Therapeutic Exploitation of Neuroendocrine Transdifferentiation Drivers in Prostate Cancer

Abstract

Neuroendocrine prostate cancer (NEPC), an aggressive and lethal subtype of prostate cancer (PCa), often arises as a resistance mechanism in patients undergoing hormone therapy for prostate adenocarcinoma. NEPC is associated with a significantly poor prognosis and shorter overall survival compared to conventional prostate adenocarcinoma due to its aggressive nature and limited response to standard of care therapies. This transdifferentiation, or lineage reprogramming, to NEPC is characterised by the loss of androgen receptor (AR) and prostate-specific antigen (PSA) expression, and the upregulation of neuroendocrine (NE) biomarkers such as neuron-specific enolase (NSE), chromogranin-A (CHGA), synaptophysin (SYP), and neural cell adhesion molecule 1 (NCAM1/CD56), which are critical for NEPC diagnosis. The loss of AR expression culminates in resistance to standard of care PCa therapies, such as androgen-deprivation therapy (ADT) which target the AR signalling axis. This review explores the drivers of NE transdifferentiation. Key genetic alterations, including those in the tumour suppressor genes RB1, TP53, and PTEN, and changes in epigenetic regulators, particularly involving EZH2 and cell-fate-determining transcription factors (TFs) such as SOX2, play significant roles in promoting NE transdifferentiation and facilitate the lineage switch from prostate adenocarcinoma to NEPC. The recent identification of several other key novel drivers of NE transdifferentiation, including MYCN, ASCL1, BRN2, ONECUT2, and FOXA2, further elucidates the complex regulatory networks and pathways involved in this process. We suggest that, given the multifactorial nature of NEPC, novel therapeutic strategies that combine multiple modalities are essential to overcome therapeutic resistance and improve patient outcomes.

Keywords: cancer; lineage plasticity; neuroendocrine; prostate; transdifferentiation.

Figure 1

Immunohistochemical (IHC) analysis of prostate adenocarcinoma and neuroendocrine prostate cancer (NEPC). Representative sections from a patient with prostate adenocarcinoma and a patient with NEPC were stained for haematoxylin and eosin (H&E) to assess histology, prostate-specific antigen (PSA) to detect luminal epithelial differentiation, and synaptophysin (SYP) to indicate neuroendocrine differentiation. Prostate adenocarcinoma tissue displays high PSA expression and an absence of SYP staining, characteristic of luminal epithelial differentiation. In contrast, NEPC tissue exhibits strong SYP expression with minimal or absent PSA staining, reflecting neuroendocrine features. Images taken at 20× magnification, with 40× magnification in the corner. Patient tumours were collected as donor tissues for the patient-derived xenograft study [23].

Figure 2

Key players in the neuroendocrine transdifferentiation process of prostate adenocarcinoma to neuroendocrine prostate cancer. Driving factors identified in the literature to facilitate the lineage plasticity and phenotypic reprogramming of prostate cancer cells, namely by the loss or downregulation of tumour suppressor genes, differentially expressed genomic and epigenetic drivers, dysregulated AR interactors and aberrations to other modulatory factors. Figure shows drivers altered state during the transdifferentiation process is upregulated/overexpressed, downregulated/loss, dysregulated, hypomethylated, hypermethylated.

Figure 3

Protein–protein interaction network of driver NE transdifferentiation genes. Confirmatory analysis of the mentioned genes shows interconnectedness to drive NE transdifferentiation. Network analysis using STRING, with the interaction score set to a confidence threshold of 0.700. Core proteins such as TP53, RB1, MYCN, and EZH2 are central to the network with greater confidence scores, highlighting their significance in NEPC transdifferentiation. Line thickness indicates the strength of the data supporting the interactions. Gene names (not encoded protein names) used: ARID1A and SMARCA4 used as main SWI/SNF signalling genes. The confidence scores are computed by STRING using evidence channels (textmining, experiments, databases, co-expression, neighbourhood, gene fusion, and co-occurrences). The colours correspond to different “evidence channels” in STRING, each representing a distinct type of data or method used to infer protein associations: multiple colours on a single line suggest multiple types of evidence for that interaction [154].