FOXA1 in prostate cancer

Abstract

Most prostate cancers initially respond to androgen deprivation therapy (ADT). With the long-term application of ADT, localized prostate cancer will progress to castration-resistant prostate cancer (CRPC), metastatic CRPC (mCRPC), and neuroendocrine prostate cancer (NEPC), and the transcriptional network shifted. Forkhead box protein A1 (FOXA1) may play a key role in this process through multiple mechanisms. To better understand the role of FOXA1 in prostate cancer, we review the interplay among FOXA1-targeted genes, modulators of FOXA1, and FOXA1 with a particular emphasis on androgen receptor (AR) function. Furthermore, we discuss the distinct role of FOXA1 mutations in prostate cancer and clinical significance of FOXA1. We summarize possible regulation pathways of FOXA1 in different stages of prostate cancer. We focus on links between FOXA1 and AR, which may play different roles in various types of prostate cancer. Finally, we discuss FOXA1 mutation and its clinical significance in prostate cancer. FOXA1 regulates the development of prostate cancer through various pathways, and it could be a biomarker for mCRPC and NEPC. Future efforts need to focus on mechanisms underlying mutation of FOXA1 in advanced prostate cancer. We believe that FOXA1 would be a prognostic marker and therapeutic target in prostate cancer.

INTRODUCTION

Prostate cancer is the second leading cause of estimated cancer deaths and accounted for 21% of all cases among men in 2020.1 Prostate cancer has the highest 5-year survival rate for all stages combined among all cancers.1 Although the overall incidence of prostate cancer has declined, distant-stage diagnoses have increased.2 Men diagnosed with primary prostate cancer are usually treated with curative treatment, and more than 30% of patients progress to metastatic disease within 10 years.3,4 Androgen deprivation therapies (ADTs) are then used to treat metastatic disease and have an initial response rate of approximately 80%.5,6 Despite the efficacy of ADT, patients ultimately develop resistance and progress to metastatic castration-resistant prostate cancer (mCRPC).5,7 Notably, Asian patients usually present with higher tumor grades in prostate cancer at diagnosis compared with Western populations.8 Therefore, there remains a need to understand the genomic underpinnings of prostate cancer development.

Forkhead box protein A1 (FOXA1), also known as hepatocyte nuclear factor 3-alpha (HNF3A), was first identified as a transcription factor for the expression of transthyretin and α₁-antitrypsin in the liver.9 FOXA1 was found to collaborate with NK2 homeobox 1 (NKX2-1) and functioned as an oncogene in lung cancer.10 Metastasis of pancreatic ductal adenocarcinoma was linked with FOXA1-dependent enhancer reprogramming.11 In the past few years, FOXA1 expression has been reported to be related to several human cancers, as both an oncogene and a tumor suppressor gene depending on the specific cancer types.12,13 Even in the same type of cancer, FOXA1 can play different roles in different subtypes.14,15

Recent findings have shown that FOXA1 plays an important role in regulating steroid receptor functions.16 Prostate cancer and breast cancer are the two most prominent hormone-related tumors, so they are not surprisingly related to FOXA1. As FOXA1 is obligatory for androgen receptor (AR)-mediated genes for CRPC, it is thought to play an important role in prostate cancer progression.17 Recent studies revealed that the genetic mutation frequency of FOXA1 is increased in mCRPC patients, suggesting that FOXA1 mutations may be a factor leading to aggressive prostate cancer.18,19 FOXA1 was also found to be the transcription factor (TF) motif most frequently enriched in prostate cancer-specific enhancer-promoter loops.20 Notably, FOXA1 mutations were found in 41% of a Chinese cohort, whereas FOXA1 was mutated in only 4% of prostate cancers in The Cancer Genome Atlas (TCGA) database.21,22 This review focuses on the possible roles of FOXA1 in prostate cancer initiation, progression, and drug resistance and as a new therapeutic target.

FOXA1 IN THE DEVELOPMENT OF THE PROSTATE

The secretory function of the prostate is dependent on epithelial AR expression, whereas morphogenesis depends on stromal AR expression.23,24 Considering the established role of FOXA1 in regulating AR transcriptional activity in the prostate, it is not surprising that FOXA1 is involved in prostate development.25 FOXA1, also known as HNF3A, was first identified as a necessary transcription factor for α₁-antitrypsin and transthyretin in the liver.9 Interestingly, FOXA1 mRNA expression levels in the prostate are much higher than those in liver tissues.26 Expression studies on the prostate showed that FOXA1 is expressed in epithelial cells but not in the stromal cells.26,27 By analyzing Foxa1-deficient mouse prostates, Gao et al.28 found an altered ductal pattern, which indicated that Foxa1 promoted epithelial cell maturation and regulated early ductal pattern formation. As AR has separate and distinct functions in the stroma and epithelium, Foxa1 probably also plays different roles in different stages. In addition to its function in ductal morphogenesis and cytodifferentiation, Foxa1 has a secretory function as well.29 Biochemical analysis confirmed similar function of FOXA1 in humans.29 In summary, FOXA1 plays a distinct role in prostate morphogenesis, differentiation and secretion.

FOXA1 AND AR IN PROSTATE CANCER

Androgens play an important role in the development and maturation of the prostate, influencing the differentiation and proliferation of the epithelial cells.30 AR, activated by androgens, is a key mediator of targeted genes that regulate the development and differentiation of the luminal epithelium. AR plays a crucial role in the growth and maintenance of the prostate gland. Mice with AR function defects do develop prostate gland or prostate cancer.31,32 Clinical evidence has shown that AR signaling participates in all stages of prostatic disease, including advanced and metastatic prostate cancer.33,34 In addition, the development of CRPC also requires AR signaling.

Regulation of AR by FOXA1

Different AR-binding sites (ARBs) were identified in tumor tissues and benign/normal tissues.35 This difference between tumor and benign tissues in ARB preference is called the AR malignancy shift (AMS). FOXA1 was found at more than 60% of ARBs driving the transcriptional programs in prostate cancer cells and was considered one of the components of the AMS.36,37 Generally, FOXA1 serves as a pioneer factor for AR. It can bind to target sites in silent chromatin and trigger regulation in the genome before other transcription factors, which is why FOXA1 is called “pioneer factor”.38 FOXA1 is considered an auxiliary factor for AR to promote prostate cancer progression. However, genome-wide elucidation of the AR cistrome in prostate cancer cell lines showed approximately 70% overlap between FOXA1 and AR-binding sites, indicating an important role of FOXA1 in the transcriptional regulation of prostate cancer.39 Sahu et al.39 identified three distinct classes of ARBs and AR-regulated transcription programs: (1) the sites independent of FOXA1, (2) the sites pioneered by FOXA1 to recruit for AR, and (3) the sites that are masked by FOXA1 and activated by AR only when FOXA1 was depleted. Moreover, an in vitro experiment showed that not only did FOXA1 depletion not reduce the ARBs, but it could also result in an increased number of ARBs.12,39 Furthermore, FOXA1 was reported to inhibit the transcriptional activity of AR in prostate cancer cell lines possibly by competing with AR coactivators.40 These experiments showed the two-way regulation of AR by FOXA1, depending on the cellular context.12

Relationship between FOXA1 and AR in different types of prostate cancer

In primary prostate cancer tissue, FOXA1 was found to promote cell proliferation, and high FOXA1 expression was related to high AR expression.41 FOXA1 expression was also reported to be positively associated with AR and lymph node metastases in a cohort study of prostate cancer.42 In CRPC, FOXA1 was found to be a key regulator of the cell cycle, as it could promote G₂ to M-phase transition and G₁ to S-phase transition through two distinct pathways and its activity was reported to be essential for AR function.17,43 However, AR was also reported to continuously bind some specific genomic loci that are open to chromatin structures and independent of FOXA1 in CRPC.44 Furthermore, previous studies reported that FOXA1 was downregulated in neuroendocrine prostate cancer (NEPC) and CRPC and showed negative correlation with AR.12,45,46

Previous studies showed that FOXA1 silencing did not change the expression of prostate-specific antigen (PSA) and transmembrane protease serine 2 (TMPRSS2), which are key factors in androgen signaling, suggesting that FOXA1 may not play a crucial role in AR signaling of AR-dependent prostate cancer cells.38,47 Interestingly, another study showed a reduction in TMPRSS2 after suppressing FOXA1 expression in androgen-independent cell lines.48 FOXA1 is also required for chromatin looping and enhancer activity by regulating TMPRSS2 and long noncoding RNA (lncRNA) prostate cancer-associated transcript 38 (PRCAT38).49 Regulation of some AR-specific target genes is dependent on FOXA1, whereas regulation of AR-V7-specific target genes is independent of FOXA1.50,51 Paakinaho et al.52 found that androgens could induce the chromatin binding of FOXA1 in prostate cancer cell lines independent of AR. Taken together, these results illustrate multiple functions of FOXA1 regulating AR transcriptional activity through various mechanisms. It also reveals the possible different mechanisms of FOXA1 between AR-dependent and AR-independent prostate cancer.

THE FOXA1 REGULATORY CIRCUIT IN PROSTATE CANCER

Regulation of FOXA1 DNA binding by histone modification

FOXA1 is typically recruited in the epigenetic signature consisting of lysine 4 on histone 3 (H3K4) into functional regulatory elements.45 Removal of mono- and di-methylated H3K4 may lead to suppression of FOXA1 and AR recruitment.53,54 As FOXA1 has the ability to bind nucleosomes, different binding sites to chromatin of FOXA1 are dependent on the methylation and distribution of H3K4.53 Acetylation of lysine 27 on histone 3 (H3K27ac), another histone marker of active enhancers, could be downregulated by FOXA1.49 Moreover, FOXA1 binding at the enhancers could be impaired by inhibiting the transcription coactivator p300.49 Wang et al.12 detected three classes of AR-binding events after hormone stimulation, indicating that in the absence of FOXA1, histone acetyltransferase was needed for initiating AR transcriptional events. Collectively, these recent studies have begun to elucidate the specific control mechanisms that recruit FOXA1 to chromatin.

Targets of FOXA1

NK3 homeobox 1 (NKX3-1) is known as the earliest prostate marker.55 In prostate epithelial cells of mice, null Foxa1 prostate showed decreased NKX3-1 expression.28 As FOXA1, GATA binding protein 3 (GATA3), and estrogen receptor alpha (ERα) form a network regulating the development of luminal breast epithelial cells, GATA3 was thought to be required for FOXA1.56,57 GATA3 was necessary for prostate function in adult mice.58 However, GATA3 was not expressed in prostate cancer cell lines at the mRNA level.59 Moreover, GATA3 expression was reported to be completely negative in both primary and metastatic prostate cancer tissues tested by immunohistochemistry.42 FOXA1 loss was observed to lead to obviously high expression of transforming growth factor-beta 3 (TGFB3), which is a key factor in the TGF-β pathway in multiple prostate cancer cell lines.19 Chromatin immunoprecipitation followed by sequencing (ChIP-seq) analysis showed that FOXA1 inhibits TGFB3 expression by binding to its enhancer.19 FOXA1 depletion upregulated TGFB3, which was found to activate the TGF-β pathway and induce cell invasion and epithelial–mesenchymal transition (EMT).19 FOXA1 could also directly inhibit interleukin 8 (IL-8) transcription and snail family transcriptional repressor 2 (SNAI2) expression.46,60 In addition, FOXA1 controlled cell proliferation by inhibiting insulin-like growth factor binding protein 3 (IGFBP-3) through the mitogen-activated protein kinase (MAPK) and protein kinase B (Akt) pathways.41 FOXA1 was significantly enriched in a chromatin region binding with N-Myc, and FOXA1 binding was related to androgen-dependent changes in N-Myc binding.61 It was also found that FOXA1 upregulated the expression of lncRNAs such as PRCAT38 and DSCAM antisense RNA 1 (DSCAM-AS1) by binding their promoters.37,62 Recently, FOXA1 was found to increase the production of proangiogenic factors, including endoglin, endothelin-1, and epidermal growth factor (EGF).63 In addition to being an oncogene, FOXA1 has also been found to enhance chemotherapy and immunotherapy resistance by suppressing signal transducer and activator of transcription 2 (STAT2) DNA-binding activity by binding the STAT2 DNA-binding domain.64

Regulation of FOXA1

It has been reported that nuclear factor I C (NFIC), TLE family member 3 (TLE3), and histone deacetylase (HDAC7) act as FOXA1 corepressor genes regulating breast cancer cell invasion.65,67,68 Wang et al.69 also found that knockdown of NFIC, TLE3, and HDAC7 promoted FOXA1-repressed invasion genes in LNCaP cell lines. Moreover, the small molecule JQ1 could promote prostate cancer cell invasion by directly interacting with FOXA1 and keeping FOXA1 from binding to its corepressor genes, such as NFIC, TLE3, and HDAC7, thereby activating invasive-related genes.69 In terms of posttranslational modification, histone lysine-specific demethylase 1 (LSD1) was found to increase the chromatin binding of FOXA1 by demethylating lysine 270.70 In addition, enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2) could increase the stability of FOXA1 by methylating FOXA1 protein at lysine 295.71 Octamer-binding transcription factor 4 (OCT4) can activate the FOXA1/AR axis by promoting complex formation.72 Clinical evidence showed that hyperglycemia could induce high expression of FOXA1 and IGFBP-2.73 In addition, negative regulation between FOXA1 and IGFBP-2 was observed in normal epithelial cells but not in cancer cells.73 A positive association was observed between IGFBP-2 and FOXA1 in prostate cancer tissues, and the FOXA1/IGFBP-2 axis played an important role in high glucose-induced EMT in prostate cancer cells.73 Poly(ADP-ribose) polymerase-2 (PARP-2) is a member of the PARP family, which is responsible for multiple DNA damage repair pathways.74 As PARP-2 showed higher expression levels in prostate cancer tissue than in benign prostate tissue, PARP-2 was found to enhance AR-mediated transcription in prostate cancer by interacting with FOXA1.75 In addition, PARP-2 was found to have no interaction with AR by coimmunoprecipitation (Co-IP) in prostate cancer cell lines.75 Taken together, the inhibition of PARP-2 could disrupt the function of FOXA1 in AR transcription. Analysis of a TCGA cohort showed that fatty acid-binding protein 5 (FABP5) overexpression was related to FOXA1 mutations in e-twenty-six (ETS)-negative subtypes.76 The transcription factor ONECUT2 (OC2) was identified as a key factor in mCRPC.77 Endogenous OC2 was found to bind the FOXA1 promoter and gene body to inhibit its expression.77 The silencing of nuclear export protein exportin-1 (XPO1) inhibited the FOXA1 expression.78

Cis-regulatory elements (CREs) are regions of noncoding DNA that play a critical role in regulating gene expression by binding to transcription factors.79 Six CREs in the FOXA1 regulatory plexus were identified, and their repression led to significantly decreased FOXA1 expression.80 Notably, microRNA-194 (miR-194) was found to decrease the levels of FOXA1 protein and mRNA in prostate cancer cell lines and was also negatively correlated with FOXA1 in clinical specimens.81

A recent study showed that ectopic expression of achaete-scute family BHLH transcription factor 1 (ASCL1) and NKX2-1 in prostate adenocarcinoma cells could reprogram FOXA1 to bind to neuroendocrine elements and promote the development of NEPC.82 The FOXA1 regulatory circuit is summarized in Table 1.

Table 1.

Modulators of forkhead box protein A1 function in prostate cancer

Gene	Function	Expression in prostate cancer	Reference
FOXA1-regulated genes
TMPRSS2	Induced	Increased	49
PRCAT38	Induced	Increased	37
NKX3-1	Induced	Increased	28 126
DSCAM-AS1	Induced	Increased	62
TGFB3	Repressed	NA	19
IL-8	Repressed	NA	60
SNAI2	Repressed	NA	46
IGFBP-3	Repressed	NA	41
AR	Repressed	NA	40
STAT2	Repressed	NA	64
Genes involved in FOXA1 expression
TLE3	Corepressor	NA	69
HDAC7	Corepressor	NA	69
NFIC	Corepressor	NA	69
MYCN	Collaborator	NA	61
Genes/proteins that may control FOXA1 expression
H3K4me1	Promoting	NA	45
H3K4me2	Promoting	NA	45
H3K27ac	Promoting	NA	82
OC2	Increased binding	Increased	77
LSD1	Increased binding	Increased	70
OCT4	Increased binding	NA	72
XPO1	Repression	Increased	78
miR-194	Repression	Increased	81
NKX2-1	Reprogramming	NA	82
ASCL1	Reprogramming	NA	82
EZH2	Increased stability	NA	71

FOXA1: forkhead box protein A1; TMPRSS2: transmembrane protease serine 2; PRCAT38: prostate cancer-associated transcript 38; NKX3-1: NK3 homeobox 1; DSCAM-AS1: DSCAM antisense RNA 1; TGFB3: transforming growth factor-beta 3; IL-8: interleukin 8; SNAI2: snail family transcriptional repressor 2; IGFBP-3: insulin-like growth factor binding protein 3; AR: androgen receptor; STAT2: signal transducer and activator of transcription 2; TLE3: TLE family member 3; HDAC7: histone deacetylase; NFIC: nuclear factor I C; OC2: ONECUT2; LSD1: histone lysine-specific demethylase 1; OCT4: octamer-binding transcription factor 4; XPO1: nuclear export protein exportin-1; miR-194: microRNA-194; NKX2-1: NK2 homeobox 1; ASCL1: achaete-scute family BHLH transcription factor 1; EZH2: enhancer of zeste 2 polycomb repressive complex 2 subunit; NA: not available

Mutation of FOXA1

Mutations, including losses of phosphatase and tensin homolog (PTEN), NKX3-1, and ETS fusion, were commonly identified.83,84,85 Recently, some new mutations were found including mediator complex subunit 12 (MED12), FOXA1, and speckle-type BTB/POZ protein (SPOP).86 FOXA1 mutation was found in 4%–9% of primary prostate tumors and in 12%–13% of mCRPCs in a Western cohort, suggesting that a high frequency of FOXA1 mutation may be related to more aggressive prostate cancer type.86,87 However, a recent study found that FOXA1 mutation was found in 41% of Chinese cohort in clinical localized prostate cancer, indicating the possibility of different mechanisms underlying the prostate cancer progression.21 Moreover, a cohort of African Americans showed an 8% mutation of FOXA1 in primary prostate tumors.88

Most FOXA1 mutations were found located in the Wing2 region of forkhead (FKHD). Genomic analysis showed that FOXA1 mutations around residue 253 in the FKHD domain caused prostate cancer cell growth under hormone-deprived conditions, leading to CRPC.89 Missense mutations at the winged-helix FKHD of FOXA1 reduced the dependency of prostate cancer cell on AR signaling and promoted prostate cancer progression.90 Moreover, TCGA studies revealed that FOXA1 mutations promoted AR-dependent transcriptional activities.22

Parolia et al.87 mapped FOXA1 mutations into two structural patterns: (1) indel, in-frame insertion, and missense mutations at FKHD; and (2) truncating frameshift mutations restricted to the regulatory domain. According to FOXA1 structure, FOXA1 mutations were categorized into three classes: class 1, which comprises all mutations within the FKHD in early prostate cancer; class 2, which includes mutations in the C-terminal domain of FKHD mainly in metastatic prostate cancers; and class 3, which comprises duplications and translocations within the FOXA1 locus mostly in mCRPC.87 Adams et al.91 defined two hotspots in the FKHD domain from 3086 human prostate cancers: (1) wing2 mutations that are detected mostly in adenocarcinomas at all stages; and (2) the highly conserved DNA-contact residue R219, which is enriched mostly in NEPC with metastatic tumors (Figure 1). Mutations of FOXA1 can increase, replace, and reprogram AR function to promote growth of prostate cancer (Figure 2).

Figure 1.

Structure of FOXA1 and classes of FOXA1 alterations. FOXA1: forkhead box protein A1; RD: regulatory domain.

Figure 2.

FOXA1 and AR-regulated pathway in AR-dependent prostate cancer and AR-independent prostate cancer. (a) Class 1 mutant FOXA1/wild-type FOXA1, opens the closed chromatin and recruits AR to promote oncogenic AR signaling. (b) In AR-independent prostate cancer, mutant FOXA1/wild-type FOXA1 reprograms or replaces AR function, and promotes AR-independent growth. FOXA1: forkhead box protein A1; Wnt: wingless-type MMTV integration site family; AR: androgen receptor; ARE: androgen-receptor response element; EMT: epithelial–mesenchymal transition.

Insertions and deletions (indels) in prostate cancer remain largely unexplored due to technical limitations.92 Notably, in the Chinese Prostate Cancer Genome and Epigenome Atlas (CPGEA), indels account for 61.8% of FOXA1 mutations.21 In-frame indel and missense of FOXA1 mutations were found to affect the nicotinate and nicotinamide metabolism pathways.21 With the application of new technologies, somatic FOXA1 untranslated region (UTR) mutations have been identified with high specificity.93

FOXA1 IN NEUROENDOCRINE PROSTATE CANCER

Advanced prostate cancer is commonly treated with ADT, and most patients finally develop androgen resistance and develop CRPC.94 New generation AR antagonists with higher affinity have been used for CRPC. Under treatment with these AR antagonists (e.g., abiraterone and enzalutamide), some CRPC patients develop into therapy-induced NEPC (t-NEPC).95,96,97

Studies have shown that lineage plasticity plays a critical role in prostate cancer therapy resistance.95 N-Myc (encoded by MYCN), a transcription factor, is necessary for neurodevelopment and is overexpressed in retinoblastoma, neuroblastoma, and glioblastoma multiforme.98,99,100,101,102,103 Interestingly, N-Myc is not found in the epithelial lineage of prostate tissue but is overexpressed in some CRPCs and a majority of NEPC.104,105 Genomic locus binding comparison revealed the nearby binding of N-Myc and FOXA1.61 Overlap between FOXA1 and N-Myc was also confirmed, and FOXA1 played an important role in regulating N-Myc binding.61 In addition, mutations of FOXA1 play an important role in the neuroendocrine process, and two different R219 mutants of FOXA1 blocked luminal differentiation and activated neuroendocrine transcription.91 More recently, miR-194 was reported to promote the neuroendocrine phenotype by decreasing FOXA1 binding.81 FOXA2, another member of the FOXA family, was found to be a NEPC TF.106,107 However, Baca et al.82 analyzed NEPC patient-derived xenografts (PDXs) and found that FOXA1 was a key factor for NEPC even if there was no FOXA1 mutation or no FOXA2 expression. FOXA1 plays a completely different role in prostate adenocarcinoma and NEPC. FOXA1 was reprogrammed to neuroendocrine-specific regulatory elements in NEPC. Possible regulation pathways are summarized in Figure 3.

Figure 3.

Pathways of FOXA1 involved in different types of prostate cancer. EMT: epithelial–mesenchymal transition; IFN: interferon; CRPC: castration-resistant prostate cancer; NEPC: neuroendocrine prostate cancer; FOXA1: forkhead box protein A1; TMPRSS2: transmembrane protease serine 2; PRCAT38: prostate cancer-associated transcript 38; NKX3-1: NK3 homeobox 1; DSCAM-AS1: DSCAM antisense RNA 1; TGFB3: transforming growth factor-beta 3; IL-8: interleukin 8; SNAI2: snail family transcriptional repressor 2; IGFBP-3: insulin-like growth factor binding protein 3; AR: androgen receptor; STAT2: signal transducer and activator of transcription 2; TLE3: TLE family member 3; HDAC7: histone deacetylase; NFIC: nuclear Factor I C; OC2: ONECUT2; LSD1: histone lysine-specific demethylase 1; OCT4: octamer-binding transcription factor 4; XPO1: nuclear export protein exportin-1; miR-194: microRNA-194; NKX2-1: NK2 homeobox 1; ASCL1: achaete-scute family BHLH transcription factor 1; EZH2: enhancer of zeste 2 polycomb repressive complex 2 subunit; EGF: epidermal growth factor; AR: androgen receptor; lncRNA: long noncoding RNA; miRNA: microRNA; CHGA: chromogranin A.

CLINICAL RELEVANCE OF FOXA1

The prognostic role of FOXA1 in many different cancers has been widely explored.86,108,109 Studies on the relationship of FOXA1 with prognosis in prostate cancer have shown controversial results.110 High FOXA1 expression in primary prostate cancer showed poor disease-specific survival in 350 patients with a hazard ratio (HR) of 2.89 (95% confidence interval [CI]: 1.02–8.21).39 In addition, FOXA1 expression levels were significantly associated with Gleason score and pT stage in primary prostate cancer.41,111 Notably, patients with high AR expression and low FOXA1 expression had better prognosis than patients with high expression of both AR and FOXA1.39 Another cohort study of 102 patients also showed that a high expression level of FOXA1 predicted poor biochemical recurrence (HR: 5.0, 95% CI: 1.2–21.1, P = 0.028).112 Other studies showed that high FOXA1 expression was related to poor disease-free survival, but FOXA1 expression was lower in metastatic tumors.12,113,114 Based on serum PSA, high FOXA1 expression was found to be related to shorter relapse-free survival.111 An overview of the different roles of FOXA1 in prostate cancer is provided in Table 2.

Table 2.

Overview of different roles of forkhead box protein A1 in prostate cancer

Study	Cancer type	Expression relative to normal	Expression relative to primary cancer	Clinical relevance of high FOXA1 expression	Method	Activity
Mirosevich et al.25 2006	Adenocarcinoma, neuroendocrine small cell carcinoma	No difference	NA	NA	IHC	Unclear
Sahu et al.39 2011	Primary prostate cancer	Increased	NA	Poor cancer-specific survival	IHC	Oncogenic
Wang et al.12 2011	Metastatic prostate cancer	NA	Decreased in metastatic tumor	Better cancer-specific survival	PCR	Tumor suppressive
Jain et al.42 2011	Metastatic prostate cancer	NA	Increased in metastatic lesion	More metastasis, high Gleason score, high PSA, and more angiolymphatic invasion	IHC	Oncogenic
Imamura et al.41 2012	Primary prostate cancer	Increased	NA	Poor biochemical recurrence-free survival	IHC	Oncogenic
Gerhardt et al.111 2012	Primary prostate cancer	Increased	NA	Poor progression-free survival	IHC	Oncogenic
	Metastatic prostate cancer	NA	Increased in metastatic tumor	Poor progression-free survival	IHC	Oncogenic
	CRPC	NA	Increased	Poor progression-free survival	IHC	Oncogenic
Jin et al.46 2013	Primary prostate cancer	Slightly increased	NA	NA	PCR	Unclear
	Metastatic prostate cancer	Decreased	Decreased in metastatic tumor	Inhibited prostate cancer metastasis	PCR	Tumor suppressive
Robinson et al.112 2014	Primary prostate cancer	Increased	NA	Poor biochemical recurrence-free survival	IHC	Oncogenic
Tsourlakis et al.116 2017	Primary prostate cancer	Increased	NA	Poor PSA recurrence-free survival	IHC	Oncogenic
Mansor et al.73 2020	Primary prostate cancer	No difference	NA	NA	IHC	Unclear

FOXA1: forkhead box protein A1; IHC: immunohistochemistry; PCR: polymerase chain reaction; CRPC: castration-resistant prostate cancer; PSA: prostate-specific antigen; NA: not available

ETS-related gene (ERG) status and TMPRSS2:ERG rearrangement are commonly observed in prostate cancer. In a small-sized cohort study, no significant association was found between ERG expression and prognosis of prostate cancer.115 High FOXA1 expression was found to be related to worse recurrence-free survival in ERG-negative prostate cancer but not in ERG-positive prostate cancer.116

Moreover, another cohort study from Spain found that FOXA1 or SPOP1 mutations were more common in Gleason grade group 5 and were associated with PSA recurrence in ERG wild-type tumors.117 However, a cohort study of African Americans failed to show an association between FOXA1 mutation and metastasis (HR: 0.34, 95% CI: 0.04–2.76, P = 0.310).88 The HRs and 95% CIs for tumor progression survival in association with FOXA1 mutation are listed in Table 3. Notably, FOXA1 3’-UTR mutations were found in early-stage prostate cancer.93 Combined with other databases, FOXA1 3’-UTR indel mutations were commonly seen in both primary and metastatic prostate cancers.118 A cohort study of 202 mCRPC patients also revealed that FOXA1 3’-UTR indel mutations did not influence OS from mCRPC treatment initiation and PFS on first-line abiraterone or enzalutamide therapy.119 Although FOXA1 mutations were also commonly observed in breast cancer and bladder cancer, the FOXA1 3’-UTR mutations were not detected in these cancers, suggesting that FOXA1 UTR indels could be a specific marker for the diagnosis and screening of prostate cancer.93,120

Table 3.

Clinical relevance of forkhead box protein A1 mutation

Study	Cancer progression	Univariate analysis		Multivariate analysis		Clinical relevance	Main population
		HR (95% CI)	P	HR (95% CI)	P
Adams et al.91 2019	Progression to biochemical recurrence	2.92 (2.12–4.02)	0.0013	NA	NA	Significant worse in FOXA1 mutation group	Caucasian
	Progression to metastatic disease	4.59 (3.17–6.64)	0.0001	NA	NA	Significant worse in FOXA1 mutation group	Caucasian
Hernández-Llodrà et al.117 2019	Progression to biochemical recurrence	NA	0.0009	5.76 (1.27–26.06)	0.023	Significant worse in FOXA1 mutation group	Caucasian
Faisal et al.88 2020	Progression to metastatic disease	0.57 (0.08–4.29)	0.586	0.34 (0.04–2.76)	0.31	Better in FOXA1 mutation group but not significant	African American

FOXA1: forkhead box protein A1; HR: hazard ratio; CI: confidence interval; NA: not available

CONCLUSIONS

While the role of AR in promoting prostate cancer progression has been established, recent studies have focused on the coregulators of AR such as FOXA1. As relationships between FOXA1 and AR have been widely researched, FOXA1 may play an independent role in the carcinogenesis of prostate cancer. The FOXA1 expression level in primary prostate cancer tissue is high and differs in various subtypes of prostate cancer. High FOXA1 expression usually predicts poor prognosis in prostate cancer. FOXA1 regulates the expression of multiple genes, especially AR in the development and progression of localized prostate cancer, metastatic prostate cancer, CRPC, and NEPC. Both primary and metastatic prostate cancers usually respond well to initial ADT, while treatment for mCRPC and NEPC patients is difficult. However, FOXA1 was found to be downregulated in NEPC and CRPC, indicating different mechanisms in different types of prostate cancers. In addition, the types of mutations in FOXA1 varied among early-stage prostate cancer, CRPC, and NEPC. Notably, the mutation rate of FOXA1 is higher in metastatic prostate cancer and NEPC than that in primary prostate cancer. Given the oncogenic activity of FOXA1 in both AR-independent and AR-dependent prostate cancers, FOXA1 could be a therapeutic target in prostate cancers, including mCRPC and NEPC.121 In addition, FOXA1 could be a biomarker for mCRPC and NEPC. Future efforts need to focus on the mechanisms underlying the mutation of FOXA1 in advanced prostate cancer.

Some significant barriers in the molecular mechanism should be considered. We usually thought that high-frequency mutations of FOXA1 were markers of metastatic prostate cancer or NEPC. However, a recent Chinese cohort study showed that FOXA1 was the most highly mutated gene (41%) in primary prostate cancer.21 Ethnic factors must be considered in the subsequent studies of FOXA1 as they may be involved in different molecular pathways. Moreover, it was reported that FOXA1 was observed in all types of prostate cancers, while FOXA2 was only detected in NEPC and advanced prostate cancer.25 In particular, FOXA2 may be a factor promoting androgen-independent prostate cancer.25 This result suggested that members of the FOXA family may play a crucial role in prostate cancer. Future studies should focus on the specific gene regulation between FOXA1 and tumor progression. Some drugs targeting gene mutations have been discovered. The mutation frequency of SPOP is even higher than that of FOXA1 in primary prostate cancer and advanced prostate cancer, and SPOP-mutated mCRPC appears to be highly sensitive to abiraterone.22,86,122 The PARP inhibitor olaparib has been used for prostate cancer patients with BRCA1 DNA repair-associated (BRCA1) and BRCA2 DNA repair-associated (BRCA2) mutations.123 Although there are currently no specific inhibitors against FOXA1, recent technologies have led to nucleotide-based inhibitors through antagonizing transcriptional programs by multiple pathways.124,125 In conclusion, we believe that FOXA1 could be a prognostic marker and therapeutic target in various stages of prostate cancer.

AUTHOR CONTRIBUTIONS

CL and JL conceived the structure of the manuscript and revised the manuscript. HYD, TRZ, and TY drafted the initial manuscript. HYD and LD made the figures and tables. All authors read and approved the final manuscript.

COMPETING INTERESTS

All authors declared no competing interests.