Clinical Outcomes of Chemotherapy Naïve Men with Metastatic Castration Resistant Prostate Cancer and Low Baseline Prostate Specific Antigen Treated with Enzalutamide vs Placebo
The Journal of Urology
Mary-Ellen Taplin, Andrew J. Armstrong, Ping Lin, Andrew Krivoshik, De Phung, Teresa Parli, Bertrand Tombal and Tomasz M. Beer
Purpose: Metastatic castration resistant prostate cancer with low baseline prostate specific antigen represents an early stage in the natural history of castration resistant prostate cancer progression (low volume disease), low prostate specific antigen producing disease or disease that is less dependent on androgen receptor biology (high volume disease). We analyzed outcomes in men with low prostate specific antigen and a high disease burden who received the oral androgen receptor inhibitor enzalutamide in the PREVAIL (Safety and Efficacy Study of Oral MDV3100 in Chemotherapy-Naive Patients with Progressive Metastatic Prostate Cancer) study.
Materials and Methods: In this exploratory analysis low baseline prostate specific antigen was defined as less than 10 ng/ml. Post hoc analyses included radiographic progression-free and overall survival in the once daily enzalutamide and placebo arms. Patients were stratified post hoc by high volume disease, defined as more than 4 bone metastases and/or visceral disease, and low volume disease, defined as 4 or fewer bone metastases with no visceral disease.
Results: Of 1,717 patients enrolled in PREVAIL 242 (14.1%) had low baseline prostate specific antigen, including 110 with high volume disease. Enzalutamide decreased the risk of radiographic progression relative to placebo in patients with low baseline prostate specific antigen (HR 0.20, 95% CI 0.10–0.42). This decrease was irrespective of tumor burden (high volume disease HR 0.17, 95% CI 0.06–0.51 and low volume disease HR 0.25, 95% CI 0.09–0.70). Median overall survival was not reached in patients with low baseline prostate specific antigen in either treatment arm.
Chemotherapy naïve men with metastatic castration resistant prostate cancer and low baseline prostate specific antigen irrespective of disease burden may benefit from enzalutamide. This indicates that targeting the androgen receptor signaling pathway is a therapeutic option in similar patients.
Castration resistant prostate cancer is characterized by disease progression following medical or surgical castration. 1,2Disease progression in the castration resistant state can manifest as a rise in serum PSA level, metastases to bone, lymph nodes or visceral sites, or a combination of these conditions. 3Thus, patients with mCRPC are a heterogeneous population regarding clinical characteristics and underlying disease biology. 3–5
mCRPC disease progression is often accompanied by increasing serum PSA due to continued engagement of the AR signaling pathway. In some patients mCRPC develops with low PSA levels, which may signify early onset classic mCRPC and/or indicate the presence of androgen or AR independent disease, termed neuroendocrine, anaplastic, small cell, aggressive variant and nonadeno. 3Analysis of mCRPC in patients with clinically defined anaplastic features revealed a high frequency of combined tumor suppressor defects in RB1, Tp53 and PTEN. 6In contrast, analysis of resistant tumors with pathological neuroendocrine features suggested that neuroendocrine prostate cancer evolves from 1 or more prostate adenocarcinomas, characterized by an AR indifferent cell state that is resistant to AR targeted therapies. 7An analysis of patients with mCRPC who had low PSA level and a high or low metastasis burden provides an opportunity to assess the response to AR inhibition at earlier or later states in the disease continuum and also provides guidance to select mCRPC therapy.
In the phase III PREVAIL trial the oral AR inhibitor enzalutamide significantly improved OS and rPFS vs placebo in men with chemotherapy naïve mCRPC. 8The study enrolled 1,717 asymptomatic or mildly symptomatic patients who had progressed on ADT based on imaging and/or PSA. 8Neuroendocrine or small cell features were among the study exclusion criteria. However, the histological status of mCRPC tissue was not known.
PREVAIL patients were a heterogeneous cohort as evidenced by the baseline prevalence of bone disease (83%), lymph node disease (51%) or visceral disease (12%). 8PREVAIL subgroup analyses that were not focused on patients with low PSA revealed that enzalutamide decreased the risk of radiographic and PSA progression in men with liver and lung metastases, lung only metastases, and bone and/or nodal disease. 9,10
In this post hoc analysis we evaluated clinical outcomes in the subgroup of men in PREVAIL who had low baseline PSA (ie less than 10 ng/ml) with a low or a high metastatic burden. 11We hypothesized that men with low PSA and a low disease burden would respond to enzalutamide at least as well as the entire PREVAIL cohort and men with low PSA and a high disease burden might not have a clinical benefit with enzalutamide, reflecting potential nonAR related tumor growth.
Materials and Methods
The full design and methodology have been published previously. 8The study was approved by the independent review board at each participating site and performed according to the provisions of the Declaration of Helsinki and the GCP (Good Clinical Practice) guidelines of ICH (International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use). All patients provided written informed consent before participation.
Patients were randomized 1:1 to oral enzalutamide (160 mg) or placebo once daily until treatment was discontinued because of intolerance, confirmed radiographic progression or a skeletal related event plus initiation of another therapy for prostate cancer. All patients were required to remain on ADT while being treated in the study.
We evaluated the co-primary end points of rPFS and OS, and the secondary end points of best overall soft tissue response, time to PSA progression and PSA response in the subgroup of patients with low baseline PSA (ie less than 10 ng/ml). A minimum PSA level of 2.0 ng/ml was required for study eligibility. The cutoff PSA of less than 10 ng/ml for the current analysis was chosen by author consensus as to our knowledge there is no existing definition for the low PSA state in mCRPC. A smaller subgroup analysis was also performed in men with baseline PSA less than 5 ng/ml and high volume disease.
We defined rising PSA as the best PSA response greater than baseline. Similarly since there is no established definition of high vs low volume disease in mCRPC, high volume was defined as more than 4 bone metastases and/or visceral disease, and low volume was defined as 4 or fewer bone metastases with no visceral disease.
The co-primary end point rPFS was defined as time from randomization to the first objective evidence of radiographic disease progression, which was assessed at a blinded independent central review facility, or death from any cause within 168 days after treatment discontinuation, whichever occurred first. Additional post hoc analyses featured evaluation of the co-primary and secondary end points in the subset of patients with low baseline PSA and high vs low volume disease in each subgroup at baseline.
Estimates of the median and 95% CI for time to event analyses were determined by the Kaplan-Meier method. The HR was determined using an unstratified Cox regression model with treatment as the only covariate and it was relative to placebo with a value of the upper 95% CI limit less than 1.00, indicating a treatment benefit in favor of enzalutamide. Since there were few missing values, no imputation techniques were done. Data were analyzed with SAS®, version 9.4.
Patients with Low Baseline Prostate Specific Antigen and Treatment
Of 1,717 patients 242 (14.1%) entered PREVAIL with PSA less than 10 ng/ml, including 110 with high volume but low PSA disease and 29 with visceral metastases. Overall men with mCRPC and low baseline PSA had more favorable prognostic characteristics for OS than the overall PREVAIL population, including lower levels of alkaline phosphatase and lactate dehydrogenase, lower pain scores, higher levels of hemoglobin and albumin, younger age and improved functional status ( table 1). 12–14Patients with low baseline PSA and high volume disease had characteristics similar to those of the overall PREVAIL population.
Among these patients with low baseline PSA, baseline demographic and disease characteristics were similar in the 122 in the enzalutamide group and the 120 in the placebo group ( table 1). The extent of baseline disease was well balanced between the 2 treatment groups with similar proportions of patients who had bone, lymph node or visceral (lung and/or liver) metastases. The median number of bone metastases in the enzalutamide and placebo groups was 2 (IQR 1–5). An equal percent of patients in the enzalutamide and placebo groups had bone disease at baseline (84.4% or 103 of 122 and 84.2% or 101 of 120, respectively).
The rPFS rate at 12 months was 78% and 42% in the low baseline PSA enzalutamide and placebo subgroups compared with 65% and 14% in the enzalutamide and placebo groups, respectively, of the overall PREVAIL population. 8Among men with low baseline PSA the median time that patients received a study drug was substantially longer in the enzalutamide group than in the placebo group (17.8 months, IQR 11.3–23.2 vs 7.8, IQR 3.7–14.7). This was longer than that observed in the overall PREVAIL population (enzalutamide 16.6 months, IQR 10.1–21.1 and placebo 4.6 months, IQR 2.8–9.7). 8The trend was observed in patients with low baseline PSA and high volume disease (enzalutamide 17.3 months, IQR 8.3–23.2 and placebo 8.3, IQR 3.6–13.8) and low volume disease (17.9 months, IQR 13.5–23.9 and 6.9, IQR 3.7–15.0, respectively).
Radiographic Progression-Free and Overall Survival in Patients with Low Baseline Prostate Specific Antigen
In patients with low baseline PSA median rPFS was not reached in the enzalutamide group while it was 8.3 months in the placebo group ( fig. 1). Thus, treatment with enzalutamide decreased the risk of radiographic progression relative to placebo (HR 0.20, 95% CI 0.10–0.42). This finding was consistent with the result in the overall PREVAIL population (enzalutamide not reached vs placebo 3.9 months, HR 0.19, 95% CI 0.15–0.23). 8Kaplan-Meier analysis of rPFS revealed a clear separation between the enzalutamide and placebo groups in patients who had low baseline PSA with high volume disease (HR 0.17, 95% CI 0.06–0.51) and with low volume disease (HR 0.25, 95% CI 0.09–0.70, fig. 1, B).
Median OS was not yet reached in the enzalutamide or placebo group among patients with low baseline PSA. Overall survival data were immature to analyze at this time point in patients with low baseline PSA.
Of patients with low baseline PSA and measurable soft tissue disease at baseline 15 of 37 (40.5%) in the enzalutamide group and 6 of 33 (18.2%) in the placebo group achieved an objective response, of which most were a partial response ( table 2). Similar findings were observed in patients with at least 1 target or nontarget lesion at screening except the proportions of patients in the enzalutamide group who achieved a complete or a partial response were approximately the same.
Time to Prostate Specific Antigen Progression and Response
Among patients with low baseline PSA median time to PSA progression was 19.4 months in the enzalutamide group and 3.0 months in the placebo group (HR 0.17, 95% CI 0.12–0.25). These intervals were longer than that in the overall PREVAIL population (enzalutamide 11.2 months vs placebo 2.8, HR 0.17, 95% CI 0.15–0.20). 8Kaplan-Meier analysis of time to PSA progression revealed a clear separation in favor of enzalutamide over placebo among patients with low baseline PSA and high volume or low volume disease ( fig. 2, A).
A confirmed PSA response (at least a 50% reduction) in patients with low baseline PSA was achieved in 98 of 119 (82.4%) treated with enzalutamide compared with 9 of 116 (7.8%) treated with placebo for a 74.6% difference (95% CI 66.2–83.0). These proportions were slightly higher than in the enzalutamide group (78.0% or 666 of 854 patients) and the placebo group (3.5% or 27 of 777) in the overall PREVAIL population, although the between group difference in the PSA response was similar (74.5% difference, 95% CI 71.5–77.6). 8A similar trend was observed regarding PSA decreases of at least 30% and at least 90% in the entire low baseline PSA population as well as in those with low baseline PSA who had high and low volume disease ( table 3). Figure 2, B and C shows a PSA waterfall plot indicating that regardless of disease volume more patients in the enzalutamide group than in the placebo group who had low baseline PSA responded to treatment.
The incidence of rising PSA (ie the best PSA response greater than baseline) was comparable in the low baseline PSA cohort and the total PREVAIL population in the enzalutamide arm (5.9% or 7 of 119 patients and 8.2% or 70 of 854) and in the placebo arm (69.8% or 81 of 116 and 81.5% or 633 of 777, respectively).
Patients with Baseline Prostate Specific Antigen Less Than 5 ng/ml and High Volume Disease
The same trend in favor of enzalutamide in 18 patients vs placebo in 23 regarding rPFS benefit, time to PSA progression and PSA response (at least 30%, 50% and 90% reductions) was also detected in the 41 patients with baseline PSA less than 5 ng/ml who had high volume disease.
In PREVAIL patients with low PSA (less than 10 ng/ml) at the start of the study achieved a substantial rPFS benefit with enzalutamide regardless of disease burden with treatment effects similar to those observed in the full study population. 8The extended time to radiographic progression in the enzalutamide group was accompanied by a high objective response rate relative to the placebo group. Additionally, in this population of patients with low PSA at baseline PSA decreases from baseline were seen in almost all men treated with enzalutamide with at least a 50% confirmed PSA response in 82.4%. Taken together these data indicate that most men enrolled in PREVAIL had AR dependent mCRPC even when the disease burden appeared disproportionately high compared to the serum PSA concentration. Thus, clinical features like PSA level and disease burden are not sufficient to prospectively identify patients with an AR independent phenotype. Metastasis biopsy is recommended when these variants are suspected.
Our findings question the view that low PSA in progressing patients with mCRPC who have more than 4 bone metastases and/or visceral disease is likely indicative of nonAR driven growth. 3,15A lack of decreasing PSA with enzalutamide was not materially different between the low baseline PSA cohort and the total PREVAIL population (5.9% vs 8.2%). Time to PSA progression and rPFS outcomes in patients with low baseline PSA were better than those observed in the full study population regardless of treatment assignment. The low PSA subgroup with high and with low volume disease had similar or more favorable prognostic characteristics compared to the overall study population, of which each was associated with improved OS. This suggests that this subset was not far along in the mCRPC continuum.
The eligibility criteria for this trial required a minimum PSA of 2.0 ng/ml and excluded known small cell or neuroendocrine histological variants, which limited the extrapolation of these data in these patients. Therefore, the low PSA in these patients reflected an earlier disease state. Nevertheless, patients with high volume disease and PSA less than 10 ng/ml at baseline had a favorable response to enzalutamide relative to placebo with respect to rPFS and time to PSA progression. Additionally, rPFS and time to PSA progression among enzalutamide recipients in this subgroup were similar to those in the entire enzalutamide treatment group in PREVAIL. 8Thus, the patients in our analysis likely continued to have hormone sensitive disease despite established metastases and low PSA rather than belonging to a cohort with distinct nonAR signaling tumor biology.
Lacking metastatic biopsies, it is not possible to assess histological transformation, tumor AR and AR variant status or other biomarkers of AR independence on outcomes. Because PSA is AR regulated, it is an attractive candidate biomarker of AR pathway activation in patients with mCRPC. However, our findings question the clinical usefulness of serum PSA levels as a measure of AR dependence in mCRPC. The lower PSA seen in these patients may reflect relatively lower AR signaling which remains sufficient to support tumor growth and may indicate that AR driven tumor growth remains intact but PSA expression has been lost. 16,17PREVAIL did not have enough participants with an even higher burden of metastasis or aggressive visceral disease and low PSA to further the analysis.
Our results suggest that although serum PSA may be a useful measure of AR signaling and the efficacy of anti-AR therapy in individual patients, it is neither a robust predictor of the benefit of AR targeted therapy in the general mCRPC patient population nor a reliable measure of the degree to which individual tumors rely on androgens for growth. Molecular analysis of tumor specimens from patients with mCRPC and low PSA is required to reveal the correlation between tumor subtypes and the clinical response to AR directed therapy.
Patients with mCRPC and low baseline PSA who had a low or a high metastatic burden and were enrolled in PREVAIL achieved substantial benefit from treatment with enzalutamide. The finding suggests that this subgroup of patients remained sensitive to this AR targeted therapy. More broadly this analysis highlights that PSA level is only one of many important clinical factors that must be considered when selecting the most appropriate treatments for patients with mCRPC. This analysis provides practicing physicians with prognostic landmarks for patients with low PSA metastatic prostate cancer who are considering second-generation ADT.
Dr. Malcolm Darkes and Shannon Davis, Ashfield Healthcare Communications, provided medical writing and editorial support.
1. Cookson MS, Roth BJ, Dahm P et al: Castrationresistant prostate cancer: AUA guideline. J Urol 2013; 190: 429.
2. Kirby M, Hirst C and Crawford ED: Characterising the castration-resistant prostate cancer population: a systematic review. Int J Clin Pract 2011; 65: 1180.
3. Zhang T and Armstrong AJ: Clinical phenotypes of castration-resistant prostate cancer. Clin Adv Hematol Oncol 2013; 11: 707.
4. Scher HI and Heller G: Clinical states in prostate cancer: toward a dynamic model of disease progression. Urology 2000; 55: 323.
5. Scher HI and Sawyers CL: Biology of progressive, castration-resistant prostate cancer: directed therapies targeting the androgen-receptor signaling axis. J Clin Oncol 2005; 23: 8253.
6. Aparicio AM, Shen L, Tapia EL et al: Combined tumor suppressor defects characterize clinically defined aggressive variant prostate cancers. Clin Cancer Res 2016; 22: 1520.
7. Beltran H, Prandi D, Mosquera JM et al: Divergent clonal evolution of castration-resistant neuroendocrine prostate cancer. Nat Med 2016; 22: 298.
8. Beer TM, Armstrong AJ, Rathkopf DE et al: Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med 2014; 371: 424.
9. Alumkal JJ, Chowdhury S, Loriot Y et al: Effect of visceral disease site on outcomes in patients with metastatic castration-resistant prostate cancer treated with enzalutamide in the PREVAIL Trial. Clin Genitourin Cancer 2017; doi: 10.1016/ j.clgc.2017.02.007.
10. Evans CP, Higano CS, Keane T et al: The PREVAIL study: primary outcomes by site and extent of baseline disease for enzalutamide-treated men with chemotherapy-naive metastatic castration-resistant prostate cancer. Eur Urol 2016; 70: 675.
11. Sella A, Konichezky M, Flex D et al: Low PSA metastatic androgen-independent prostate cancer. Eur Urol 2000; 38: 250.
12. Armstrong AJ, Garrett-Mayer ES, Yang YC et al: A contemporary prognostic nomogram for men with hormone-refractory metastatic prostate cancer: a TAX327 study analysis. Clin Cancer Res 2007;
13: 6396. 13. Halabi S, Small EJ, Kantoff PW et al: Prognostic model for predicting survival in men with hormone-refractory metastatic prostate cancer. J Clin Oncol 2003; 21: 1232.
14. Smaletz O, Scher HI, Small EJ et al: Nomogram for overall survival of patients with progressive metastatic prostate cancer after castration. J Clin Oncol 2002; 20: 3972.
15. Beltran H, Tomlins S, Aparicio A et al: Aggressive variants of castration-resistant prostate cancer. Clin Cancer Res 2014; 20: 2846.
16. Nelson PS: Molecular states underlying androgen receptor activation: a framework for therapeutics targeting androgen signaling in prostate cancer. J Clin Oncol 2012; 30: 644.
17. Nelson PS: Targeting the androgen receptor in prostate cancerda resilient foe. N Engl J Med 2014; 371: 1067.