A Narrative Review of Implementing Precision Oncology in Metastatic Castration-Resistant Prostate Cancer in Emerging Countries

Mitoxantrone, the first cytotoxic chemotherapy, was approved for mCRPC based on improved palliative responses in pain-related measures [4] but showed no survival benefit [5]. In 2004, docetaxel was the first systemic chemotherapy to demonstrate survival benefit in mCRPC. Docetaxel was studied in two prospective phase III trials: the TAX 327 trial [6] and the Southwest Oncology Group (SWOG) trial [7]. In both studies, docetaxel prolonged median overall survival (OS) by 1.9–2.4 months, thereby establishing docetaxel as the standard of care for mCRPC in 2004 [6, 7]. Cabazitaxel, another taxane, has demonstrated activity in docetaxel-resistant prostate cancers. It was approved by the FDA in 2010 based on the results of the TROPIC trial, in which patients receiving the combination cabazitaxel/prednisone had significantly longer progression-free survival (PFS) and OS than those receiving the combination mitoxantrone/prednisone (PFS: 2.8 vs. 1.4 months; OS: 15.1 vs. 12.7 months, respectively) [8].

Androgen receptor-regulating genes presumably participate in various cellular processes that contribute to the initiation and progression of prostate cancer [9]. Until 2004, patients with mCRPC who progressed on novel ART therapies were treated with additional secondary hormonal agents, including antiandrogens such as bicalutamide and nilutamide [10]. Between 2011 and 2012, abiraterone, a new androgen biosynthesis inhibitor, and enzalutamide, an AR blocker with a higher affinity to AR, were approved as the first- and second-line therapies for mCRPC, respectively.

The approval of abiraterone was based on a multi-national phase III trial, COU-AA-301, in which the combination abiraterone/prednisone (1000 mg) improved the OS compared to placebo (14.8 vs. 10.9 months, respectively) in patients with mCRPC who had progressed with prior ART therapy and docetaxel [11]. Furthermore, the phase III COU-AA-302 trial conducted in mCRPC patients who showed progression when on ART therapy, but had no prior treatment with docetaxel, demonstrated superior OS (34.7 vs. 30.3 months) and median radiographic PFS (16.5 vs. 8.3 months) in the abiraterone group (1000 mg) as compared to the placebo group [12, 13].

The clinical efficacy of enzalutamide has been established in two phase III trials: PREVAIL [14] and AFFIRM [15]. It is also evident that enzalutamide is a treatment option for mCRPC patients in both the pre- and post-docetaxel settings and represents a reasonable choice for men who are not candidates for chemotherapy [16]. In 2017, another AR blocker, apalutamide, was successfully evaluated in a phase II study in patients with progressive mCRPC with and without prior chemotherapy with the combination abiraterone + prednisone (AAP); the PSA response rate after 12 weeks for AAP-naïve patients and patients who were treated with AAP previously was 88 and 22%, respectively, [17].

Sipuleucel-T is the first and only immunotherapy to be approved by the FDA in 2010 for the treatment of mCRPC. An independent phase III study (D9902B; the IMPACT [Immunotherapy for Prostate Adenocarcinoma Treatment) trial]) revealed that the use of sipuleucel-T prolonged the median survival of men with mCRPC by 4.1 months compared to the placebo (25.8 vs. 21.7 months, respectively) [18].

Bone metastases occur in most of the patients with mCRPC, primarily affecting the structural integrity of bone and causing patient disability, pain, reduced quality of life (QOL), and death. Radium-223 was approved by the FDA in 2013, based on the data of the phase III ALSYMPCA (Alpharadin in Symptomatic Prostate Cancer Patients) trial. The phase II ALSYMPCA trial showed an OS gain of 2.8 months over placebo (14.0 vs. 11.2 months, respectively; p = 0.002) in men with mCRPC (and symptomatic bone metastasis) [19]. The trial also demonstrated reduced pain and improved symptomatic skeletal events in patients with mCRPC without visceral disease [20]. Additionally, a retrospective cohort study showed that the concomitant use of other bone-modifying agents, such as denosumab or zoledronic acid, with other relatively new agents is a common clinical practice for the treatment of CRPC patients with bone metastases [21].

In a long-term placebo-controlled randomized clinical trial (RCT), treatment with 4 mg zoledronic acid was associated with skeletal-related events (SREs) versus placebo in men with hormone refractory prostate cancer at 24 months (38 vs. 49%; difference − 11.0%, 95% confidence interval [CI] − 20.2 to − 1.3%; p = 0.028). The median time to first SRE was 488 days compared to 321 days with placebo (p = 0.009). The ongoing risk of SREs was reduced by 36% with zoledronic acid compared to placebo (risk ratio 0.64, 95% CI 0.485–0.845; p = 0.002) [22]. In a phase 3 RCT, denosumab (120 mg) was found to be better than zoledronic acid (4 mg) at preventing SREs (median time to first SRE 20.7 vs. 17.1 months, respectively; hazards ratio [HR] 0.82, 95% CI 0.71–0·95; p = 0·0008). However, hypocalcemia occurred more frequently in the denosumab group than in the zoledronic acid group [23].

To date, no clear recommendations are available for guiding the appropriate treatment sequence in mCRPC. Currently, the choice for further treatment following the development of castration resistance is unclear. Furthermore, cross-resistance is also commonly observed between abiraterone and enzalutamide when these drugs are used sequentially for the treatment of mCRPC. The Kyoto–Baltimore Collaboration report suggested that abiraterone as the first-line treatment before enzalutamide prolonged the PFS (HR 0.56; p < 0.001), but not OS. This effect is relative to enzalutamide as first-line treatment before abiraterone in patients with chemotherapy-naïve castration-resistant prostate cancer [24]. However, OZM-054, a phase II, randomized trial compared the clinical benefit (defined as PSA decline ≥ 50% or stable disease for ≥ 12 weeks) of cabazitaxel and the combination abiraterone/enzalutamide in patients with mCRPC expressing poor outcomes [25]. In the first-line treatment, a significantly higher number of patients benefited from receiving cabazitaxel over abiraterone/enzalutamide (90 vs. 70%; p = 0.02). In terms of second-line therapy upon cross-over, there was no difference between treatment groups in PSA50 (50% decline in PSA), measurable disease response, or stable disease at > 12 weeks (75 vs. 85%; p = 0.483). In addition, the study did not demonstrate an OS benefit (p = 0.143) with upfront cabazitaxel over abiraterone/enzalutamide in these patients [25]. The PROREPAIR-B cohort study demonstrated significantly longer PFS among men with mCRPC who received abiraterone or enzalutamide upfront compared with those who received first-line docetaxel (10.8 vs. 8.3 months; p < 0.001). However, no significant differences in the OS were observed with both treatment sequences [26]. Another retrospective study of treatment sequences in real-world practice among men with mCRPC showed no significant difference in OS with abiraterone first followed by enzalutamide, or the reverse sequence; abiraterone or enzalutamide first followed by docetaxel, or the reverse sequence; or docetaxel first followed by cabazitaxel [27]. In the CARD study, which included mCRPC patients progressing after docetaxel and abiraterone/enzalutamide, cabazitaxel significantly improved median imaging-based PFS compared to enzalutamide/abiraterone (8.0 vs. 3.7 months) and OS (13.6 vs. 11.0 months; HR for death 0.64, 95% CI 0.46–0.89; p = 0.008). The median PFS was 4.4 months with cabazitaxel versus 2.7 months with abiraterone/enzalutamide (HR for progression or death 0.52, 95% CI 0.40–0.68; p < 0.001) [28].

A family history of prostate cancer has been long recognized as a major risk factor (60%) [30]. Patients with mCRPC can have genomic aberrations that interfere with the DDR pathway [31]. These include somatic (23%) [32] and/or germline (11.8%) alterations in DDR genes, such as BRCA1 (germline 0.9%, somatic 0.9%), BRCA2 (germline 8.6%, somatic 7.7%), ataxia-telangiectasia mutated (ATM) (germline 2.3%, somatic 4.5%), and CHEK2 (germline 4.1%, somatic 0.9%), and also in other genes with direct and indirect roles in homologous recombination repair (HRR), such as BRIP1, BARD1, CDK12, CHEK1, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D, or RAD54L [33, 34].

Germline BRCA1 and BRCA2 mutations in men are associated with a significant increase in aggressive prostate cancer risk. By the age of 65 years, BRCA1 and BRCA2 carriers are at 3.75- and 8.6-fold increased risk of prostate cancer compared to non-carriers [35, 36]. In addition, genetic mutations in HRR genes, BRCA1, BRCA2, and ATM have been found to be associated with an increased sensitivity to PARP inhibition [37]. Mateo et al. [38] conducted a phase II trial in which they studied the efficacy of a PARP inhibitor, olaparib, in mCRPC patients who progressed on standard therapy. Next-generation sequencing (NGS) identified deleterious mutations (somatic and germline) in DDR genes, including BRCA1/2, ATM, Fanconi’s anemia genes (FANCA), and CHEK2 in 16 of 49 evaluated patients (33%). Among these 16 patients, 14 (88%) had a response to olaparib, including seven patients with loss of BRCA2 (4 with biallelic somatic loss and 3 with germline mutations) and four of five with an ATM mutation. This discovery that approximately half of these treatment-actionable genetic alterations lie in the germline DNA (and are therefore heritable) have had profound implications in the field of precision medical oncology in prostate cancer.

Results from the TOPARP study led to the design and conduct of the PROfound study, a prospective, randomized, open-label, phase III clinical trial with the aim to evaluate the efficacy and safety of olaparib versus enzalutamide or abiraterone in men with mCRPC who had failed prior treatment with a novel ART therapy (abiraterone or enzalutamide) and had a qualifying tumor mutation in ≥ 1 of the 15 predefined genes (prospectively identified by the FoundationOne^® CDx investigational NGS test) [36]. Those with alterations in BRCA1, BRCA2, and ATM were included in cohort A (n = 245) based upon prevalence in the prostate population, while those with alterations in BRIP1, BARD1, CDK12, CHEK1, CHEK2, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D, or RAD54L were included in cohort B (n = 142). Patients in each cohort were randomized 2:1 to receive olaparib 300 mg twice daily (BID) or the physician’s choice of enzalutamide (160 mg/day) or abiraterone (1000 mg/day + prednisone 5 mg BID) [39].

Olaparib resulted in a significant improvement in median radiographic PFS (rPFS; assessed by blinded independent central review [BICR]) compared to physicians’ choice of treatment (enzalutamide/abiraterone) among patients in cohort A (7.4 vs. 3.55 months, respectively; HR 0.34; p < 0.0001) [39]. Olaparib also resulted in a significantly improved OS (median 18.5 vs. 15.11 months; HR 0.67, p = 0.0063) and a significantly greater objective response rate (33.3 vs. 2.3% as assessed by BICR; odds ratio 20.86; p < 0.0001) and median time to pain progression (not reached vs. 9.92 months based on the Brief Pain Inventory [Short Form] worst pain [item 3] and opioid use; HR 0.44; p = 0.0192). When cohort A and B were combined, improvement in rPFS shown in patients receiving olaparib was maintained (median 5.82 vs. 3.52 months; HR 0.49; p < 0.0001). Important toxicities included nausea, anemia elevation in liver enzymes, and gastrointestinal and hematological side effects. Additional toxicities observed in this study included a venous thromboembolism prevalence and potential induction of myelodysplastic syndrome/acute myeloid leukemia [39].

PROfound validated the concept of PARP sensitivity across multiple genes associated with HRR in mCRPC. It also highlighted the importance of genomic testing in prostate cancer and showed that identifying patients with DDR alterations in individual patients expands the treatment artillery available to include PARP inhibitors. However, the availability of more treatment options necessitate a better understanding of the optimal treatment sequence for individual patients.

Several other PARP inhibitors, such as rucaparib, niraparib, veliparib, and talazoparib, have been evaluated in phase II trials as monotherapy (Table 2). Similar to the study using olaparib, patients are being selected based upon the presence of DDR alterations, either from archival tumor specimens, cell-free DNA, or both. And again, like olaparib, each of these PARP inhibitors demonstrated impressive response rates in genomically selected patients. The most significant reported adverse events were hematologic (grade 3/4 anemia and thrombocytopenia), along with nonhematologic events, including gastrointestinal events, asthenia, and hypertension. [40] The details of ongoing phase II and III trials evaluating PARP inhibitors as monotherapy or in combination with other therapies as first-line or later-lines treatments in mCRPC are shown in Table 3.

In order to identify the DDR alterations that make patients eligible for treatment with PARP inhibitors, archival tumor tissues or cell-free tumor DNA are retrieved for NGS. However, a proportion of the DDR alterations detected are likely to be germline. While not addressed by any of the PARP inhibitor studies, the role of offering germline testing in such cases is important, given the implications for family members on cancer predisposition and the possibility to deploy risk-reduction strategies at an early stage [32, 33].

It is important for pathologists and clinicians to understand the role and effect of potential targeted therapies in the light of germline (inherited) and somatic (acquired) mutations that occur in prostate cancer.

While germline genetic testing helps identify inherited pathogenic mutations in genes associated with familial cancer risk, tumor-directed somatic testing may guide treatment decision-making. Germline genetic testing can be performed on lymphocyte DNA from blood or a combination of lymphocyte and buccal cells from saliva, obtained non-invasively; on the other hand, somatic testing is complex and requires prostate tumor material from biopsies, or in some cases, circulating tumor cells/DNA (ctDNA) in the blood. Additionally, outside the clinical trials in the real-world settings, obtaining sufficient and high-quality tumor tissue for complicated somatic analysis is not a trivial process in patients with mCRPC [41]. Using archived tissue samples may enable wider testing, but there are risks of missing the evolution of somatic mutations in the tumor tissue due to genetic instability [42]. As a result, repeat testing of tumor DNA is required during the disease course.

Although somatic testing may help identify potential germline mutations [43], it should never be used to substitute for germline testing, primarily because of the risk for false-positives and false-negatives owing to the variations in reporting between different commercially available tests. However, contemporary sampling of metastatic disease sites or cell-free ctDNA using liquid biopsy may be more informative and a novel way to identify genomic alterations and track patient’s genomic landscape over time [44].

Despite extensive investigations, the identification of rare germline mutations in prostate cancer genes has been extremely challenging. Four major factors contributing to the challenges include: (1) the genetic heterogeneity of prostate cancer, (2) the high rate of sporadic disease, (3) large well-annotated patient populations needed to establish associations between germline pathogenic mutations and prostate cancer, and (4) the high cost of sequencing.