-tissue disease. It is active in men with CPRC who have been treated previously with a range of therapies, from LHRH analogs, to docetaxel-based chemotherapy and ketoconazole. Abiraterone appears more potent than currently used antiandrogens and lacks the highly toxic adverse effects associated with ketoconazole.

MDV3100

A novel second-generation antiandrogen that has had demonstrated success in preclinical and clinical studies is MDV3100, an orally available, small-molecule AR antagonist. MDV3100 was specifically engineered to bind to AR with greater affinity than the approved antiandrogen bicalutamide.¹⁷ MDV3100 also reduces the efficiency of nuclear translocation of AR, prevents DNA binding to androgen response elements, and impairs recruitment of coactivators.¹⁸

In 2009, Tran and associates reported that MDV3100 inhibited tumor growth and induced apoptosis in the AR-amplified VCaP cell line in vitro. This cell line proved resistant to bicalutamide.¹⁸ Additional in vitro experiments compared bicalutamide and MDV3100 in a CRPC cell line engineered to overexpress wild-type AR (LNCaP/AR). Bicalutamide demonstrated agonist effects; MDV3100 did not.¹⁸ Finally, in castrated mice with LNCaP/AR xenograft tumors, treatment with MDV3100 resulted in apoptosis and tumor regression, whereas bicalutamide only slowed tumor growth.¹⁸

It is theorized that the agonist activity of bicalutamide in tumors with AR amplification is why some patients experience antiandrogen withdrawal syndrome, evidenced by a swift decline in PSA levels after discontinuing therapy.¹⁷ AR amplification is evident in as many as 20% to 30% of hormone-refractory prostate tumors but is rarely seen in hormone-sensitive tumors.¹⁹ MDV3100 is purely antagonistic, which might partially explain why it was more effective than bicalutamide in murine models with AR amplification.

Based on the highly positive results from preclinical studies, the Prostate Cancer Clinical Trials Consortium selected MDV3100 for a phase I/II human trial. A US multicenter trial was initiated in 2007, with preliminary data presented at various oncology and urology conferences.

Phase I/II Data

Scher and associates enrolled 140 patients (median age, 68 y) with progressive, metastatic CRPC at 5 centers in the United States.¹⁷ This was a dose-escalation trial, with the first cohort of 3 patients initiated on a 30 mg dose of MDV3100. After establishing the safety of this dose, 24 more patients were enrolled, half of who had received previous chemotherapy. Starting doses for these and subsequent patients accrued ranged from 60 mg to 600 mg daily, but doses for all patients were later dropped to ≤240 mg, which was established as the maximum tolerated dose.

Median duration of therapy with MDV3100 was 21 weeks. Each month, patients were assessed for toxicities and blood samples were analyzed for platelet count, serum chemistry, and levels of PSA and creatinine. CTC count was obtained at baseline and reassessed at weeks 4 and 12. Patients who had metastatic disease at enrollment underwent imaging studies every 3 months versus every 6 months for those without metastases. Every patient also received several electrocardiograms throughout the course of the study. Twenty-two patients enrolled in a concurrent trial at Memorial Sloan-Kettering Cancer Center that assessed therapy response using 16β-18F-fluoro-5α-dihydrotestosterone (FDHT) and 2-18F-fluoro-deoxy-D-glucose (FDG) PET-CT scanning.

Antitumor effects were evident at all dose levels. More than half (54%) the patients had received prior chemotherapy, but no significant difference was observed between this cohort and the chemotherapy-naïve cohort in the proportion of patients achieving a >50% decrease in PSA level from baseline (51% vs 62%, respectively; P = .23). A significant difference was noted between these two groups at 12 weeks when looking at PSA declines only for those patients who remained on MDV3100 a minimum of 12 weeks (Figure 3). In the prior-chemotherapy arm, 36% of patients experienced a decline in PSA levels >50% from baseline compared with 57% of chemotherapy-naïve patients (P = .02).¹⁷

The extent of PSA decline was dose dependent, but only up to 150 mg daily. No additional clinical benefit was seen with doses above 150 mg to 240 mg daily, even though patients at daily dose levels of 360 mg and 480 mg had substantially higher plasma concentrations of MDV3100.

In patients with soft tissue lesions, 22% had partial response and 29% experienced stable disease (SD). In patients with bone metastases, 56% had SD. Of the 51 patients who had unfavorable CTC counts at baseline (≥5 cells per 7.5 mL of blood), counts dropped to <5 cells per 7.5 mL for 26 (49%). A lower proportion of patients in the prior-chemotherapy group than in the chemotherapy-naïve group who had baseline unfavorable CTC counts experienced conversion to favorable counts after treatment (37% vs 75%, respectively; P = .02). Almost all patients (91%) who started with a favorable CTC count retained this status throughout the study.

Slightly more than three-quarters of patients who showed CTC conversion had a PSA decline ≥50%. In the phase II study of abiraterone in patients post-docetaxel, Reid and colleagues found no significant correlation between maximal changes in CTC counts and PSA levels in the overall population, but they did report a significant correlation between these endpoints in patients with ERG gene-arranged tumors.¹⁴ Scher and associates did not stratify outcomes according to ERG gene arrangements, but they concluded that there was a general—if inconsistent—association between PSA declines or lack of progression and favorable CTC counts.¹⁷ Some studies have linked favorable CTC counts at baseline and with therapy to improved survival in metastatic prostate cancer.^20,21

Median time to radiological progression was 47 weeks overall, but it was not reached for the subset of patients who were chemotherapy-naïve. For patients previously treated with chemotherapy, median TTPP was 29 weeks. Based on Prostate Cancer Clinical Trials Working Group 2 criteria of a confirmed PSA rise ≥25% from nadir, median TTPP progression was 32 weeks overall, 41 weeks for patients who had not had prior chemotherapy, and 29 weeks for patients who had received prior chemotherapy.¹⁷

Like abiraterone acetate, MDV3100 appears to be a much more potent antiandrogen than those in current clinical use. In this select population, MDV3100 was associated with relatively high rates of biochemical response, objective activity, stable disease on bone imaging, and relatively protracted periods before evidence of biochemical and radiographic progression.

Fatigue appeared to be the most frequent grade 3-4 adverse event, experienced by 11% of patients. The onset of fatigue seemed to coincide with the time it took MDV3100 plasma concentration to reach a steady state, which was approximately 4 weeks. The fatigue typically resolved 2 to 4 weeks after doses were reduced.

All other grade 3-4 adverse events affected <5% of the study population, and with the exception of anemia, occurred more frequently in patients taking doses of MDV3100 >240 mg. Treatment discontinuation was also more likely at doses >240 mg. Two seizures were reported in patients taking ≥360 mg of MDV3100, but these were not confirmed as treatment related. There were no treatment-related cardiovascular effects reported.

The low rate of serious adverse events is a major improvement over some of the treatments used today for metastatic CRPC and, coupled with the demonstrable activity of MDV3100, justify investigation in phase III trials. At least one phase III trial has already been initiated, comparing MDV3100 with placebo in men with CRPC that has progressed after therapy with docetaxel.

CONCLUSION

Androgen deprivation therapy has been the mainstay of the management of advanced prostate cancer for nearly 7 decades. Despite significant changes in patient presentation as a consequence of a ubiquitous screening paradigm, those patients not cured with local therapy have seen limited progress in management over the past several decades. Significant new insights into the biology of the AR have been translated into clinical developments with agents such as abiraterone acetate and MDV3100, both of which are currently in late-stage clinical development with ongoing human trials. Recent FDA approvals of sipuleucel-T (Provenge) and cabazitaxel (Jevtana) provide additional management options for patients with advanced prostate cancer.

The approval of new therapies is always associated with a subsequent learning curve. Adding agents such as MDV3100 and abiraterone acetate to our therapeutic armatorium will create opportunities and challenges. Determining how to optimally integrate these therapies into our management paradigm will require additional clinical trials. In addition, the added costs of multiple new therapies will need to be taken into consideration, given our evolving healthcare environment. Although the potential to evolve the management of CRPC into a chronic disease paradigm has not quite arrived, perhaps it is not as quixotic as it appeared to be just a short time ago.