The arsenal of therapeutic options for men with castration-resistant prostate cancer has steadily grown as ongoing research uncovers more about prostate cancer biology and resistance mechanisms.
William Oh, MD
As ongoing research continues to deepen our understanding of prostate cancer biology and resistance mechanisms, our arsenal of therapeutic options for men with castration-resistant prostate cancer (CRPC) has steadily grown. Four general classes of therapeutic agents have emerged so far: cytotoxic chemotherapy (docetaxel [Taxotere] and cabazitaxel [Jevtana]), androgen receptor (AR)-pathway directed therapies (abiraterone [Zytiga], enzalutamide [Xtandi], and apalutamide [Erleada]), bone-targeted therapy (radium-223 [Xofigo]), and active cellular immunotherapy (sipuleucel-T [Provenge]). Each of the drugs in these classes has demonstrated significant clinical benefits in randomized phase III clinical trials in CRPC,1-9 but the duration of response for each remains relatively shortlived and the survival benefits are moderate. There remains a great need for additional treatment options. Here we reexamine the case for platinum therapy.Platinum agents have long been studied in metastatic CRPC (mCRPC), but in patient cohorts that are largely molecularly unselected (Table). Trial experience with single-agent carboplatin,10-13 cisplatin,14,15 and oxaliplatin16 have generally yielded modest rates of prostatespecific antigen (PSA) decline, improvement in pain, and radiographic response.17 Although platinum doublet combinations with a taxane,18-20 anthracycline,21,22 etoposide,23,24 or gemcitabine25 are usually associated with slightly higher rates of PSA decline, they are also associated with increased toxicity. Triplet combination regimens, such as paclitaxel/ estramustine/carboplatin26 or docetaxel/estramustine/ carboplatin27-29 have been found to have significant clinical activity. In one study, docetaxel/estramustine/carboplatin was associated with a ≥50% decline in PSA in 95% of enrolled patients.29 However, these studies were relatively small and these regimens have not been further developed. Of the 3 platinum drugs, cisplatin and oxaliplatin use is often limited by renal impairment and neuropathic toxicity. Carboplatin is generally well tolerated, but has a heavier footprint in terms of myelosuppressive effects.
Satraplatin, a fourth-generation oral platinum analogue, was tested in the randomized phase III SPARC study. A total of 950 men with chemotherapy- refractory mCRPC were enrolled and received either satraplatin plus prednisone or placebo plus prednisone. Although satraplatin was associated with a small, but significant, benefit in progression-free survival (11.1 vs 9.7 weeks; HR, 0.67; 95% CI, 0.57-0.77; P <.001), improvement in time to pain progression (66.1 vs 22.3 weeks; HR, 0.64; 95% CI, 0.51-0.79; P <.001), and PSA response (25.4% vs 12.4%; P <.001), there was no demonstrated difference in median overall survival (OS) compared with placebo (61.3 vs 61.4 weeks; HR, 0.98; 95% CI, 0.84-1.15; P = .80).30
Although SPARC was ultimately a negative trial and further development of satraplatin stopped, the results again confirm that a subset of patients with mCRPC is responsive to platinum therapy, even as a single agent. Unfortunately, predictive markers to identify the cohort most likely to benefit from platinum therapy could not be developed from SPARC, as correlative and translational studies were not included in the study design.
Currently, outside of small cell carcinoma of the prostate, where platinum therapy is considered standard offering based on limited study findings31-33 and an extrapolation of efficacy data in small cell lung cancer, platinum compounds have not been formally adopted into any professional practice guidelines. This is due to a lack of defined OS benefit in the general nonselected population with prostate cancer. It could be argued that adoption of platinum drugs has also been poor due to inadequate characterization of the subtype of patients with CRPC who could benefit from platinum agents. Neuroendocrine prostate cancer (NEPC), a heterogeneous histology34 that most commonly transdifferentiates from preexisting adenocarcinoma as an adaptive response to long-term exposure to androgendeprivation therapy (ADT) and AR blockade,35,36 is classically considered to be responsive to platinumbased therapies. The response rates are as high as 70% in those with poorly differentiated histology.37
Histologically, NEPC is archetypally characterized as the presence of small round blue neuroendocrine cells with a lack of AR expression. Phenotypically, NEPC frequently metastasizes to visceral organs, has disease burden out of proportion to serum PSA levels, and may express elevated serum neuroendocrine markers such as chromogranin A, synaptophysin, and neuron specific enolase. Clinically speaking, these findings are neither specific nor sensitive, so more definitive molecular characterization is needed to recognize this subgroup of patients who may be good candidates for platinum compounds.Genomic instability has long been known to be a hallmark of cancer, leading to genomic heterogeneity that gives variable clinical courses and response to treatment. With recent advances in genomic profiling, changes to the cancer genome—both germline and somatic—have been better characterized.
Germline mutations that disrupt DNA damage repair, especially the process of homologous recombination, such as BRCA1, BRCA2, ATM, CHEK2, PALB2, and RAD51D, have prognostic consequence. Various study results show them to be associated with increased risk of prostate cancer diagnosis at a younger age, higher histologic grade on diagnosis, more aggressive clinical course, and increased risk of cancer-specific mortality.38-41 It has also been suggested that homologous recombination is associated with relative insensitivity and worse outcomes with ADT.
The presence of germline mutations in DNA damage repair genes had previously been thought to be relatively low, with The Cancer Genome Atlas reporting a prevalence of 4.6% in men with localized prostate cancer.42 However, recent study results demonstrate a significantly higher proportion of men with mCRPC harboring these germline mutations, 11.8%, irrespective of family history of cancer.43 When considering both germline and somatic genomic alterations, the prevalence may be as high as 20% in mCRPC.44-46 Outside of specific mutations in the defined DNA damage repair pathway genes, a “BRCAness” genomic signature may exist for an even wider population of patients with mCRPC.
These genomic findings have therapeutic implications in that there has been evidence to suggest increased sensitivity to platinum therapy in ovarian47,48 and breast cancers49,50 with germline BRCA1/2 mutations. Platinum compounds exert their antineoplastic activity by forming covalent DNA adducts, causing double-stranded DNA breaks. Loss-of-function mutations or alterations to BRCA1, BRCA2, ATM, CHEK2, PALB2, and RAD51D, among other genes, all of which serve important roles in the repair process for double-stranded DNA breaks— including homologous repair, nonhomologous end joining, or altered DNA damage checkpoints—put the disease at a handicap in overcoming the effects of platinum agents.45,51
Early evidence suggests that prostate cancers that harbor deficiencies in the DNA damage repair pathway are also associated with a higher likelihood of response to platinum agents and poly (ADP-ribose) polymerase (PARP) inhibitors.52-54 In a retrospective analysis at Dana-Farber Cancer Institute of 141 men with CRPC who received at least 2 cycles of carboplatin/docetaxel chemotherapy, 75% (6/8) of men carrying deleterious BRCA2 variants experienced a >50% decline in PSA within 12 weeks of treatment initiation compared with 17% (23/133) in those who lacked the BRCA2 variant (P <.001).52 Twenty-five percent of BRCA2 carriers had a >90% PSA decline versus 4.5% of noncarriers.52 Median OS from time of platinum therapy initiation was 18.9 months for carriers versus 9.5 months for noncarriers (P = .03). Prospective studies are needed to fully realize the benefits of platinum compounds in molecularly selected patients (NCT02311764, NCT02598895).
Given the higher than expected prevalence of germline mutations and their prognostic and therapeutic implications, the question arises whether it would be appropriate to actively screen all men with mCRPC for these mutations. If so, what type of testing should be done? Would germline testing be sufficient by itself? Should somatic mutations be screened for as well? The prospect of obtaining metastatic biopsies on all patients with mCRPC, for the purpose of genomic testing, may prove to be a difficult task, especially as some lesions, such as osseous metastases, may not lend themselves to easy collection. Next-generation sequencing of circulating tumor cells or cell-free DNA analysis could play a role in identifying patients at higher likelihood of benefit with a platinum-based regimen.In the absence of having genomic data to optimally guide treatment selection, is there still a place for the routine use of platinum therapy in the standard management of mCRPC? Emerging data suggest that prolonged therapeutic exposure to potent AR-pathway— targeting agents places selective pressures on the tumor, inducing adaptive changes and leading to the emergence of “treatment-induced lineage crisis,” characterized by poorly differentiated and aggressive prostate cancers, visceral and bulky metastasis, and low PSA production.55 Lineage crisis often occurs in the form of treatment-induced neuroendocrine differentiation or epithelial-to-mesenchymal transition. There may be a role for platinum-based agents in maintaining lineage plasticity and collateral sensitivity and preventing the onset of treatment-induced lineage crisis and therapeutic resistance.
Currently enrolling at the Mount Sinai Health System is the Prostate Cancer Intensive Non—Cross Reactive Therapy (PRINT) for CRPC trial (NCT02903160), a single-arm investigator-initiated phase II study aimed at evaluating the clinical benefits of treating an intrinsically heterogeneous tumor with a rapidly cycling, non–cross reactive regimen of FDA-approved prostate cancer therapeutic agents. While optimal timing and sequence of different drugs in mCRPC remains a topic of debate, the standard approach so far has been to treat with a single agent until resistance develops, at which point the patient is transitioned to a different agent. Inherent genetic instability will lead to substantial tumor heterogeneity56 along with varied amounts of genetic dysregulation and varied patterns of antineoplastic resistance.57 This argues that a one-size-fits-all, sequential, single-agent approach will ultimately be met with limited success and subpopulations resilient to single-agent therapies will be identified, leading to a multidrug-refractory phenotype.
Our rapidly-cycling treatment regimen contains 3 consecutive treatment modules, each lasting 12 weeks:
By switching between non—cross reactive therapies, the selective pressure exerted by any single therapeutic agent’s intermittent use is limited, allowing resistance rates to stabilize or decrease in its absence. Additionally, the therapeutic agents are delivered as rationally designed combinations to achieve optimal therapeutic dosing and minimize overlapping adverse events. The rationale for inclusion of carboplatin in this regimen is 2-fold: to improve disease response and to help maintain lineage plasticity.
The primary endpoint of this study is time to disease progression following completion of the study regimen. If this novel treatment strategy significantly lengthens the time until disease progression, this trial would serve as proof-of-principle for future treatment regimen design in mCRPC. Additionally, as this regimen was designed to only include fully approved therapies, it would be immediately accessible to any treating oncologist.Platinum compounds remain relevant in mCRPC. Going forward, next-generation sequencing will be needed to further develop and expand on the development of biomarkers to predict for platinum responsiveness and to better identify patients appropriate for prospective trials involving platinum-based regimens. A better understanding of whether biomarkers associated with platinum sensitivity and their association with response to PARP inhibitors, and vice versa, will be important. It’s also important to discover how these 2 classes of drugs with similar mechanisms of actions, but potentially different mechanisms of resistance, should interplay.