Pharmacokinetics, Safety, and Antitumor Effect of Apalutamide with Abiraterone Acetate plus Prednisone in Metastatic Castration-Resistant Prostate Cancer: Phase Ib Study

This article is featured in Highlights of This Issue, p. 3497

Prostate cancer remains a global health problem and a leading cause of male morbidity and mortality (1). Since the 1940s, androgen receptor (AR) inhibition has been a mainstay of therapy for patients with prostate cancer who are in need of systemic treatment (2). While androgen deprivation therapy (ADT) is initially highly effective as a primary therapy, progression eventually occurs in the castrate setting because of continued activation of the AR (3–5). Even for castration-resistant prostate cancer (CRPC), AR-mediated signaling remains an important driver of disease progression (6, 7). As such, pharmacologic inhibition of AR activation is important for the management of CRPC.

Apalutamide is a next-generation AR inhibitor that binds directly to the AR ligand-binding domain with high affinity and prevents AR nuclear translocation, DNA binding, and transcription of AR gene targets (8); it is approved for the treatment of patients with nonmetastatic CRPC worldwide and of patients with metastatic castration-sensitive prostate cancer (mCSPC) in the United States and Europe. The SPARTAN study demonstrated that apalutamide prolonged metastasis-free survival (MFS) by more than 2 years and provided a 55% reduction in the risk of symptomatic progression in patients with nonmetastatic CRPC (9). More recently the TITAN study in men with mCSPC demonstrated that the addition of apalutamide to ADT significantly extended overall survival (OS), with a 33% reduction in the risk of death, and significantly improved radiographic progression-free survival (rPFS), with a 52% lower risk of radiographic progression or death compared with placebo plus ADT (10).

Abiraterone acetate (AA) is a prodrug of abiraterone, which is an AR signaling inhibitor that blocks adrenal and intratumoral androgen biosynthesis (5). Abiraterone is a specific inhibitor of P-450c17 (CYP17). AA in combination with prednisone (AA-P) significantly improved OS compared with prednisone and placebo in patients with metastatic CRPC (mCRPC) who were either pretreated with docetaxel or docetaxel naïve (11–14). In pivotal phase III trials, patients with mCRPC were treated with AA 1,000 mg daily in combination with prednisone 10 mg daily. The LATITUDE study evaluated AA (1,000 mg daily) plus prednisone (5 mg daily) and demonstrated that the addition of AA-P to ADT significantly improved OS and rPFS in men with newly diagnosed, high-risk mCSPC (15).

Despite the OS benefit observed with AA-P in patients with prostate cancer, primary and secondary resistance to AA-P have both been observed, leading to eventual disease progression (16). Persistent AR signaling may be a major mechanism of resistance to AA-P (16). Apalutamide and AA-P provide clinical benefit in patients with prostate cancer by targeting the AR axis via complementary but different mechanisms. Thus, treatment intensification with the combination of apalutamide and AA-P could delay disease progression more effectively than treatment with either AR signaling inhibitor alone. In a phase II study of patients with mCRPC, apalutamide treatment led to a ≥50% decline in 12-week PSA levels in 88% (22/25) of AA-P–naïve patients and 22% (4/18) of patients with prior exposure to AA-P (17). As apalutamide is a strong CYP3A4 inducer (18), and as CYP3A4 is involved in the metabolism of AA-P, this raises the question of whether coadministration of apalutamide with abiraterone acetate will affect the pharmacokinetics of AA-P treatment (18). In this phase Ib study, we evaluated potential drug–drug interactions and safety of apalutamide in combination with AA-P in patients with mCRPC while describing early evidence of antitumor activity.

This multicenter, open-label, phase Ib drug–drug interaction study (NCT02123758) was conducted at six study centers in the United States (2), Canada (2), the Netherlands (1), and Great Britain (1) from July 10, 2014, to June 27, 2016, in accordance with the principles of Good Clinical Practices and the Declaration of Helsinki. At each participating site, the study protocol was approved by all required regulatory bodies (independent ethics committees, institutional review boards) and conducted with required institutional oversight. All patients provided written informed consent prior to participation.

Patients aged ≥18 years with metastatic histologically or cytologically confirmed adenocarcinoma of the prostate, progressing despite surgical or medical castration (testosterone levels of <50 ng/dL), were enrolled. Key eligibility criteria included an Eastern Cooperative Oncology Group performance status (ECOG PS) score ≤2 and metastatic disease documented by bone scan and CT or MRI scans. No eligibility restrictions were placed on the number of prior hormonal or cytotoxic interventions. Patients with known brain metastases or pathologic findings consistent with small-cell carcinoma of the prostate were excluded from the study. Patients with a history of seizure or conditions that predispose to seizures were also excluded.

This was an open-label, single-sequence crossover study (Supplementary Fig. S1). Patients received AA 1,000 mg daily with prednisone 10 mg (administered as 5 mg twice daily) beginning on cycle 1 day 1 (C1D1). Apalutamide 240 mg once daily was started on cycle 1 day 8 (C1D8). Each treatment cycle was 28 days. Patients who were not surgically castrate continued to receive gonadotropin hormone–releasing hormone analogues at study entry. To be eligible for the study and to avoid interference with the pharmacokinetic results, patients had to be discontinued from AA-P for at least 2 weeks before receiving the first dose of AA-P on C1D1 in the study. Likewise, patients had to be discontinued from enzalutamide for at least 8 weeks before first dose of study drug on C1D1. The long washout for enzalutamide was defined taking into account the enzyme-inducing properties of enzalutamide.

The primary objective of this study was to evaluate the effect of apalutamide on steady-state abiraterone and prednisone pharmacokinetics in patients with mCRPC. Cohort 1 focused on pharmacokinetics of abiraterone and apalutamide; cohort 2 was added to measure pharmacokinetics of prednisone. The pharmacokinetic parameters for abiraterone (cohorts 1 and 2 combined) and prednisone (cohort 2) were assessed when AA-P was administered alone on C1D7 or in combination with apalutamide on C2D8 (day 36).

Plasma samples were analyzed to determine concentrations of abiraterone, apalutamide and its metabolite N-desmethyl apalutamide, prednisone, and prednisolone using validated, specific, and sensitive LC-MS methods. Pharmacokinetic parameters evaluated were maximum plasma concentration (C_max), area under the concentration–time curve from time 0 to 24 hours (AUC_0–24), and minimum plasma concentration (C_min) of abiraterone, apalutamide, and N-desmethyl apalutamide, and C_max and AUC from 0 to 12 hours of prednisone and prednisolone.

All patients who received one or more dose of study medication were analyzed for safety. Safety monitoring included collection of adverse events (AEs), laboratory safety, vital signs, and electrocardiography. Treatment-emergent AEs (TEAE) were coded using the Medical Dictionary for Regulatory Activities; TEAEs and laboratory values were graded according to National Cancer Institute Common Terminology Criteria for Adverse Events, Version 4.03. Patients were monitored for safety during screening and treatment phases and up to 30 days after the last dose of study drugs.

Antitumor activity evidence was evaluated primarily using serial serum PSA measurements. Maximal change in PSA relative to baseline at any time on study was reported for each patient. PSA measurements were performed at screening, on day 1 of cycles 1 to 3, and then on day 1 of every third cycle from cycle 4 onward. Disease imaging frequency was not study mandated and was performed at the investigators' discretion as part of standard of care. Disease progression was based on modified Response Evaluation Criteria in Solid Tumor 1.1 and Prostate Cancer Working Group 2 criteria (19, 20).

Statistical analysis for the primary objective was performed by applying linear mixed model on log-transformed pharmacokinetic parameter data, and the results were presented in original scale after antilog transformation. The linear mixed model included treatment as fixed effect and subject as random effect. Individual and mean plasma concentration and pharmacokinetic parameter estimations were summarized with descriptive statistics for each treatment. All treated patients with baseline measurements and at least one postbaseline efficacy measurement were analyzed for PSA kinetics.

The data sharing policy of Janssen Pharmaceutical Companies of Johnson & Johnson is available at https://www.janssen.com/clinical-trials/transparency. As noted on this site, requests for access to the study data can be submitted through Yale Open Data Access (YODA) Project site at http://yoda.yale.edu.

As of June 27, 2016 (clinical cut-off date), 57 patients with mCRPC had received one or more dose of either AA-P or the combination of apalutamide and AA-P. Baseline demographics and characteristics are listed in Table 1. Despite the nature of their disease and previous treatments, most (96%) had an ECOG PS of 0 to 1. The majority of patients (95%) had received at least one type of prior systemic therapy for mCRPC. More than half of patients (51% [29/57]) had received prior therapy with a taxane (docetaxel, 51%; cabazitaxel, 30%). The majority [74% (42/57)] of patients received systemic anticancer therapy for prostate cancer with an AR axis agents (AA-P or enzalutamide) before entering the study. A total of 29 (51%) patients received prior treatment with AA-P and 23 (40%) patients received prior treatment with enzalutamide. One third (19/57) of patients received prior treatment with AA-P only, 23% (13/57) of patients with enzalutamide only, and 18% (10/57) of patients received prior treatment with both AA-P and enzalutamide.

At the time of the clinical cutoff, 47 (83%) patients had discontinued treatment. Clinical progression (with or without PSA progression) was the most common reason for treatment discontinuation (68%), followed by patient withdrawal (7%), PSA progression (without clinical progression; 3.5%), and physician decision and death (1 patient each). The overall median treatment duration of apalutamide plus AA-P was 17 weeks (range, 3–92).

The mean plasma concentration–time profiles are presented in Fig. 1, and pharmacokinetic analysis results are summarized in Table 2 and Supplementary Table S1. An approximately 23%, 14%, and 5% decrease was observed in abiraterone C_max, AUC_0–24, and C_min, respectively, when apalutamide was administered with AA-P (Fig. 1A; Table 2). Mean plasma concentration–time profiles of prednisone and prednisolone were lower when apalutamide was coadministered with AA-P compared with AA-P alone (Fig. 1B and C). Treatment with apalutamide decreased exposure to prednisone and its metabolite prednisolone by 51% to 61% and 26% to 42%, respectively (Table 2). The pharmacokinetic profiles of apalutamide and N-desmethyl apalutamide following administration of apalutamide 240 mg once daily together with AA-P were similar to those previously reported for apalutamide monotherapy (Fig. 1D) (21).

Fifty-seven patients (100%) received one or more dose of study drug and were thus included in the safety analysis. A summary of TEAEs is presented in Table 3. The majority of TEAEs were grade 1 to 2 in severity. Fatigue was the most common TEAE, affecting more than half of the patients (56%). The frequency of grade 3 fatigue was less than 9% (n = 5). Nausea and vomiting were reported in 40% of patients, but only one patient experienced grade 3 nausea. Other commonly reported TEAEs (≥25% of patients) included back pain (33%), decreased appetite (32%), arthralgia (28%), constipation (30%), and diarrhea (26%). Hypokalemia was reported in 25% of patients, with grade 3 or higher hypokalemia reported in 7% of patients. A grade 4 hypokalemia event was observed in one patient and occurred with associated vomiting. Hypokalemia events were managed with potassium supplementation. Hypertension events (7%) of grade 2 or grade 3 severity were reported in four patients with a history of hypertension. Grade 3 TEAEs reported in more than one patient included fatigue, hypokalemia, pneumonia, hyponatremia, hypertension, back pain, and fall. Grade 4 TEAEs included septic shock and fall, hypokalemia, and pyrexia (one patient each). One patient experienced grade 3 fatigue resulting in permanent discontinuation of apalutamide. This patient was maintained on AA-P.

Twenty-two (39%) patients reported one or more serious AE (SAE) during the study. SAEs that occurred in more than one patient included pneumonia (n = 4), back pain (n = 3), and constipation, pyrexia, and fall (n = 2 each), the majority of which were not considered related to study drugs. Six deaths were reported in the study. All 6 patients died from disease progression after discontinuation of therapy. None of the deaths were considered related to study treatment.

Fifty-seven patients received one or more dose of study treatment and thus were assessed for potential antitumor activity. Eighteen (32%) patients had a confirmed PSA response of ≥50% reduction from baseline at any time, and six (11%) patients had a confirmed PSA response of ≥90% reduction from baseline at any time. Six of 42 (14%) AR signaling inhibitor–treated patients benefited from this combination as evidenced by confirmed PSA responses of ≥50% reduction from baseline. In contrast, 80% (12/15) of AR signaling inhibitor–naïve patients had a confirmed PSA response of ≥50% reduction from baseline (Fig. 2). Further breakdown of ≥50% and ≥90% PSA reduction responses for those patients who had received prior therapy with an AR signaling inhibitor are presented in Supplementary Table S2 and Supplementary Fig. S2. None of the patients who received prior therapy with AA-P or enzalutamide achieved a confirmed PSA response of ≥90% reduction from baseline while 6 (40%) patients naïve to AR signaling inhibitors had a confirmed PSA response of ≥90%.

Patients naïve to AR signaling inhibitors were more likely to have a PSA decline than were patients who had received prior AR signaling inhibitors (Fig. 2). Moreover, AR signaling inhibitor–naïve patients had a greater mean decrease in PSA than did those who had received prior therapy. Overall, the shortest treatment durations were observed in patients pretreated with AA-P or enzalutamide and who had previously received taxane-based chemotherapy (Fig. 3). Patients who received prior AA-P (n = 19) or prior enzalutamide (n = 13) had a median duration of treatment of 28.0 weeks and 15.9 weeks, respectively (Supplementary Table S2). The median treatment duration for patients who had previously received both AA-P and enzalutamide (n = 10) was 10.4 weeks. Among AR signaling inhibitor–naïve patients (n = 15), the median duration of treatment was 36 weeks.

This phase Ib study in patients with mCRPC demonstrated that apalutamide combined with AA-P was well tolerated and, based on serial PSA monitoring, provided evidence of potential antitumor activity in 80% of AR signaling inhibitor–naïve patients and in 14% of patients who had received prior AR signaling inhibitors.

AA is a substrate of CYP3A4 (22), and prednisone, upon conversion to the active moiety prednisolone, is further metabolized by CYP3A4 (23). Because apalutamide is a strong CYP3A4 inducer, the involvement of CYP3A4 in the metabolism of AA-P suggests that pharmacokinetic exposure of abiraterone and prednisone may be lowered when AA-P is coadministered with apalutamide. In this study, apalutamide decreased the AUC of abiraterone by approximately 14%. This degree of change in abiraterone exposure was within the intrasubject variability of abiraterone pharmacokinetics (19% for AUC observed in healthy subjects or 70% in patients with mCRPC when AA-P was administered alone) (24). Thus, the modest reduction in abiraterone exposure is not hypothesized to be clinically relevant.

Apalutamide is metabolized to its active metabolite, N-desmethyl apalutamide, through CYP3A4 and CYP2C8 (18). Abiraterone is a weak inhibitor of CYP2C8 in vivo, and prednisone 10-mg daily does not cause meaningful effects on CYP3A4 (25); therefore, AA-P treatment was not hypothesized to affect the metabolism of apalutamide. The results herein demonstrate that the pharmacokinetics of apalutamide, when administered in combination with AA-P, were similar to those when apalutamide is administered as a monotherapy (21), thereby confirming the lack of effect of AA-P on the pharmacokinetics of apalutamide.

In this study, apalutamide decreased the AUC of prednisone and prednisolone by approximately 50%. The frequency of mineralocorticoid excess–related AEs did not appear to be increased substantially in this study compared with that reported in AA-P arms of pivotal phase III trials (10-mg prednisone daily refs.) (12, 13, 17). Moreover, results from the phase III STAMPEDE (26) and LATITUDE (15) studies demonstrate that a 5-mg dose of prednisone is acceptable. However, when apalutamide is administered with AA-P, a dose of 5-mg prednisone may be insufficient. The pharmacokinetic interaction between apalutamide and prednisone is not a concern when a 10-mg daily dose of prednisone is used with apalutamide in combination with AA-P.

As hypothesized, apalutamide in combination with AA-P demonstrated more antitumor activity in AR signaling inhibitor–naïve patients and in less heavily pretreated patients. Patients who were pretreated with an AR signaling inhibitor and had previously received taxane-based chemotherapy had the shortest median treatment duration and lowest likelihood of achieving a maximal PSA response as change from baseline at any time. This observation suggests that a heavily pretreated population of patients with mCRPC may be less likely to benefit from combination treatment of apalutamide plus AA-P (Fig. 2B).

The Alliance for Clinical Trials in Oncology recently presented results from A031201 (NCT01949337). This phase III trial showed that the addition of AA-P to enzalutamide did not prolong survival in men with mCRPC compared with enzalutamide alone (27). However, there are important differences between the Alliance trial and the phase III ACIS study as well as our study. The control arm in ACIS received AA-P. In addition, it will be important to review differences in the patient populations as the majority of patients in A031201 were at low risk according to Halabi criteria. Finally, patients randomized to the combination regimen in this cooperative group were not provided AA-P by the sponsor and drugs were obtained on a subject-by-subject basis. In contrast, a company-sponsored study provides the drug to patients and tracks patients' drug compliance and disposition.

Although the safety, pharmacokinetics, and potential antitumor activity of apalutamide are evident from our results, there are certain limitations to this study. First, the sample size was relatively small. Second, as this was a phase Ib study, it had a single treatment arm. Finally, the study population was heterogeneous with regard to prior therapies and disease burden.

However, the documented tolerability, lack of clinically relevant pharmacokinetic interaction, and antitumor activity of apalutamide combined with AA-P in this phase Ib study support further clinical development in patients with mCRPC. The phase III randomized ACIS trial that examines apalutamide in combination with AA-P (10-mg daily dose of prednisone) in chemotherapy-naïve patients with mCRPC, with a primary end point of radiographic progression-free survival (NCT02257736), is currently ongoing.

In conclusion, the combination of apalutamide and AA-P was well tolerated and provided some evidence of potential antitumor activity without a clinically meaningful decrease in AA-P bioavailability or obvious differences in toxicity compared with the historical safety profile of either AR signaling inhibitor treatment alone among patients with mCRPC. These data support the use of a 10-mg daily dose of prednisone and the daily dosing of 1,000 mg of AA with 240 mg of apalutamide.

E.M. Posadas is an unpaid consultant/advisory board member for Janssen. K.N. Chi is a paid consultant for Janssen, Astellas, Sanofi, Bayer, Roche, AstraZeneca, Pfizer, Point Biopharma, Daiichi Sankyo, and Bristol-Myers Squibb; reports receiving commercial research grants from Janssen; and reports receiving speakers bureau honoraria from Janssen and Astellas. R. de Wit is a paid consultant for Janssen, Merck, and Sanofi. G. Attard is a paid consultant for Janssen, AstraZeneca, Pfizer, Astellas, Essa Pharmaceuticals, Bayer Pharmaceuticals, Veridex, Ventana, Novartis, and Sanofi; reports receiving commercial research grants from Janssen, AstraZeneca, Innocrin Pharma, and Arno Therapeutics; reports receiving speakers bureau honoraria from Janssen, Astellas, Pfizer, Bayer Pharmaceuticals, Janssen, AstraZeneca, and Sanofi; and holds ownership interest (including patents) in Abiraterone. T.W. Friedlander is a paid consultant for Janssen, AbbVie, and Dendreon, and reports receiving commercial research grants from Janssen. M.K. Yu is an employee of and holds ownership interest (including patents) in Janssen R&D. P. Hellemans is an employee of and holds ownership interest (including patents) in Johnson & Johnson. C. Chien was formerly an employee of and holds ownership interest (including patents) in Johnson & Johnson. J.J. Jiao is an employee of Johnson & Johnson. F. Saad reports receiving speakers bureau honoraria from and is an unpaid consultant/advisory board member for Janssen. No potential conflicts of interest were disclosed by the other authors.

Conception and design: R. de Wit, M.J.A. de Jonge, T.W. Friedlander, M.K. Yu, P. Hellemans, C. Chien, J.J. Jiao, F. Saad

Development of methodology: R. de Wit, G. Attard, T.W. Friedlander, P. Hellemans, C. Chien, J.J. Jiao, F. Saad

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): E.M. Posadas, K.N. Chi, R. de Wit, M.J.A. de Jonge, G. Attard, T.W. Friedlander, F. Saad

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): E.M. Posadas, K.N. Chi, R. de Wit, M.J.A. de Jonge, G. Attard, T.W. Friedlander, P. Hellemans, C. Chien, J.J. Jiao, F. Saad

Writing, review, and/or revision of the manuscript: E.M. Posadas, K.N. Chi, R. de Wit, M.J.A. de Jonge, G. Attard, T.W. Friedlander, M.K. Yu, P. Hellemans, C. Chien, C. Abrams, J.J. Jiao, F. Saad

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): T.W. Friedlander, F. Saad

Study supervision: K.N. Chi, T.W. Friedlander, M.K. Yu, C. Abrams, F. Saad

The authors would like to thank the patients who participated in this trial, and their families, as well as the investigators, study coordinators, study teams, and nurses. Scientific and statistical input was provided by Vijay Chauhan, Martha Gonzalez, and Géralyn Trudel of Janssen Research & Development. This study was funded by Janssen Research & Development. Editorial assistance was provided by Ann P. Tighe, PhD, of Parexel, with funding from Janssen Global Services, LLC.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.