The microenvironment associated with prostate cancer includes low cytolytic activity of natural killer (NK) cells,5 high secretions of TGF-beta by prostate tissue (which inhibits NK and lymphocyte function),6 and recruitment of T regulatory cells that down-regulate antitumor immunity.7 As such, the prostate cancer microenvironment has been described as an immunosuppressive state. Furthermore, based on the chronicity of the prostate cancer disease spectrum, the immune microenvironment is likely dynamic, with changes over time/clinical states and with treatment exposure.8 For example, there is increased tumor-infiltrating lymphocytes in the prostate bed following androgen deprivation therapy,9 and higher levels of PD-1 ligand and PD-L2 expression on the surface of enzalutamide treated prostate cancer cells.10 Several aspects make prostate cancer attractive for immunotherapy-based treatment options, including a high level of tumor-associated antigens such as prostate specific antigen (PSA), prostate acid phosphatase (PAP), and prostate-specific membrane antigen (PSMA).
Given the further emergence of immune checkpoint inhibitors since then, including several phase 3 trials accruing patients and presenting initial results, this article will focus specifically on the current landscape of immune checkpoint inhibitors and prostate cancer.
Arguably, the most focus among checkpoint inhibitors in the mCRPC arena is on pembrolizumab, initially receiving FDA approval in a tumor agnostic indication for MSI-high (MSI-H) mutation CRPC patients in 2017. Pembrolizumab is a human IgG4 monoclonal antibody targeting PD-1 activity. A study from the Memorial Sloan Kettering Cancer Center assessed the prevalence of MSI-H/dMMR prostate cancer among 1,033 patients treated at their institution,11 finding that 32 (3.1%) had MSI-H/dMMR disease. This included 23 patients (2.2%) that had tumors with high MSIsensor scores and 7 of the 32 MSI-H/dMMR patients (21.9%) with pathogenic germline mutation in a Lynch syndrome-associated gene. Eleven patients with MSI-H/dMMR CRPC received anti-PD-1/PD-L1 therapy and six of these had a greater than 50% decline in PSA levels. Based on these data, many experts in the field of advanced prostate cancer feel that mCRPC patients should be tested for MSI-H status based on the potential for pembrolizumab eligibility.
The KEYNOTE-028 study was a trial of pembrolizumab in advanced solid tumors among patients with PD-1 expression ≥1% of tumor or stromal cells. Among 245 men screened, there were 35 PD-1% (14.3%) and 23 patients who enrolled.12 There were four partial responses, for an objective response rate of 17.4%, and 8 of 23 (34.8%) patients had stable disease. Median duration of response was 13.5 months, and median PFS and OS were 3.5 and 7.9 months, respectively. Furthermore, the 6-month PFS and OS rates were 34.8% and 73.4%, respectively. Off-label use of pembrolizumab among a heavily pre-treated population of mCRPC patients has been reported. At the 2019 GU ASCO meeting, Tucker and colleagues presented data on 51 patients, 86% of which had received three or more prior lines of therapy. Most patients had previously received abiraterone (88%), docetaxel (86%), enzalutamide (80%), and sipuleucel-T (74%). Among these patients, 16% had a >50% confirmed PSA decline with pembrolizumab, with 8% having >90% PSA decline. Fifty-nine percent of men were treated with some form of concurrent therapy along with pembrolizumab, most commonly enzalutamide (47%).
In 2020, Antonarakis and colleagues published results from the phase II KEYNOTE-199 study in the Journal of Clinical Oncology.13 This study included three cohorts of patients with mCRPC treated with docetaxel and one or more targeted endocrine therapies. Cohort 1 enrolled patients with RECIST-measurable PD-L1 positive disease, cohort 2 patients with RECIST-measurable PD-L1-negative disease, and cohort 3 patients with bone-predominant disease, regardless of PD-L1 expression. All patients received pembrolizumab 200 mg every three weeks for up to 35 cycles, with a primary endpoint of objective response rate in cohorts 1 and 2. There were 133 patients enrolled in cohort 1, 66 in cohort 2, and 59 in cohort 3, with an objective response rate of 5% in cohort 1 and 3% in cohort 2. Disease control rate was 10% in cohort 1, 9% in cohort 2, and 22% in cohort 3, whereas median OS was 9.5 months in cohort 1, 7.9 months in cohort 2, and 14.1 months in cohort 3. Shortly after publication, Dr. Antonarakis presented updated data for these cohorts at GU ASCO 2020, demonstrating that over a median follow-up of 9.6 months the objective response rate of patients in cohort 1 was 6% and 3% for cohort 2.14 Thus, although the number of patients experiencing a response was low, those that had a response had a durable response with further follow-up.
The KEYNOTE-199 platform has additional cohorts, including cohort 4 (RECIST-measurable disease) and cohort 5 (bone-predominant disease) consisting of chemotherapy-naive patients with mCRPC treated with enzalutamide plus pembrolizumab after progression with enzalutamide. Graff and colleagues presented results of these two cohorts of patients at GU ASCO 2020.15 Among 126 patients (81 in cohort 4 and 45 in cohort 45), the median follow-up was 13.7 months over which time the objective response rate was 12% for patients in cohort 4. The disease control rate for patients in cohort 4 was 51% and was also 51% in cohort 5. PFS at 12 months was 17% and 23% for cohort 4 and 5, respectively, and OS at 12 months was 70% and 75%, respectively.
At the ESMO 2020 virtual annual meeting, Dr. Omlin presented updated results of cohorts 4 and 5 with a specific focus on time to cytotoxic chemotherapy and time to new anticancer therapy.16 The median time to PSA progression for cohort 4 was 4.2 months (95% CI 4.1-4.4) and cohort 5 was 4.2 months (95% CI 4.2-4.4). The median rPFS was 4.2 months (95% CI 2.5-6.0) for cohort 4 and 4.4 months (95% CI 3.4-6.2) for cohort 5. Whereas median overall survival was not reached (95% CI 15.9-NR) for cohort 4 and 18.8 months (95% CI 14.0-NR) for cohort 5. In cohort 4, the median time to next cytotoxic chemotherapy was 11.1 months (95% CI 8.5 – NR) and time to new anticancer therapy was 9.4 months (95% CI 7.2-11.1). In cohort 5, the median time to next cytotoxic chemotherapy was 11.3 months (95% CI 9.0 – 14.5) and time to new anticancer therapy was 9.5 months (95% CI 5.9-12.1). Whether the addition of pembrolizumab to enzalutamide has meaningful benefit in a phase III trial (ie. KEYLYNK-010, KEYNOTE-921, KEYNOTE-641) remains to be seen, though IMbassador250 failed to show a benefit for the combination of atezolizumab and enzalutamide.
KEYNOTE-365 is a phase Ib/II study evaluating pembrolizumab plus other agents for men with mCRPC, as highlighted in the study schema below.
At the GU ASCO 2020 meeting, results were presented for cohort A (pembrolizumab plus olaparib in docetaxel pretreated patients),17 cohort B (pembrolizumab plus docetaxel in abiraterone or enzalutamide pretreated patients),18 and cohort C (pembrolizumab plus enzalutamide in abiraterone pretreated patients).19 For Cohort A, 84 patients were treated of which 42 discontinued treatment, primarily due to disease progression (n=29). The overall PSA response rate was 9%, objective response rate was 8%, and disease control rate was 22%. For Cohort B, 104 patients were treated of which 72 discontinued treatment, again primarily due to disease progression (55%). The overall PSA response rate was 28%, objective response rate was 18%, and disease control rate was 51%. For Cohort C, 102 patients were treated of which 73 discontinued treatment, primarily due to disease progression (60%). The overall PSA response rate was 22%, objective response rate was 12%, and disease control rate was 35%.
At the virtual 2020 ESMO annual meeting, Dr. Mourey and colleagues presented updated results from cohort C including the time to symptomatic skeletal event, radiographic bone progression, and radiographic soft tissue progression.20 For all patients, median rPFS was 6.1 months (95% CI, 4.4-6.5) and median OS was 20.4 months (95% CI, 15.5-not reached). At 12 months, the rPFS rate was 24.6% and OS rate was 72.8% by Kaplan-Meier. The median time to radiographic bone progression was 8.3 months (95% CI 6.7-10.8), the median time to radiographic soft tissue progression was 15.2 months (95% CI 6.7 to not reached), and the median time to symptomatic skeletal-related event was not reached (95% CI 18.2 months to not reached).
At the GU ASCO 2021 meeting, updated data of Cohort B were presented, with an additional year of follow-up making the median follow-up 32.4 months.21 Overall, 74% of patients experienced some PSA response from baseline, and 34% of patients had a confirmed PSA response of greater than 50%. The overall response rate in patients with measurable disease (52 patients) was 23.1%, and 73.1% of patients experienced disease control on treatment. There were no complete responses, 12 partial responses, and 26 stable disease patients of any duration. Over 90% of patients had some tumor reduction from baseline. The rPFS of the cohort was 8.5 months, with an overall survival of 20.2 months.
Given these promising results in single-armed studies, a number of ongoing phase 3 trials are assessing the role of pembrolizumab in this disease space, as highlighted in the table below.
Like pembrolizumab, nivolumab is a human IgG4 monoclonal antibody targeting PD-1 activity. Initial results of the phase II KEYNOTE-650 were presented at the 2019 GU ASCO meeting, testing the combination of nivolumab plus ipilimumab for men with mCRPC. There were two cohorts for this study: cohort 1 – asymptomatic or minimally symptomatic, who had progressed after at least 1 second-generation hormone therapy with no prior chemotherapy, and cohort 2 – progression after chemotherapy. Overall response rates were 26% in cohort 1 and 10% in cohort 2, including two patients in each cohort who had a complete response. Median time to response was approximately two months. PSA response rate was 18% in cohort 1 and 10% in cohort 2.
There are several ongoing trials utilizing nivolumab in the setting of advanced prostate cancer, all of which were highlighted at the GU ASCO 2020 meeting. A single-arm, phase II trial of a DNA vaccine encoding prostate acid phosphatase (pTVG-HP) in addition to nivolumab is being tested in men with non-metastatic, PSA recurrent prostate cancer after undergoing radical prostatectomy.22 The primary outcomes will be percentage of subjects within acceptable toxicity boundaries and PSA complete response rate. A single-arm, phase I/II trial is assessing nivolumab plus radium-223 in mCRPC patients, specifically evaluating circulating tumor DNA levels as a biomarker of response.23 The primary outcomes of this study will be to assess the safety of this combination (phase Ib) and to assess circulating tumor DNA reduction after 6 weeks of nivolumab treatment (phase II). Finally, a phase Ib/II trial is assessing the combination of rucaparib with nivolumab in men with mCRPC.24 The primary outcome will be dose-limiting toxicity rate between the combination of rucaparib and nivolumab.
At the 2021 GU ASCO meeting, Dr. Fizazi and colleagues presented the final analysis results for Arm B (nivolumab plus docetaxel) of CheckMate 9KD.25 CheckMate 9KD Arm B enrolled patients with chemotherapy-naive mCRPC with ongoing androgen deprivation therapy and ≤ 2 prior novel antiandrogen therapies (ie. abiraterone, enzalutamide, etc). Patients received nivolumab 360 mg plus docetaxel 75 mg/m2 Q3W plus prednisone 5 mg BID for ≤ 10 cycles, followed by nivolumab 480 mg Q4W until disease progression/unacceptable toxicity (for up to 2 years). The co-primary endpoints were objective response rate and PSA response rate (defined as a ≥ 50% PSA reduction). Secondary endpoints included radiographic progression-free survival, overall survival, and safety. Of 84 treated patients with a median age of 71 years (range: 53-88), 27% had visceral disease and 54% had measurable disease; median PSA was 49.5 ng/mL (range: 1.2-1085). The median number of docetaxel cycles was 8 (range: 1-10), and the median number of nivolumab doses was 11 (range: 1-27). Over a median follow-up of 15.2 months, there were 76 patients (90.5%) that had discontinued therapy, most commonly secondary to disease progression (59.5%) and drug toxicity (17.9%). There was 1 (2.2%) complete objective response and 17 (37.8%) partial responses in 45 patients with measurable disease.
For the 18 patients with an objective response, the median time to response was 2.0 (range: 1.6-7.3) months, with a median duration of response of 7.0 months (95% CI 6.4-12.4). Among 81 PSA-evaluable patients, the median time to PSA progression was 8.7 months (95% CI 7.3-10.4). A reduction from baseline in the sum of diameters of target lesions was observed in most evaluable patients (79.5%), and the median change from baseline for all patients was -32.1%. Tumor reductions and PCWG3 responses were observed both in patients who had or had not previously received novel antiandrogen therapies. A reduction from baseline in PSA was observed in most evaluable patients (84.0%) and median change from baseline for all patients was -54.6%. Finally, the median radiographic progression-free survival for all patients was 9.0 months (95% CI 8.0-11.6) and the median overall survival was 18.2 months (95% CI 14.6-20.7).
Finally, the NEPTUNE trial is a phase 2 trial assessing nivolumab plus ipilimumab among men with mCRPC and a positive immunogenic signature. Initial results of this trial were presented at the AACR 2021 annual meeting. There were 184 men with mCRPC screened that had progressed following at least one line of therapy and 62 (34%) men had a positive immunogenic signature. A positive screen meant that the patient had at least one of either DNA mismatch repair deficiency, DNA damage repair deficiency, or a high level (>20%) of tumor-infiltrating lymphocytes. Among the 62 immunogenic signature positive men, 35 were treated with nivolumab plus ipilimumab every 3 weeks for four cycles, followed by nivolumab every 4 weeks for up to 1 year. Over a median follow-up of 7.2 months, 10 patients had either a radiologic response, a PSA response of at least 50%, conversion of circulating tumor cells at 9 weeks, or a combination of the three metrics, resulting in a composite response rate of 28.6%.
The following table summarizes key ongoing clinical trials incorporating nivolumab for the treatment of advanced prostate cancer. Notably, only CheckMate 7DX is a randomized comparison.
Atezolizumab is a monoclonal antibody of IgG1 against the protein programmed cell death-ligand 1 (PD-L1). Following initial approval in advanced bladder cancer, it has also been used in lung and breast cancers. In mCRPC, there are two notable trials utilizing atezolizumab, each in a combination therapy regime.
The phase 3 IMbassador250 trial of atezolizumab in combination with enzalutamide versus enzalutamide alone for patients with mCRPC did not have favorable results.26 First presented at the American Association for Cancer Research (AACR) virtual meeting 2020, IMbassador250 was a phase III, randomized clinical trial (NCT03016312) evaluating atezolizumab plus enzalutamide versus enzalutamide alone in men with mCRPC who had progressed on abiraterone and docetaxel or were not candidates for a taxane regimen. There were 759 patients randomized to atezolizumab plus enzalutamide (n = 379) or enzalutamide alone (n = 380). There was no difference in the primary efficacy endpoint of OS between groups: stratified HR 1.12, 95% CI 0.91-1.37. Further, the median OS was 15.2 months (95% CI 14.0-17.0) in the atezolizumab plus enzalutamide group and 16.6 months (95% CI 14.7-18.4) in the enzalutamide group. Treatment-related adverse events were observed in 77.8% undergoing atezolizumab plus enzalutamide and 51.1% of patients receiving enzalutamide alone. These negative results led to early termination of the study.
In a discussion of the IMbassador250 trial, Dr. Sharma from the University of Texas MD Anderson Cancer Center provided several potential hypotheses for why the addition of atezolizumab to enzalutamide did not improve survival outcomes for patients with mCRPC:
- The prostate cancer tumor microenvironment has few T cells, thus activation of T cells by atezolizumab may not be expected to improve cancer control
- There may be multiple immunosuppressive pathways in the prostate cancer tumor microenvironment, thus targeting a single pathway may not yield a meaningful tumor biology response
- Since prostate cancer does not typically care a high mutation burden, T cells may not recognize and target prostate cancer sufficiently
- The impact of enzalutamide on the immune response in tumor microenvironment is unknown
Atezolizumab has also been combined with cabozantinib in the COSMIC-021 (NCT03170960) trial, a multinational phase 1b study, evaluating the combination of cabozantinib with atezolizumab in various solid tumors. At the GU ASCO 2020 meeting, interim results were presented for patients treated with this combination who had mCRPC.27 There were 44 eligible patients who were required to have had radiographic progression in soft tissue after enzalutamide and/or abiraterone, had measurable disease, and an ECOG performance status of 0 or 1. Among these patients, 34% had visceral metastases, 61% had extrapelvic lymph node metastases, 27% had prior docetaxel and 52% had ≥2 prior novel hormonal therapies. Over a median follow-up of 10.6 months (range 3.4 to 17.9), the median duration of treatment was 5.3 months, and the objective response rate per RECIST 1.1 criteria was 32%. Among these, two patients experience a complete response, 12 patients had partial response, and 21 patients had stable disease for an overall disease control rate of 80%. This data was well-received and plans for further evaluating mCRPC patients with the combination of cabozantinib and atezolizumab are ongoing.
The following table summarizes key phase 3 ongoing clinical trials incorporating atezolizumab for the treatment of advanced prostate cancer.
Ipilimumab is a monoclonal antibody targeting CTLA-4, which has been used both individually and in combination with nivolumab for the treatment of melanoma and renal cell carcinoma. Among men with mCRPC, ipilimumab was tested in phase III for those who had progressed on docetaxel chemotherapy, randomizing 799 patients to ipilimumab or placebo after bone-directed radiotherapy.28 The primary endpoint was OS, with no difference between the groups (ipilimumab 11.2 months vs placebo 10 months; HR 0.85, p=0.053); however, there was a small benefit in PFS favoring ipilimumab (4.0 vs 3.1 months; HR 0.70, p < 0.0001). More recently, Beer et al.29 reported findings of another phase III trial randomizing 602 patients (2:1) with metastatic chemotherapy-naïve CRPC to ipilimumab vs placebo. Similar to the post-docetaxel patients, there was no difference in OS between the groups (HR 1.11, 95% CI 0.88-1.39), however men receiving ipilimumab had improved PFS (5.6 months vs 3.8; HR 0.67, 95% CI 0.55-0.81) compared to those receiving placebo.
Although there have been cautiously encouraging data for phase Ib/II trials assessing immune checkpoint inhibitors and mCRPC patients (primarily for pembrolizumab), the negative results of the IMbassador250 trial have somewhat tempered enthusiasm for immune checkpoint inhibitor therapy in advanced prostate cancer. However, much is yet to come from pembrolizumab combinations, with several phase III trials accruing and results anticipated in the next several years.
Published Date: June 2021