Precision Medicine Approaches When Prostate Cancer Akts Up

In this issue of Clinical Cancer Research, de Bono and colleagues report the results of a randomized phase Ib/2 trial of abiraterone acetate with or without the Akt inhibitor, ipatasertib (GDC-0980), in men with metastatic castration-resistant prostate cancer (mCRPC; ref. 1). PTEN loss due to deletion, structural variation, or mutation is prevalent in mCRPC, present in 40%–60% of patients with mCRPC (2). PTEN is a tumor suppressor that usually inhibits the PI3K–Akt–mTOR pathway. When PTEN expression is lost, either through genomic or nongenomic means, there is constitutive signaling through PI3K and Akt, resulting in downstream mechanisms of tumorigenesis such as activation of the cell cycle and proliferation, increased metabolic pathways such as glycolysis, protein translation, and cell growth, and upregulation of angiogenic and cell survival pathways (Fig. 1A). PTEN loss can occur in primary tumors, but is more often found in mCRPC, supporting its role in disease progression; in addition, patients with PTEN loss in primary tumors have a more aggressive course of disease compared with patients with PTEN-intact tumors.

Preclinical modeling of PTEN loss in prostate cancer has underscored the reciprocal regulation of the PI3K–Akt–mTOR and androgen receptor (AR) pathways. When PI3K is inhibited, feedback upregulation and activation of receptor kinase receptors such as insulin growth factor (IGF) or HER2/3 among others leads to the activation of AR. When AR is inhibited, such as with the AR-targeting therapies abiraterone acetate or enzalutamide, downstream transcription targets such as FKBP5 are downregulated, impairing PHLPP function and increasing phosphorylated Akt (Fig. 1A; ref. 3). These reciprocal interactions limit the clinical efficacy of single-agent approaches to pathway engagement.

Given our increasing understanding of the basic signaling mechanisms important in mCRPC, several treatment strategies targeting the PI3K–Akt–mTOR have been studied in patients with mCRPC. In a phase II trial of mCPRC, temsirolimus (an intravenous mTOR inhibitor) showed PSA responses in 4 of 15 patients (28.5%) but median time to PSA progression of only 2 months and 52% of patients with serious adverse events (AE; ref. 4). In addition, buparlisib (BKM120, a pan-PI3K inhibitor) was studied as monotherapy and in combination with enzalutamide in men with mCRPC, and did not show significant disease activity [median progression-free survival (PFS) for monotherapy was 1.9 months and 3.5 months with concurrent enzalutamide], with no responses observed. Concurrent enzalutamide had significant drug–drug interactions through the cytochrome P450 pathway induction by enzalutamide, leading to an 80% reduction in serum levels of buparlisib. In addition, 43% of patients had grade 3 AEs, most common including fatigue, anemia, hyperglycemia, confusion, myalgia, and urinary tract obstruction (5). Of note, although none of the patients in these studies were specifically selected on the basis of PTEN loss, no responses were observed, supporting the concept that targeting multiple reciprocal pathways is necessary to prevent the rapid emergence of drug resistance.

Ipatasertib (GDC-0068) has been developed as a competitive ATPase inhibitor of Akt. Careful preclinical work showed inhibition of key downstream Akt targets mTOR, eIF4G, and S6, as well as metabolic targets pFOXO3a, pGSK3b, and pATP citrate lyase; in addition, the PTEN-deficient PC3 prostate cancer xenograft models showed dose-dependent inhibition of tumor growth (6). A phase I study of 9 patients showed that ipatasertib was tolerable at doses of 100 mg, 200 mg, and 400 mg daily, with main side effects of diarrhea, asthenia/fatigue, hyperglycemia, and rash (6). On-treatment biopsies, particularly in those PTEN-null patients, demonstrated significant increases in tumor necrosis (6).

In this issue of Clinical Cancer Research, de Bono and colleagues report on an international, multicenter, randomized phase II study of abiraterone acetate with ipatasertib 400 mg daily, ipatasertib 200 mg daily, or placebo. The coprimary endpoints were radiographic PFS (rPFS) in the intention to treat population, as well as patients with PTEN loss. An improvement in median rPFS of 3 months was considered clinically meaningful for future development in phase II. PTEN status was primarily detected via an IHC assay for protein loss at the Institute of Cancer Research (ICR), and also verified through 3 other assays testing for PTEN loss at both protein and DNA levels. Importantly, study investigators and patients were blinded to PTEN status, and PTEN status was mostly determined from primary prostate biopsies rather than new metastatic biopsies. Patients were stratified by prior enzalutamide treatment, number of prior chemotherapy regimens, and type of progression (PSA progression only vs. other).

The study randomized 253 patients; all patients received abiraterone acetate and in addition, 84 patients received ipatasertib 400 mg, 86 patients received ipatasertib 200 mg, and 83 patients received placebo. Baseline characteristics were mostly similar across treatment arms, with more chemotherapy refractory patients in the placebo-treated arm (25% vs. 18%–20%) and more patients with metastatic disease at diagnosis (68% vs. 51%–56%) and higher mean enrollment PSA (379 μg/L vs. 230–261 μg/L) in the ipatasertib 400 mg treated arm (1).

For the primary outcome, rPFS did not significantly differ (HR 0.75, P = 0.17) between treatment cohorts for the intention-to-treat population, suggesting that combined AR and Akt inhibition may be ineffective in unselected men with mCRPC (1). However, for the 43% of patients with PTEN protein loss, analyzed as a coprimary subgroup, the addition of ipatasertib 400 mg daily improved median rPFS from 4.6 months to 11.5 months [HR 0.39; 90% confidence interval (CI), 0.22–0.70], whereas PTEN-intact patients had no significant benefits (HR 0.84; 90% CI, 0.51–1.37) from the addition of ipatasertib (1). Of note, there were discordances between PTEN loss at the DNA level (36%–44%) and PTEN loss at the protein level (43%–56%), indicating likely epigenetic mechanisms of PTEN loss in some men (1). This enrichment for PTEN loss did not lead to improvements in PSA response suggesting its effects are AR independent. The study was underpowered for overall survival (OS) differences.

The positive effects on rPFS were tempered by AEs. Sixty-four percent of patients treated with ipatasertib 400 mg daily with abiraterone developed grade 3 or higher AEs, most commonly diarrhea, rash, hyperglycemia, pneumonia, asthenia, and fatigue, with 14 deaths (5.5%) attributed to AEs that ranged from bradycardia, sepsis, aortic aneurysm rupture, and heart failure. These grade 5 events were not felt by investigators to be related to ipatasertib.

In the age of precision medicine, selecting therapeutic targets within patients with mCRPC beyond AR has been challenging despite promising candidates (Fig. 1B). This trial of ipatasertib specifically found clinically meaningful activity of ipatasertib in PTEN-deficient mCRPC patients. In addition, the investigators capitalized on the molecular understanding of the reciprocal regulation of AR and PI3K, inhibiting both with abiraterone and ipatasertib, respectively. Unlike the prior trial with buparlisib and enzalutamide, there were much fewer drug interactions between ipatasertib and abiraterone. This trial thus represents an early success in the evolution of treatments targeting the PI3K–Akt–mTOR pathway in mCRPC. However, caution must be noted, given the mixed results in this smaller study of ipatasertib; although these justify the larger ongoing phase III study (NCT03072238), we will need to see prospective confirmation. Unfortunately, OS is a secondary endpoint in this larger 1,100 patient randomized study, despite rPFS lacking clear surrogacy in this setting for OS. A host of other mechanisms of resistance to abiraterone in mCRPC is likely to be important for further cancer adaptation, including alternatively spliced AR variants such as AR-V7, AR-indifferent disease leading to lineage plasticity, and alternative reciprocal pathways that emerge from combined blockade. Further work is likely needed to examine resistance mechanisms through serial biopsies and tumor evolution assessments during combined Akt/AR inhibition to improve OS in men with mCRPC.

T. Zhang reports receiving commercial research grants from Abbvie, Acerta Pharma, Janssen, Merck, Merrimack, Novartis, OmniSeq, Pfizer, and PGDx, speakers bureau honoraria from Exelixis and Genentech, and is a consultant/advisory board member for AstraZeneca, Foundation Medicine, Janssen, Pfizer, and Sanofi Aventis. D.J. George reports receiving other commercial research support from AstraZeneca, Bayer Pharmaceuticals, Dendreon, Innocrin, Janssen Pharmaceuticals, Pfizer, and Sanofi, speakers bureau honoraria from Bayer Pharmaceuticals and Sanofi, and is a consultant/advisory board member for Astellas/Medication, AstraZeneca, Bayer Pharmaceuticals, Innocrin, Janssen Pharmeceuticals, Myovant Sciences, Pfizer, and Sanofi. A.J. Armstrong reports receiving commercial research grants from Dendreon, Genentech, Janssen, Novartis, Pfizer, and Sanofi Aventis, and is a consultant/advisory board member for AstraZeneca, Bayer, and Clovis. No other potential conflicts of interest were disclosed.

Conception and design: T. Zhang, A.J. Armstrong

Development of methodology: A.J. Armstrong

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A.J. Armstrong

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): T. Zhang, A.J. Armstrong

Writing, review, and/or revision of the manuscript: T. Zhang, D.J. George, A.J. Armstrong

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A.J. Armstrong

Study supervision: A.J. Armstrong