Genomic Profiling of Prostate Cancer: An Updated Review

Abstract

Understanding the genomic profiling of prostate cancer is crucial, owing to the emergence of precision medicine to guide therapeutic approaches. Over the last decade, integrative genomic profiling of prostate tumors has provided insights that improve the understanding and treatment of the disease. Minimally invasive liquid biopsy procedures have emerged to investigate cancer-related molecules with the advantage of detecting heterogeneity as well as acquired resistance in cancer. The metastatic castration-resistant prostate cancer (mCRPC) tumors have a highly complex genomic landscape compared to primary prostate tumors; a number of mCRPC harbor clinically actionable molecular alterations, including DNA damage repair (e.g., BRCA1/2 and ATM) and PTEN/phosphoinositide 3-kinase signaling. Heterogeneity in the genomic landscape of prostate cancer has become apparent and genomic alterations of TP53, RB1, AR, and cell cycle pathway are associated with poor clinical outcomes in patients. Prostate cancer with mutant SPOP shows a distinct pattern of genomic alterations, associating with better clinical outcomes. Several genomic profiling tests, which can be used in the clinic, are approved by the U.S. Food and Drug Administration, including MSK-IMPACT, FoundationOne CDx, and FoundationOne Liquid CDx. Here, we review emerging evidence for genomic profiling of prostate cancer, especially focusing on associations between genomic alteration and clinical outcome, liquid biopsy, and actionable molecular alterations.

INTRODUCTION

Prostate cancer is the second most frequent cancer among males and the cause of an estimated 385,000 deaths worldwide in 2018 [1]. Prostate carcinogenesis and progression are correlated with loss of specific chromosome regions and candidate tumor suppressor genes, such as loss of 8p21 and NKX3.1, loss of 10q and PTEN, loss of 13q and RB1, and loss of 17p and TP53 [2]. Recurrent gene fusions of TMPRSS2 and ETS transcription factor genes are frequently detected in prostate cancer, suggesting that the androgen-responsive promoter elements of TMPRSS2 mediate the overexpression of ETS family members [3]. Prostate cancer development and disease progression are driven by the androgen receptor (AR) signaling pathway, which has led to the use of androgen deprivation therapy (ADT) for patients with advanced prostate cancer. Sustained AR signaling is the primary driver of castration-resistant prostate cancer (CRPC), leading researchers to develop novel treatments targeting the AR pathway, such as abiraterone and enzalutamide [4]. Molecular mechanisms behind AR reactivation in CRPC include AR gene amplification, AR mutations (e.g., T878A, F876L, L702H, L701H, and T877A), AR splice variants (AR-Vs), changes of androgen biosynthesis, and changes in AR cofactor [5]. Recently, novel mechanisms of AR activation have been reported, such as amplification of an upstream enhancer of AR and AR gene rearrangements [6,7,8]. During disease progression, a subset of metastatic CRPC (mCRPC) tumors loses AR dependence and often have neuroendocrine features [9].

Recently, precision medicine has emerged to guide therapeutic approaches for patients with prostate cancer by understanding each altered gene or pathway in an individual, leading to the improvement of clinical outcomes [10]. A phase 3 clinical trial demonstrated that the alteration of BRCA1/2 or ATM was associated with response to poly (adenosine diphosphate–ribose) polymerase (PARP) inhibitor olaparib in patients with mCRPC [11]. An Akt inhibitor, ipatasertib, showed antitumor activity in patients with PTEN-loss tumors, in a phase 2 study [12]. Over the last decade, the integrative genomic profiling of human prostate tumors had provided the foundations for discoveries that can impact disease understanding and treatment [13,14,15]. Furthermore, minimally invasive liquid biopsy procedures have emerged to investigate cancer-related molecules with the advantage of detecting heterogeneity as well as acquired resistance in cancer [16,17]. Here, we review emerging evidence for genomic profiling of prostate cancer, especially focusing on association of genomic alteration and clinical outcome, liquid biopsy, and actionable molecular alterations (Fig. 1). In this review, we identified the relevant studies using electronic databases, including PubMed and Web of Science.

Fig. 1. Overview of genomic profiling of prostate cancer. The specific gene/pathway alterations are associated with clinical outcomes. Genomic profiling is useful to identify actionable molecular alterations. cfDNA: cell free DNA, ctDNA: circulating tumor DNA, CTC: circulating tumor cell, AR: androgen receptor, PARP: poly (adenosine diphosphate–ribose) polymerase, MSI-H: microsatellite instability-high, dMMR: deficiency in mismatch repair genes.

MAIN BODY

1. Genomic landscape of prostate cancer

Common genetic alterations in primary prostate cancer include losses of NKX3.1 and PTEN [2] and fusion of ETS family transcription factor genes with androgen-responsive promoters [3]. In addition, a significant proportion of primary prostate tumors harbor large-scale genomic rearrangements [18,19]. Recurrent somatic mutations were identified in multiple genes, including SPOP and FOXA1, in patients with primary prostate cancer [20]. In 2015, The Cancer Genome Atlas (TCGA) presented a comprehensive molecular analysis of 333 primary prostate cancers, in which the tumors fell into subtypes according to specific gene fusions or mutations (SPOP, FOXA1, and IDH1) [14]. AR activity varied widely in a subtype-specific manner, with SPOP and FOXA1 mutant tumors having the highest levels of AR-induced transcripts [14]. In 2015, Robinson et al [15] demonstrated that aberrations of AR, ETS genes, TP53, and PTEN were detected in 40% to 60% of cases in patients with mCRPC. The mCRPC tumors have a highly complex genomic landscape compared to primary prostate tumors (Fig. 2) [21,22]. Genomic alterations in AR, TP53, RB1, and PTEN are enriched during disease progression [23,24,25]. Approximately 90% of mCRPC harbor clinically actionable molecular alterations, including AR signaling, DNA damage repair and phosphoinositide 3-kinase (PI3K) signaling [15].

Fig. 2. Gene alterations in the different stages of prostate cancer. Localized PCa, TCGA (n=333) [14]; mCSPC, MSK (n=424) [38]; mCRPC, SU2C/PCF Dream Team (n=444) [36]. The frequency of each gene alteration was calculated based on clinical data provided by cBioPortal (https://www.cbioportal.org/) The Figures from the cBioportal are permitted to use in the publications (https://docs.cbioportal.org/1.-general/faq#can-i-use-figures-from-the-cbioportal-in-my-publications-or-presentations) [21,22]. PCa: prostate cancer, TCGA: The Cancer Genome Atlas, mCSPC: metastatic castration-sensitive prostate cancer, MSK: memorial sloan kettering, mCRPC: metastatic castration-resistant prostate cancer, SU2C/PCF: stand up to cancer/prostate cancer foundation.

In 2018, two studies, Quigley et al [6] and Viswanathan et al [7], demonstrated the structural alterations driving mCRPC using whole-genome sequencing. Tandem duplications affect an upstream enhancer of AR in 70% to 87% of cases, correlating with increased AR expression [6,7]. Progression on androgen pathway inhibitors, abiraterone and enzalutamide, was associated with gains in AR and AR enhancer [7]. Tandem duplication hotspots also occur near MYC, associated with post-translational MYC regulation [6]. Classes of structural variations were linked to distinct DNA repair deficiencies, including associations of CDK12 mutation with tandem duplications, TP53 inactivation with inverted rearrangements and chromothripsis, and BRCA2 inactivation with deletions [6,7,26].

The ethnic and racial background can influence the incidence and mortality of prostate cancer, partly due to the interplay of socioeconomic factors and environmental exposures [27]. To date, most prostate cancer genomics data have been derived from Western populations. Thus, precision oncologic studies have under-represented patients from Asia and Africa, limiting comprehensive understanding of disparities in the diagnosis and prognosis of prostate cancer among these populations [28]. The incidence and mortality rates of prostate cancer for Asians are lower than Western populations [29]. In 2020, Li et al [30] reported on the genomic landscape of primary prostate cancer in Asian populations, in which 41% of tumors contained mutations in FOXA1 and 18% had deletions in CHD1. Lower incidence of FOXA1/CHD1 alterations in Western populations and lower incidence of TMPRSS2:ERG fusion gene and PTEN loss in Asian populations compared with counterparts were reported [30,31,32,33]. Thus, the genomic alteration signatures in Asian patients were markedly different from those of Western cohorts.

2. Association of genomic alteration and clinical outcome

Heterogeneity in the genomic landscape of prostate cancer has become apparent through several comprehensive profiling studies. Growing evidence suggests that the genomic alterations correlate with clinical outcomes (Table 1). In 2014, Hieronymus et al [34] reported an association between biochemical recurrence and the pattern of DNA copy number alteration (CNA) in primary prostate cancer, raising the possibility of CNA as a prognostic biomarker. Since 2018, several studies have demonstrated the association of specific gene/pathway alterations and clinical outcomes based on the genome-wide study of prostate cancer [25,35,36,37,38,39]. Wang et al [35] reported that the gene-based pathway of cell cycle progression was associated with shorter time to treatment change (TTTC) in patients with mCRPC who were treated with abiraterone (hazard ratio [HR], 2.11; 95% confidence interval [CI], 1.17–3.80; p=0.01). Abida et al [36] demonstrated that RB1 alteration was associated with poor overall survival (OS), whereas alterations in RB1, AR, and TP53 were associated with shorter TTTC in patients with mCRPC treated with abiraterone or enzalutamide. Chen et al [37] reported that two DNA alterations in RB1 were predictive of poor OS (median 14.1 mo vs. 42.0 mo; p=0.007), and CTNNB1 mutations were exclusive to enzalutamide-resistant patients (p=0.01), associating with poor OS (median 13.6 mo vs. 41.7 mo; p=0.025) in patients with mCRPC treated with enzalutamide. Stopsack et al [38] reported that rates of castration resistance (HR, 1.84; 95% CI, 1.40–2.41) and death (HR, 3.71; 95% CI, 2.28–6.02) were higher in high-volume metastatic castration-sensitive prostate cancer (mCSPC), associating with genomic alterations. Rates of castration resistance differed 1.5-fold to 5-fold according to alterations in AR, cell cycle pathway, MYC pathway, TP53, WNT pathway (inverse), and SPOP (inverse), whereas OS rates differed 2-fold to 4-fold according to AR, cell cycle pathway, WNT pathway (inverse), and SPOP (inverse) [38]. Mateo et al [25] reported that patients with RB1 loss in the primary prostate cancer had a worse prognosis. Among men with matched hormone-naive and mCRPC biopsies, RB1/TP53/AR aberrations were enriched in later stages [25]. Deek et al [39] reported that the frequency of driver mutations in TP53 (p=0.01), WNT (p=0.08), and cell cycle (p=0.04) genes increased across the mCSPC spectrum. Mutations in TP53 were independently associated with shorter radiographic progression free survival (PFS) (HR, 1.59; p=0.03) and the development of CRPC (HR, 1.71; p=0.01) [39]. Hamid et al [40] reported that deleterious tumor suppressor genes, TP53, PTEN, and RB1, were associated with an increased risk of relapse and death in patients with CSPC.

Table 1. Genomic alterations in prostate cancer tissue samples associated with clinical outcome.

Author	Year	Patients	Number of patients	Therapy	Endpoint	Genomic alterations	Outcome
Hieronymus et al [34]	2014	Localized PCa	168	Px	Risk of BCR	CNA burden	HR, 1.99; 95% CI, 1.11–3.55; p=0.021
Wang et al [35]	2018	mCRPC	77	ABI	TTTC	Cell cycle progression scores (≥50)	HR, 2.11; 95% CI, 1.17–3.80; p=0.01
Boysen et al [41]	2018	mCRPC	89	ABI	TTTC	SPOP	HR, 0.37; 95% CI, 0.20–0.69; p=0.002
Abida et al [36]	2019	mCRPC	128	ABI or ENZ	TTTC	RB1	CPE=0.818; p<0.001
						AR	CPE=0.651; p=0.005
						TP53	CPE=0.609; p=0.046
					OS	RB1	CPE=0.768; p=0.002
Chen et al [37]	2019	mCRPC	101	ENZ	OS	RB1	Median 14.1 mo vs. 42.0 mo; p=0.007
						CTNNB1	Median 13.6 mo vs. 41.7 mo; p=0.025
Hamid et al [40]	2019	Localized PCa	205	Local therapy	PFS	TP53, PTEN, and RB1	HR, 1.95; 95% CI, 1.22–3.13; p=0.005
					Time to CRPC	TP53, PTEN, and RB1	HR, 3.36; 95% CI, 1.01–11.16; p=0.04
Stopsack et al [38]	2020	mCSPC	424	N/A	Time to CRPC	AR	HR, 5.30; 95% CI, 2.97–9.46
						Cell cycle pathway	HR, 2.12; 95% CI, 1.50–3.00
						MYC pathway	HR, 2.04; 95% CI, 1.35–3.10
						TP53	HR, 1.57; 95% CI, 1.17–2.12
						WNT pathway	HR, 0.66; 95% CI, 0.47–0.95
						SPOP	HR, 0.63; 95% CI, 0.39–1.00
					OS	AR	HR, 4.06; 95% CI, 1.71–9.68
						Cell cycle pathway	HR, 2.03; 95% CI, 1.18–3.50
						WNT pathway	HR, 0.45; 95% CI, 0.22–0.90
						SPOP	HR, 0.33; 95% CI, 0.13–0.84
Mateo et al [25]	2020	Primary PCa	203	N/A	OS	RB1	Median 2.32 y vs. 4.28 y; p=0.006
Swami et al [42]	2020	mCSPC	121	ADT	PFS	SPOP	Median 35 mo vs. 13 mo; HR, 0.47; 95% CI, 0.25–0.87; p=0.016
					OS	SPOP	Median 97 mo vs. 69 mo; HR, 0.32; 95% CI, 0.12–0.88; p=0.027
Deek et al [39]	2021	mCSPC	294	N/A	rPFS	TP53	HR, 1.59; 95% CI, 1.04–2.41; p=0.03
					Time to CRPC	TP53	HR, 1.71; 95% CI, 1.16–2.52; p=0.01

PCa: prostate cance, BCR: biochemical recurrence, CNA: copy number alteration, HR: hazard ratio, CI: confidence interval, mCRPC: metastatic castration-resistant prostate cancer, ABI: abiraterone, TTTC: time to treatment change, ENZ: enzalutamide, CPE: concordance probability estimate, OS: overall survival, mCSPC: metastatic castration-sensitive prostate cancer, PFS: progression free survival, N/A: not applicable, ADT: androgen deprivation therapy, rPFS: radiographic PFS.

Prostate cancer with mutant SPOP shows a distinct pattern of genomic alterations, defining a new molecular subtype [20]. Boysen et al [41] reported that SPOP mutations were associated with a higher response rate to abiraterone (odds ratio, 14.50; 95% CI, 2.92–71.94; p=0.001) and a longer time on abiraterone (HR, 0.37; 95% CI, 0.20–0.69; p=0.002) in patients with mCRPC. Swami et al [42] reported that SPOP mutations were significantly associated with better PFS (median 35 mo vs. 13 mo; HR, 0.47; 95% CI, 0.25–0.87; p=0.016) and OS (97 mo vs. 69 mo; HR, 0.32; 95% CI, 0.12–0.88; p=0.027) in patients with mCSPC treated with ADT. Although AR is a ubiquitination degradation substrate of SPOP E3 ligase, prostate-cancer-associated SPOP mutants cannot bind to and promote AR degradation [43]. The SPOP mutant tumors have the highest AR transcriptional activity among prostate cancer subtypes [14]. Thus, the SPOP mutant tumors may primarily be driven by AR signaling and in turn will be responsive to AR targeted therapies [42].

Taken together, genomic alterations of TP53, RB1, AR, and cell cycle pathway are associated with poor clinical outcomes in patients with prostate cancer, whereas SPOP mutations are associated with better clinical outcomes (Table 1).

3. Liquid biopsy

A liquid biopsy is a minimally invasive procedure to investigate the cancer-related molecules in circulating tumor cells (CTCs) and cell-free tumor nucleic acids. There is a high consistency between metastatic tumor tissue and matched circulating tumor DNA (ctDNA) or CTCs [44,45,46,47]. Liquid biopsies have the advantage of detecting acquired resistance in prostate cancer [17,48]. In 2016, Ulz et al [16] performed whole-genome sequencing on plasma samples derived from patients with metastatic prostate cancer, and identified driver aberrations in cancer-related genes, including gene fusions (TMPRSS2:ERG), focal deletions (PTEN, RYBP, and SHQ1), and amplifications (AR and MYC). In serial plasma analyses, the focal amplifications were detected in 40% of cases, suggesting a high plasticity of prostate cancer genomes with newly occurring focal amplifications as a driving force in progression [16]. Although ADT rapidly reduces ctDNA availability [49], the emergence of AR amplification in ctDNA is detected during treatment with abiraterone and enzalutamide [50]. Tumor fraction in cell free DNA (cfDNA) correlates with metastatic burden, and the decline of ctDNA can be a promising biomarker for therapeutic response in patients with CRPC [51]. Decreases in cfDNA concentration independently associated with outcome in patients with metastatic prostate cancer who were treated with PARP inhibitor olaparib (HR for OS at week 8, 0.19; 95% CI, 0.06–0.56; p=0.003) [52].

Recently, a number of studies demonstrated the association between genomic alterations in liquid biopsy and clinical outcome in prostate cancer (Table 2). As sustained AR signaling pathway remains a key driver for CRPC progression [5], considerable efforts have been made to profile AR aberrations using circulating nucleic acids [53]. Resistance to AR pathway inhibitors, abiraterone and enzalutamide, has been observed in patients with CRPC harboring AR copy number gain/amplification [54,55,56,57,58,59], somatic AR mutations [54,55,56], and constitutively active AR-Vs, such as AR-V3, AR-V7, and AR-V9 [58,60]. AR copy number gain has also been associated with poor outcomes in patients receiving chemotherapy [58,61], likely reflecting aggressive intrinsic disease biology. Furthermore, genomic alterations of RB1, TP53, MYC, cell cycle pathway, and DNA repair pathway are detected in liquid biopsy, and are reported to be associated with poor clinical outcomes in patients with prostate cancer [55,62,63,64,65,66].

Table 2. Genomic alterations in liquid biopsy associated with clinical outcome.

Author	Year	Sample	Patients	Number of patients	Therapy	Endpoint	Genomic alterations	Outcome
Azad et al [54]	2015	Plasma cfDNA	mCRPC	39	ENZ	c/rPFS	AR gain/mut	Median 2.3 mo vs.7.0 mo; p<0.001
Wyatt et al [55]	2016	Plasma cfDNA	mCRPC	65	ENZ	PFS	AR gain/amp	HR, 2.92; 95% CI, 1.59–5.37; p=0.001
							Multiple AR mut	HR, 3.94; 95% CI, 1.46–10.64; p=0.007
							RB1 loss	HR, 4.46; 95% CI, 2.28–8.74; p<0.001
							MET gain	HR, 4.53; 95% CI, 1.97–10.45; p<0.001
							MYC gain	HR, 2.58; 95% CI, 1.39–4.77; p=0.003
Conteduca et al [56]	2017	Plasma cfDNA and CTC	CRPC	171	ABI or ENZ	PFS	AR gain	HR, 2.22; 95% CI, 1.48–3.34; p<0.001
							AR mut	HR, 2.59; 95% CI, 1.24–5.44; p=0.012
						OS	AR gain	HR, 4.26; 95% CI, 2.76–6.55; p<0.001
							AR mut	HR, 3.80; 95% CI, 1.77–8.15; p=0.001
				94	ENZ	rPFS	AR gain	HR, 8.06; 95% CI, 3.26–19.93; p<0.001
						OS	AR gain	HR, 11.08; 95% CI, 2.16–56.95; p=0.004
De Laere et al [60]	2017	Plasma cfDNA and CTC	CRPC	17	ABI or ENZ	PFS	ARVs	HR, 4.53; 95% CI, 1.424–14.41; p=0.0105
Kohli et al [57]	2018	Plasma cfDNA	mCRPC	70	ABI	OS	AR amp	HR, 5.25; 95% CI, 2.21–12.46; p=0.0002
Annala et al [62]	2018	Plasma cfDNA	mCRPC	202	ABI or ENZ	PFS	BRCA2/ATM	HR, 6.14; 95% CI, 3.35–11.26; p<0.001
							TP53	HR, 2.70; 95% CI, 1.86–3.91; p<0.001
Conteduca et al [61]	2019	Plasma cfDNA	mCRPC	163	DTX	OS	AR gain	HR, 1.61; 95% CI, 1.08–2.39; p=0.018
De Laere et al [63]	2019	Plasma cfDNA and CTC	mCRPC	168	ABI or ENZ	PFS	TP53	HR, 1.88; 95% CI, 1.18–3.00; p=0.008
Sonpavde et al [64]	2019	Plasma cfDNA	mCRPC	163	N/A	OS	MYC amp	HR, 5.85; 95% CI, 2.17–15.77; p<0.001
Fettke et al [58]	2020	Plasma cfDNA/cfRNA	mCRPC	67	ABI, ENZ, DTX, CBT	c/rPFS	AR gain	HR, 3.2; 95% CI, 1.3–8.0; p=0.01
						OS	AR gain	HR, 2.8; 95% CI, 1.1–7.2; p=0.04
Du et al [59]	2020	Plasma cfDNA	mCRPC	88	ABI	TTTC	AR amp	HR, 3.27; 95% CI, 1.78–6.84; p=0.0003
							OPHN1 amp	HR, 3.70; 95% CI, 1.08–7.00; p=0.0002
Ritch et al [65]	2020	Plasma cfDNA	mCSPC	210	ADT	Time to CRPC	dMMR	Median 9.1 mo vs.18.2 mo; p=0.00025
Kohli et al [66]	2020	Plasma cfDNA	mCRPC	69	N/A	OS	RB1	HR, 4.2; 95% CI, 2.0–8.7; p=0.00015
			mCSPC	73	N/A	OS	ATM, BRCA1, BRCA2, and CHEK2	HR, 4.0; 95% CI, 1.4–11.8; p=0.0000475

cfDNA: cell free DNA, mCRPC: metastatic castration-resistant prostate cancer, ENZ: enzalutamide, c/rPFS: clinical/radiographic progression free survival, HR: hazard ratio, CI: confidence interval, CTC: circulating tumor cell, CRPC: castration-resistant prostate cancer, ABI: abiraterone, OS: overall survival, ARVs: androgen receptor splice variants, DTX: docetaxel, N/A: not applicable, cfRNA: cell free RNA, CBT: cabazitaxel, ADT: androgen deprivation therapy, dMMR: deficiency in mismatch repair genes.

4. Actionable molecular alterations

DNA repair alterations are observed in about one fourth of prostate cancer, in which most commonly mutated genes include BRCA2, BRCA1, and ATM [23]. These gene alterations can occur at either a somatic or a germline level [23]. Although the mutations in DNA-repair genes occurred more often in Black men than in White men [28], the germline alterations in DNA-repair genes were identified in 31% of the patients in Asian populations, including mutations in BRCA2 (5.3%) [67]. The germline mutations in BRCA1/2 and ATM are associated with prostate cancer risk [68], as well as aggressive prostate cancer phenotype [69,70,71,72,73,74]. Family history of cancer remains a foundation of genetic risk assessment, especially inquiring about prostate cancer as well as non-prostate cancers, including breast, ovary, pancreas, and melanoma, with their known association with mutations in BRCA1/2. [75]. BRCA1/2 and ATM are involved in homologous recombination repair. Tumors that lose the homologous recombination pathway are preferentially sensitive to PARP inhibition via the mechanism of synthetic lethality [76]. A randomized, phase 3 trial evaluated the PARP inhibitor olaparib in men with mCRPC who had disease progression while receiving a new hormonal agent (e.g., enzalutamide or abiraterone) [11]. Among patients who had at least one alteration in BRCA1, BRCA2, or ATM, radiological PFS was significantly longer in the olaparib group than in the control group (median 7.4 mo vs. 3.6 mo; HR, 0.34; 95% CI, 0.25–0.47; p<0.001) [11].

The solid tumors which harbor deficiency in mismatch repair genes (dMMR), such as MSH2, MSH6, PMS2, and MLH1, can be effectively treated by the anti–programmed cell death protein 1 (PD-1) antibody pembrolizumab, regardless of tissue of origin [77]. In 2019, Abida et al [78] reported that 32 of 1,033 patients with prostate cancer (3.1%) had microsatellite instability (MSI)–high or dMMR, of whom 7 (21.9%) carried a germline mutation in a Lynch syndrome–associated gene. The dMMR prostate cancers are associated with higher MSI scores, and enriched for higher T cell infiltration and PDL1 protein expression [79]. Screening for MSI-H/dMMR in advanced prostate cancer is beneficial for identifying patients who have potential for durable responses to anti–PD-1/PD-L1 therapy.

Approximately 40% to 60% of mCRPC tumors have a functional loss of PTEN, a tumor suppressor phosphatase, which causes hyperactivation of the PI3K–Akt–mTOR pathway [13,15]. Ipatasertib (GDC-0068) is a novel selective ATP-competitive small-molecule inhibitor of all three isoforms of Akt. Sensitivity to ipatasertib is associated with high tumoral levels of phosphorylated Akt, PTEN protein loss or genetic mutations, and PIK-3CA kinase domain mutations [80]. In a phase 2 study, combined treatment with abiraterone and ipatasertib showed superior antitumor activity to abiraterone alone in patients with mCRPC, especially in patients with PTEN-loss tumors [12]. A phase 3 trial is ongoing to test the efficiency of ipatasertib plus abiraterone in patients with mCRPC (IPATential150, NCT03072238).

5. Neuroendocrine prostate cancer

Neuroendocrine prostate cancer is an aggressive variant of prostate cancer, which may arise de novo or in patients who were previously treated with hormonal therapies [81]. A subset of mCRPC tumors show small-cell neuroendocrine features during disease progression on metastatic biopsy [82]. This phenomenon may reflect an epithelial plasticity that enables tumor adaptation in response to AR-targeted therapies [9]. Neuroendocrine prostate cancer is associated with worse OS, even when platinum-based chemotherapy is used [81,83]. In 2016, Beltran et al [9] demonstrated that CRPC with neuroendocrine features (CRPC-NE) is associated with low AR signaling and a paucity of somatic AR gene alterations, concurrent loss of RB1 and TP53 (in 53.3% of CRPC-NE vs 13.7% of CRPC-Adenocarcinoma; p<0.0004), changes in DNA methylation profile, and upregulation of mRNA encoding the histone methyltransferase EZH2. There was high concordance between ctDNA and biopsy tissue genomic alterations in patients with CRPC-NE, supporting the use of ctDNA profile to recognize transformation to CRPC-NE during the course of CRPC treatment [84].

6. Clinical utility of genomic profiling

Tumor genomic profiling is a fundamental component of precision medicine, enabling the identification of genomic alterations in genes and pathways that can be targeted therapeutically. In 2017, the U.S. Food and Drug Administration (FDA) approved two comprehensive next generation sequencing panel assays, MSK-IMPACT and FoundationOne CDx [85]. At Memorial Sloan Kettering Cancer Center, MSK-IMPACT was developed and implemented to detect protein-coding mutations, CNAs, and selected promoter mutations and structural rearrangements in 341 (and, more recently, 468) cancer-associated genes [85,86]. FoundationOne CDx, a similar 324 gene assay, was developed to identify actionable genomic aberrations in cancer [85]. For the effective analysis of genomic tests, the quality of tumor tissue samples is crucial. Although formalin-fixed paraffin-embedded blocks obtained from prostate tumor biopsies are widely used to identify clinically actionable molecular alterations, DNA degradation can occur during mid- to long-term storage of samples [87]. Genomic heterogeneity is commonly detected in primary prostate cancer [88,89,90]. Furthermore, genomic alterations can occur during CRPC progression [16,91]. Thus, a metastatic biopsy provides a reasonable assessment for genomic profiling in patients with mCRPC [92]. In 2020, FoundationOne Liquid CDx, a novel 324-Gene cfDNA-based comprehensive genomic profiling assay, was approved by the FDA [93]. This laboratory test can be used as a companion diagnostic tool that can identify if patients with mCRPC harbor BRCA1/2 alterations which may benefit from treatment with PARP inhibitors [93]. After eliminating clonal hematopoiesis variants, ctDNA was detected in 87.9% of patients with prostate cancer showing its high detectability [94]. Thus, cfDNA-based genomic tests provide a noninvasive approach to elucidate a patient's genomic landscape and actionable information.

CONCLUSIONS

The integrative genomic profiling of prostate tumors has provided comprehensive information and novel discoveries which improve our understanding of the disease. A number of mCRPC harbor clinically actionable molecular alterations, including changes to DNA damage repair pathway and PTEN/PI3K signaling. The genomic alterations of TP53, RB1, AR, and cell cycle pathway are associated with poor clinical outcomes, whereas SPOP mutation is associated with better clinical outcomes. Several genomic profiling tests are emerging to identify patients who could benefit from targeted therapy. Thus, the genomic profiling of prostate cancer provides useful information for diagnosis and treatment in this new era of precision medicine.