Prostate cancer (PCa) is the second most common cancer in men worldwide and a leading cause of cancer-related deaths. In the United States alone, approximately 260,000 new cases were projected for 2022, representing 27% of all new cancer diagnoses in men.
PCa is characterized by extreme heterogeneity -- meaning that even within a single tumor, cancer cells differ dramatically from one another in their genetic makeup, behavior, and response to treatment. Traditional laboratory methods like immunohistochemistry struggle to capture this complexity.
Single-cell omics, especially single-cell RNA sequencing (scRNA-seq), overcomes this limitation by measuring gene activity in individual cells rather than averaging signals across thousands of cells. This reveals the unique molecular fingerprint of each cell type present in the tumor and its surrounding environment.
This review synthesizes the growing body of research applying single-cell omics to prostate cancer, covering tumor cells, cancer-associated fibroblasts, the immune environment, spatial gene expression patterns, and drug resistance mechanisms.
The cells lining the prostate gland -- called tumor-associated epithelial cells (TAECs) -- are the most abundant cell type in PCa. scRNA-seq has revealed that these cells are far more diverse than previously appreciated, with distinct subtypes including luminal, basal, and newly discovered cell types like hillock cells and club cells.
Some epithelial cell populations show high androgen signaling activity, which is significant because androgen is the primary driver of prostate cancer growth. Cells with high androgen signaling are associated with specific gene markers (LTFhigh and NKX3-1high) and are linked to disease progression and eventual resistance to hormone therapy.
A key study identified four distinct tumor cell transcriptional subtypes in untreated patients: EMT-like (involved in invasion), luminal A-like, luminal B/C-like, and basal-like. Each subtype corresponds to distinct tumor clonal patterns and has different implications for prognosis -- highlighting why one-size-fits-all treatment often fails.
The heterogeneity of TAECs is clinically important because it means tumors contain mixtures of cells at different stages of aggressiveness, with different sensitivities to androgen-blocking therapies. Understanding which cell states predominate in a given patient could guide more targeted treatment strategies.
Cancer-associated fibroblasts (CAFs) are connective tissue cells that make up over 50% of the cells in a solid tumor's surrounding environment. They are not passive bystanders -- they actively communicate with cancer cells through chemical signals and physical interactions to promote tumor growth and spread.
Single-cell sequencing has classified prostate CAFs into six distinct subgroups (PCAFs), each with different functional roles. Some subgroups recruit immune cells to the tumor site through a signaling molecule called CCL2, while others produce signals that drive cancer cell migration and invasion through CXCL8 cytokines.
Crucially, blocking CCL2 and CXCL8 signals in laboratory experiments was shown to eliminate CAF-driven cancer cell migration -- suggesting these specific CAF subtypes could be targeted therapeutically to limit cancer spread.
CAFs also promote cancer aggressiveness through epithelial-mesenchymal transition (EMT) -- a process where cancer cells acquire invasive properties -- and can induce resistance to standard treatments. Single-cell omics has made it possible to identify which CAF subtypes are responsible for these specific effects, opening new avenues for intervention.
The immune system can recognize and attack cancer cells, and immunotherapy has revolutionized treatment of many cancers by unleashing this immune response. However, prostate cancer is classified as a 'cold tumor' -- one with relatively few immune cells infiltrating the tumor and an overall immunosuppressive environment that limits natural immune attack.
This immune coldness is one reason why CRPC often fails to respond well to immunotherapy, including immune checkpoint blockade drugs like nivolumab and ipilimumab. The complex regulation of immune cells within the prostate tumor immune microenvironment (TIME) sometimes paradoxically shows that more immune cell infiltration is associated with worse outcomes rather than better ones.
Single-cell analysis of over 30,000 cells from PCa samples revealed that cancer cells can alter the gene activity of T cells within the tumor -- for example, by increasing KLK3 (PSA) levels inside T cells. This may be one mechanism by which prostate tumors reprogram immune cells to be less effective against cancer.
A distinct population of immunosuppressive macrophages in prostate tumors has been identified by single-cell sequencing, regulated by a receptor called CSF1R (colony-stimulating factor 1 receptor). These macrophages suppress immune activity and promote tumor survival, and CSF1R emerges as a potential therapeutic target to convert the prostate tumor from cold to immunologically active.
Beyond which cell types are present, spatial transcriptomics addresses where cells are located within tumor tissue -- a dimension that standard sequencing destroys by breaking tissue into individual cells. Early spatial transcriptomics studies measured gene expression across at least 6,000 regions in prostate tumor tissue and found significant differences between adjacent areas of the same tumor.
The prostate is anatomically divided into three zones: the peripheral zone (PZ), transition zone (TZ), and central zone (CZ). These zones differ in their cancer biology. PZ tumors have higher malignant potential, higher rates of spread outside the prostate capsule, and worse clinical outcomes. TZ tumors tend to be less aggressive. CZ tumors are rare but carry the worst prognosis.
Spatial sequencing confirmed that genes associated with invasive potential -- including Ki-67 (cell proliferation), bcl-2 (cancer cell survival), MMP-2, and MMP-9 (invasion enzymes) -- are significantly more highly expressed in PZ tumors compared to TZ tumors, providing a molecular explanation for their different behaviors.
At the microscopic level, spatial studies have demonstrated that individual cancer cells at different physical locations within a single tumor can carry entirely different mutation patterns -- meaning a tumor is not genetically uniform even within a few millimeters. This spatial genetic diversity helps explain treatment failures when a drug may eliminate some cells but miss others nearby.
Cellular senescence is a state in which cells stop dividing permanently in response to stress. It was originally understood as a protective mechanism that limits tumor growth by halting abnormal cell proliferation. However, senescent cells remain metabolically active and secrete a variety of molecules that can paradoxically promote inflammation and tumor progression.
In prostate cancer, many anticancer treatments -- including chemotherapy, radiation, and androgen receptor-targeting drugs -- can induce therapy-induced senescence (TIS) in tumor cells. The most common signaling pathways involved include p53/p21 and p16/pRb pathways. While this initially suppresses tumor growth, the resulting senescent cells can secrete factors that stimulate remaining cancer cells.
This collection of secreted molecules is called the senescence-associated secretory phenotype (SASP). The SASP can create an environment that encourages surviving cancer cells to become more aggressive, resist future treatment, and potentially relapse. This makes therapy-induced senescence both a desired outcome and a potential liability.
Single-cell omics has enabled the mapping of senescent cell populations within PCa tumors, revealing that senescent cells show significant heterogeneity and are positively correlated with immune cell infiltration and PD-L1 expression -- a key immune checkpoint protein. This suggests senescent cells actively influence the immune landscape and could be targeted to improve immunotherapy outcomes.
Androgen deprivation therapy (ADT) has been the cornerstone of prostate cancer treatment since the 1940s. It works by reducing androgen (primarily testosterone and dihydrotestosterone) levels or blocking the androgen receptor (AR), starving cancer cells of their primary growth signal.
Despite initial success, most prostate cancers eventually develop resistance and progress to castration-resistant prostate cancer (CRPC). Single-cell sequencing of over 20,000 prostate epithelial cells revealed that CRPC-like cells already exist in the early stages of prostate cancer, even before any ADT treatment is given. This means resistance is not purely acquired through treatment -- a pre-existing resistant cell population may survive and expand under ADT pressure.
Studies using scATAC-seq (single-cell assay for transposase-accessible chromatin) alongside scRNA-seq identified pre-existing cellular states associated with cancer recurrence in patients treated with enzalutamide, a second-generation AR blocker. These analyses reveal how cancer cells alter the accessibility of their DNA in response to treatment -- providing a window into how resistance emerges at the epigenetic level.
This single-cell approach to studying drug resistance is clinically significant because it enables identification of the specific cancer cell subtypes that survive treatment, which can then be targeted with complementary therapies to prevent relapse and improve long-term outcomes for prostate cancer patients.
The future of prostate cancer research lies in single-cell multi-omics -- combining gene expression data (scRNA-seq) with other molecular measurements such as whole-exome sequencing, chromatin accessibility (scATAC), proteomics, metabolomics, and epigenomics all from the same individual cells. This will provide an unprecedented multidimensional portrait of each tumor.
A major frontier is integrating single-cell data with spatial transcriptomics to understand not just which cells are present and what genes they express, but where those cells are located relative to each other within the tumor tissue. This spatiotemporal view will reveal how cancer cells and immune cells interact in physical space and how these interactions change over the course of treatment.
New sample preparation advances mean that even formalin-fixed paraffin-embedded (FFPE) archival tissue blocks -- the standard format for stored clinical biopsy specimens worldwide -- can now be used for single-cell and spatial transcriptomics. This opens the possibility of studying patient samples from historical clinical trials that could not previously be analyzed at the single-cell level.
Together, these technologies will enable more precise diagnosis, better risk stratification, identification of actual drug targets, and ultimately the development of combination treatment regimens tailored to the specific cellular composition of an individual patient's tumor -- realizing the promise of true precision oncology for prostate cancer.