Prostate cancer is the second most frequent cancer in men, with increasing prevalence due to an ageing population. Advanced prostate cancer is diagnosed in up to 20% of patients, and, therefore, it is important to understand evolving mechanisms of progression. Significant morbidity and mortality can occur in advanced prostate cancer where treatment options are intrinsically related to lipid metabolism. Dysfunctional lipid metabolism has long been known to have a relationship to prostate cancer development. Research has established the pathways leading to prostate cancer development, including dysregulated lipid metabolism-associated de novo lipogenesis through steroid hormone biogenesis and β-oxidation of fatty acids. These pathways, in relation to treatment, have formed potential novel targets for management of advanced prostate cancer via androgen deprivation.
1. Prostate Cancer and Dysregulated Lipid Metabolism
The relationship between lipid metabolism and cancer was first observed by Medes et al.
[1] in 1953. Cancer tissue was found to overexpress enzymes to generate FAs and phospholipids via de novo lipogenesis in conjunction with environmental uptake of lipids. De novo lipogenesis, in turn, supports the excess energy requirements for growth and proliferation which has become a notable hallmark of cancer
[2]. Cells can utilise FAs for energy generation via β-oxidation to generate ATP, the principal energy molecule. Activation of the de novo lipogenesis pathway affects all levels of lipid enzyme regulation, occurring downstream to known oncogenic abnormalities such as activation of akt, loss of PTEN, mutation or loss of p53 or BRCA1 and steroid hormone activation
[3]. Additionally, exogenous lipids from circulation and lipolysis or stored lipids in adipocytes and intracellular lipid droplets can also be utilised
[4].
Interestingly, PCa cells differ from various other cancers as FAs are the predominant energy substrate as opposed to glucose
[5]. PCa cells undergo dysregulated lipid metabolism comprising increased de novo lipogenesis in the form of steroid hormone biosynthesis and β-oxidation of FAs for energy generation, membrane synthesis and cell division. Furthermore, FAs are stored in lipid droplets or converted to complex phospholipids as key components to cell membranes
[3]. Remarkably, epidemiologic data suggests obesity is a significant risk factor for aggressive forms of PCa, further emphasising the role of dysregulated lipid metabolism
[6]. Laboratory studies have demonstrated FA synthase (FAS) having similar properties to oncogenes in PCa mouse models and FAS inhibitors having converse effects, limiting PCa growth in similar environments
[7][8].
2. Lipid Metabolism and Androgens in Prostate Cancer
Huggins and Hodges
[9] first noted in 1941 the improvement of patients with metastatic PCa when chemically castrated with oestrogens. This led to the understanding that PCa is exquisitely influenced by androgenic activity with inhibition occurring with elimination of androgens. Androgen receptor (AR) activation is a key player in PCa growth and stimulation via multiple metabolic pathways, and its link to lipid metabolism has been observed in advanced PCa, whereby accumulation of lipid droplets in the cytoplasm occurs via AR-associated increased synthesis of cholesterol and FAs
[10][11]. Furthermore, an AR antagonist reverses the effects of lipogenesis, which is not seen in AR-negative PCa cells
[11]. Subsequently androgens have been found to influence prostate cell lipid profile through synthesis, binding, uptake, metabolism and transport of lipids
[3].
The most characterised mechanism for androgen involvement in stimulation of de novo lipogenesis is via indirect regulation of protein expression, a transcription factor family named sterol regulatory element-binding protein (SREBP). SREBP plays an important part in increasing lipid and cholesterol metabolism and, in turn, aids androgen synthesis
[12]. Specifically, SREBPs are comprised of SREBP1a and SREBP1c, and with two isoforms are master regulators of lipid homeostasis due to regulation of enzymes required for lipid synthesis and uptake. Reduction of intracellular sterol levels causes SREBP cleavage activating protein (SCAP)-SREBP complex translocation into the Golgi and is further cleaved by proteases (site-1 and site-2). This, in turn, causes SREBP to translocate to the nucleus, binding to sterol-response elements (SRE) inducing transcription of the key enzymes to de novo lipogenesis, including FAS, 3-hydroxy-4-methylglutarul coenzyme A reductase (HMG CoA-R) and LDL receptor (LDLR)
[13]. Furthermore, several studies reviewed by Wu et al.
[14] have emphasised the importance of FAs as a dominant energy source in PCa, finding increased expression of enzymes DBP and AMACR, noted earlier to be important in β-oxidation for ATP generation
[15]. Given the exquisite relationship of androgens for stimulation and growth of PCa, the mainstay of treatment of advanced PCa is ADT, which inhibits testicular testosterone production either medically or surgically to reduce circulating levels of androgen
[16].
3. Androgen Deprivation Therapy
Despite the availabilities of newer targeting agents, such as Enzalutamide or Abiraterone, classic ADT (hormonal therapy) has widespread use for local advanced to metastatic hormone sensitive PCa as neoadjuvant or adjuvant therapy with radiotherapy. Furthermore, addition of ADT to other systemic agents, such as AR targeted therapy, have recently been noted to improve overall survival in a systematic review when comparing systemic treatments for metastatic castration-sensitive (hormone sensitive) PCa
[17]. Wang et al.
[17] found Abiraterone acetate (hazard ratio (HR), 0.61; 95% confidence interval (CI), 0.54–0.70) and Apalutemide (HR, 0.67; 95% CI, 0.51–0.89) may have the most improvement to overall survival when added to ADT. A historic form of ADT, which remains a treatment option, is bilateral orchidectomy. Though superseded by more commonly used non-surgical treatments, there remain benefits with surgical castration including less cost and follow-up with potentially fewer side effects. Furthermore, Weiner et al.
[18] found that survival rates are comparable between surgical and medical castration.
ADT targets various portions of the hypothalamic–pituitary–gonadal axis and are broadly classed into antiandrogens, LHRH agonist/antagonists (i.e., gonadotrophin releasing hormone (GhRH)) and androgen pathway inhibitors
[19]. Antiandrogens block the AR to reduce testosterone cellular signalling whilst androgen pathway inhibitors target along the androgen pathway to reduce AR signalling or inhibit testosterone synthesis. LHRH agonists/antagonists target the LHRH receptor in the anterior pituitary gland. LHRH agonists stimulate the receptor leading to a temporary LH and testosterone surge, subsequently downregulating the receptor causing reduction in the LH and testicular production. Conversely, LHRH antagonists are competitive reversible agents blocking the LHRH receptors, in turn reducing LH release and therefore dropping testosterone production which avoids the initial transient rise. The first systematic review and meta-analysis of prospective studies in men undergoing ADT and its effects on body composition in PCa demonstrated significant increase in body fat on average of 7.7% (95% CI, 4.3–11.2,
p < 0.0001), body weight (2.1%,
p < 0.001) and BMI (2.2%,
p < 0.001) with reduction in lean body mass −2.8% (95% CI, −3.6, −2.0,
p < 0.0001)
[20].
4. Cellular Mechanisms of Prostate Cancer and Dysfunctional Lipid Metabolism
Several mechanisms for dysregulated lipid metabolism in PCa via genetic abnormalities have been hypothesised, and this section will mainly focus on two novel theories. The first focuses on amplification and overexpression of pyruvate dehydrogenase complex (PDC), which is a gatekeeper for conversion of pyruvate to acetyl-CoA and subsequent entry into the mitochondrial TCA cycle
[21]. Another focuses on co-deletion of promyelocytic leukemia (PML) and phosphatase and tensin homolog (PTEN) on PTEN-null PCa phenotypes
[22]. PTEN is a commonly mutated or lost tumour-suppressor gene in many cancers
[23] with partial loss in up to 70% of localised PCa
[24] and complete loss linked to metastatic castration-resistant PCa
[25]. The PTEN-null transgenic PCa mouse model was utilised in both these studies and emulates high grade intraepithelial prostate tumours at an early age and invasive PCa at late age.
Chen et al.
[21] emphasise the association of increased mitochondrial metabolism and cancer pathogenesis and progression through the investigation of PDC. A major component of PDC is PDHA1, which is activated when dephosphorylated by pyruvate dehydrogenase phosphatase (PDP). Conversely, PDC is deactivated when phosphorylated by pyruvate dehydrogenase kinases (Pdks). The authors first establish that subunits of the PDC are amplified and overexpressed in the PTEN-null transgenic mouse model similarly to clinical PCa. In vivo analysis of both mouse and human PCa models demonstrated hampering of PCa progression via suppression of lipid biosynthesis when PDHA1 is inactivated
[21]. Furthermore, the authors postulate that the PDC must be functional in both mitochondria and cancer cell nuclei for sufficient support of lipid biosynthesis. In the nucleus, SREBP transcription factor, previously described as important in de novo lipogenesis, has reduced histone acetylation at regulatory regions. However, at the mitochondrial level lipid biosynthesis suppression was noted to be due to a reduction in citrate production, which is important for ATP generation via the TCA cycle
[21]. Finally, the authors postulate that PDC, specifically PDHA1, may be a potential therapeutic target for prevention of PCa development. A recent study similarly suggested the therapeutic potential for targeting PDC components and comparatively also found a positive correlation between PDC component (PDHA1, PDP1 and PDK) expression and AR expression
[26].
Alternatively, Chen et al.
[22] aimed to explore the impact of PML and PTEN co-deletion in metastatic PCa in PTEN-null transgenic mouse of PCa. Amplification of lipid metabolism was identified on coordinated loss of these two tumour suppressor genes. Interestingly, inhibition of SREBP by fatostatin can block metastasis, which can be replicated by a high fat diet in PTEN-null mice without PML loss
[22]. Therefore, the authors note that PML deletion leads to amplification of MAPK signalling and subsequent aberrant lipid metabolism. Current evidence for statin use in PCa has no consensus guidelines; however, a recent systematic review and meta-analysis suggests statins may have a unique role in the reduction of biochemical recurrence of PCa post definitive treatment
[27].
Several other genetic drivers of aggressive PCa include p53, retinoblastoma (RB1) and Myc, which are well established mechanisms of cancer cell survival in nutrient-poor environments. P53, RB1 and Myc may also be associated with dysfunctional lipid metabolism. Tumour suppressor gene (TSG) p53 regulates cellular metabolism and is a transcription factor controlling protein expression in cell cycle arrest, DNA repair, apoptosis and senescence
[28]. Mutation in p53 has been found to increase gain of function activities (p53
R273H and p53
R280K) and encourage mutant binding to SREBPs, which of course leads to aberrant lipid metabolism, amplifying tumour progression thereby increasing fatty acid synthesis
[28]. Conversely, TSG protein RB1 (a critical transcriptional corepressor in prevention of tumour development and progression) loss of function (LOF) has been associated with progression to castration-resistant PCa
[29]. Furthermore, LOF of RB1 is associated with alteration in multiple metabolic pathways including lipid, amino acid and peptide metabolism; however, the specifics of lipid dysregulation are unknown
[29]. More recently, Myc has also been associated with key FA synthesis genes including ACLY, ACC1 and FAS in PCa
[30]. Hi-Myc transgenic mice with prostate-specific overexpression of Myc demonstrated increased circulating levels of total free FAs
[30]. As previously noted, FA metabolism is intricately related to PCa development for ATP generation through β-oxidation supporting the excess energy requirements for growth and proliferation of cancer cells
[2]. In light of these genetic alterations associated with more aggressive disease variants, recent studies have attempted to identify specific molecular features of aggressive PCa.
Aggressive variant PCa (AVPC) refers to AR-independent forms of PCa
[31]. Clinically, AVPC is characterised by rapid disease progression including hormone refractory disease and visceral metastases. PTEN, p53, RB1 and Myc have been found to have more frequent alterations in AVPC and is characterised by a combination of alterations similar to small-cell PCa and therefore have direct clinical relevance with platinum-based treatment
[31][32]. Hence, treatment of AVPC as similar to small-cell PCa has demonstrated improvement in progression-free survival (65.4% first-line and 33.8% second-line) and median overall survival (16 months (95% CI, 13.6–19.0 months)) after first line carboplatin and docetaxel and second line etoposide and cisplatin, respectively
[33].