Inside Prostate Cancer: Biology, Drivers, and the Roadblocks in Drug Discovery

Although many prostate tumors are detected while still localized, prostate cancer remains a major focus of oncology drug discovery because of its high incidence, clinical heterogeneity, and the complexity of treating the disease. In the United States, prostate cancer remains one of the most diagnosed cancers among men, with over 330,000 new cases and more than 36,000 deaths projected annually, despite significant advances in screening and treatment.¹

In response, prostate cancer drug development is undergoing rapid transformation. While androgen receptor (AR) pathway inhibitors have redefined the standard of care, the pipeline now includes radioligand therapies, antibody-drug conjugates (ADCs), T cell engagers, PARP inhibitors, AR degraders, and other precision medicines. Yet despite these advances, most patients with advanced disease ultimately develop treatment resistance. Understanding the biology that drives disease progression and resistance has therefore become central to selecting therapeutic targets, designing predictive preclinical models, and improving clinical translation.

At its core, prostate cancer is not a single biological entity. It is a spectrum of diseases shaped by hormonal signaling, genomic instability, tumor suppressor loss, microenvironmental cues, immune interactions, and selective pressure from treatment. These features influence how prostate tumors arise, how they progress, and how they adapt when exposed to therapies.

For researchers and drug developers, a deep understanding of prostate cancer biology is critical to building predictive models, identifying clinically relevant targets, and designing translational strategies that truly capture the complexity of this disease.

What is Prostate Cancer?

Prostate cancer develops in the prostate gland, a small reproductive organ located below the bladder. Approximately 95% of prostate cancers are acinar adenocarcinomas arising from glandular epithelial cells. These tumors most commonly develop in the peripheral zone of the prostate, although a subset can arise in the transition or other regions of the gland.

Prostate cancer is primarily caused by genetic changes in prostate cells, with age, family history, ethnicity, and lifestyle factors influencing risk. These mutations can cause cells to grow uncontrollably, avoid normal cell death, and accumulate to form tumors. Key genetic changes include activation of oncogenes, inactivation of tumor suppressor genes, and defects in DNA repair genes, all of which disrupt normal cell regulation and can lead to cancerous growths.

Localized disease is confined to the prostate and may be managed with active surveillance, surgery, radiation, or focal approaches depending on risk. Locally advanced disease extends beyond the prostate capsule or involves nearby tissues. Metastatic prostate cancer most commonly involves bone and lymph nodes, although lung and liver metastases also occur, particularly in advanced disease. This progression from localized to metastatic disease reflects underlying changes in tumor biology, including enhanced survival signaling, invasive behavior, immune evasion, and adaptation to new tissue environments.

How Prostate Cancer Progresses

Early-stage prostate cancer often depends strongly on AR signaling. Androgens such as testosterone and dihydrotestosterone bind the AR, which then regulates genes involved in prostate cell growth, differentiation, and survival. This dependency explains why androgen deprivation therapy can be effective, particularly in advanced or metastatic castration-sensitive prostate cancer (mCRPC).

Over time, however, many tumors acquire mechanisms that allow continued growth despite reduced androgen levels. This transition defines castration-sensitive prostate cancer (CRPC) but does not mean the tumor no longer involves AR biology; in many cases, the pathway remains active through receptor amplification, mutation, splice variants, altered co-regulators, or intratumoral androgen synthesis. In other cases, tumors become less dependent on AR signaling and rely more heavily on bypass pathways, lineage plasticity, or neuroendocrine-like features.

A subset of tumors undergo lineage switching to treatment-emergent neuroendocrine prostate cancer (NEPC), an aggressive variant characterized by reduced AR dependence, rapid disease progression, and poor clinical outcomes.

As tumors progress, they often become more invasive, more heterogeneous, and less responsive to standard therapies. The biology of metastatic prostate cancer is particularly challenging, requiring tumor cells to survive in circulation, establish themselves in distant tissues, and adapt to organ-specific microenvironments. Among these microenvironments, bone is the predominant site of metastasis, occurring in approximately 80–90% of men with mCRPC.

Key Molecular Drivers of Prostate Cancer

Among the molecular drivers of prostate cancer, AR signaling remains the central pathway. The AR acts as a transcription factor, controlling expression of genes that support prostate tumor growth and survival. Therapies that suppress androgen production or block AR activity have transformed care, yet resistance frequently emerges because tumor cells find ways to reactivate or bypass this axis.

The PI3K/AKT/PTEN pathway is another major driver of prostate cancer progression and treatment resistance. PTEN is a tumor suppressor that normally restrains PI3K signaling. When PTEN is lost or PI3K/AKT signaling becomes activated, tumor cells may gain growth and survival advantages. Importantly, AR and PI3K/AKT pathways can interact, creating compensatory signaling networks that complicate therapy. This crosstalk is one reason combination strategies targeting multiple pathways are an active area of research.

DNA damage repair pathway alterations also shape prostate cancer biology. Defects in homologous recombination repair, including changes involving BRCA1, BRCA2, ATM, and related genes, can influence tumor behavior and create therapeutic vulnerabilities. PARP inhibitors are an example of how molecular understanding can translate into targeted treatment, particularly for tumors with specific DNA repair deficiencies.

Prostate-specific membrane antigen (PSMA) is highly expressed in many advanced prostate cancers and has emerged as an important target for both imaging and therapy. However, PSMA expression can be heterogeneous and may decrease under anti-androgen therapeutic pressure or with lineage plasticity, which can limit its utility in some disease settings. As a result, additional targets such as STEAP1 and B7-H3 are emerging as promising alternatives across multiple modalities, including radioligand therapies, ADCs, T cell engagers, and other targeted approaches. Together, these strategies are reshaping treatment development and creating new opportunities for biomarker-driven drug discovery.

Loss of key tumor suppressors such as TP53, RB1, and PTEN can drive more rapid disease progression and promote lineage plasticity. This phenomenon enables tumor cells to alter their identity in response to therapeutic pressure, often shifting toward states that are less dependent on AR signaling. Such adaptability helps explain why some prostate cancers become more aggressive and resistant following prolonged AR pathway inhibition.

Tumor Heterogeneity in Prostate Cancer

Prostate cancer heterogeneity is a defining challenge of the disease. At the interpatient level, tumors can vary widely across individuals in terms of grade, genomic alterations, immune context, patterns of metastasis, and responsiveness to therapy. At the same time, intrapatient heterogeneity means that multiple distinct tumor clones may exist within a single patient, even within the same primary tumor or metastatic site.

This heterogeneity exists across both space and time. Spatial heterogeneity arises from differences between regions of the primary tumor or among distinct metastatic lesions. Temporal heterogeneity reflects the dynamic changes that occur as the disease progresses and as therapeutic pressure selects for resistant clones. Consequently, a single biopsy at one timepoint may not fully capture the evolving biology driving disease progression months or years later.

For biomarker development, heterogeneity creates both opportunities and challenges. Molecular biomarkers can help identify patients most likely to respond to targeted therapies, but biomarkers must be robust enough to account for clonal diversity and changing disease states. Liquid biopsy, longitudinal sampling, spatial profiling, and multiomic approaches are increasingly important tools for understanding how prostate tumors evolve and adapt over time.

Major Challenges in Prostate Cancer Research

One major challenge in prostate cancer research is modeling disease heterogeneity, complexity, and drug resistance. Traditional cell lines have contributed enormously to the field, but they cannot capture every aspect of tumor heterogeneity, stromal interaction, immune biology, or metastatic colonization.

In prostate cancer, model establishment has been particularly difficult because many patient tumors display poor tumorigenic properties, which has historically limited the number of available preclinical models relative to other top five cancer types. As a result, the overall prostate cancer model landscape remains comparatively narrow, making access to well-characterized and clinically relevant platforms especially important.

Capturing treatment resistance remains a persistent challenge. Resistance can emerge through multiple mechanisms, including AR reactivation, activation of bypass pathways, adaptations in DNA repair, epigenetic reprogramming, immune evasion, and lineage switching. Prostate tumors are also generally considered immunologically cold, often exhibiting limited T cell infiltration, relatively low neoantigen burden, and an immunosuppressive microenvironment shaped in part by myeloid-driven suppression and TGF-β-associated exclusion, all of which have contributed to the relatively modest success of immune checkpoint inhibitors in unselected patient populations. Because these processes often occur in parallel, a therapy that shows promise in one model may not translate effectively across the full spectrum of patient disease.

Metastasis, especially bone metastasis, is also difficult to study. The bone microenvironment is complex, comprising osteoblasts, osteoclasts, immune cells, extracellular matrix components, and a range of growth factors that can promote tumor survival and influence therapeutic response. Recreating these interactions in preclinical models is technically demanding, yet critical for advancing our understanding of advanced prostate cancer biology.  No single model can capture every dimension of metastasis, which is why researchers often use complementary systems, with cell line-based models supporting experimental metastasis studies and GEMMs providing valuable insight into tumor initiation, local invasion, and disease progression.

Finally, translating preclinical findings into clinical relevance requires careful alignment between mechanism, model, biomarker, and patient population. A target may be biologically compelling, but successful drug development depends on identifying the disease context in which that target truly matters.

Why Biology Matters for Drug Discovery

Biology-informed drug discovery starts with a clear understanding of disease mechanism, which for prostate cancer means asking whether a proposed therapy targets AR prostate cancer biology, DNA repair deficiency, PI3K/AKT pathway dependence, lineage plasticity, immune suppression, tumor-microenvironment interaction, or another actionable feature. This approach is a core part of target validation and must also be considered in the context of therapeutic modality, since the most relevant biology, assays, and model systems may differ for small molecules, radioligand therapies, ADCs, T cell engagers, and other emerging approaches.

For many new agents, resistance biology is also an important consideration early in development. Together, these factors should guide model selection, assay design, biomarker strategy, and translational planning. Linking disease mechanisms to model selection is critical. Models used to study AR signaling should faithfully preserve key aspects of AR biology. Those designed to investigate bone metastasis must incorporate features of the bone microenvironment. Similarly, models intended for CRPC research should capture clinically relevant resistance states observed in patients.

Translational strategies improve when they integrate molecular profiling, functional assays, pharmacodynamic markers, and clinically relevant endpoints. This integrated approach helps researchers move beyond whether a candidate works in a model toward understanding why it works, where it is most likely to work, and how resistance may emerge.

Prostate Cancer Models and Translational Insights

Comprehensive preclinical prostate cancer research relies on integrating multiple model systems to capture the biological complexity of the disease across stages, from hormone sensitive to mCRPC. Together, these models can reflect tumor heterogeneity, treatment history, and clinically relevant biology, while supporting both mechanistic studies and translational research.

Such platforms are often characterized using standardofcare therapies and supported by multiomic profiling approaches that include genomic, transcriptomic, proteomic, and targetspecific analyses (e.g., PSMA and AR signaling). These capabilities enable biomarkerdriven model selection, investigation of resistance mechanisms, and evaluation of diverse therapeutic modalities such as AR pathway inhibitors, PARP inhibitors, radioligand therapies, ADCs, and other targeted approaches across both in vitro and in vivo settings.

Conclusion

For oncology researchers and drug developers, biology is not background information; it is the foundation for better target selection, better models, and better translational decisions. A deep understanding of tumor heterogeneity, clonal evolution, AR signaling, and interactions with the tumor microenvironment is essential for identifying which pathways are truly driving disease progression and therapeutic resistance.

In prostate cancer specifically, factors such as lineage plasticity, neuroendocrine differentiation, immune contexture, and stromal influences all shape how tumors respond to treatment. Ignoring these biological nuances risks oversimplifying the disease and pursuing targets or compounds that may perform well in artificial systems but fail in patients.

The next step, therefore, is to align preclinical modeling strategies with this biological complexity. The next blog in this series will discuss how this biological complexity is characterized through key biomarkers and genetic alterations.

Reference

¹American Cancer Society. “Key Statistics for Prostate Cancer.” Prostate Cancer Facts. Accessed June 2026. https://www.cancer.org/cancer/types/prostate-cancer/about/key-statistics.html (cancer.org in Bing).