Lineage Plasticity and Treatment Resistance Highlight Need for Personalized mCRPC Strategies

Lineage plasticity and therapeutic resistance are reshaping the clinical landscape of metastatic castration-resistant prostate cancer (mCRPC), especially as more patients are exposed to early, intensive androgen signaling inhibition, according to Jue Wang, MD.

In the second part of our discussion, Wang highlighted how increasing use of novel androgen receptor pathway inhibitors, often initiated during the hormone-sensitive phase, may drive tumor phenotypic transformation, including neuroendocrine and sarcomatoid differentiation. In the first part of the interview, Wang explained how personalized treatment approaches could help fill gaps in mCRPC management.

Wang is a professor in the Department of Internal Medicine and a member of the Division of Hematology and Oncology at the University of Texas Southwestern Medical Center in Dallas.

OncLive: How has the treatment landscape for mCRPC evolved in recent years, and what lessons can we draw from real-world patient journeys about sequencing therapies and managing resistance?

Wang: Prostate cancer is [ripe] for targeted therapy. From the beginning, we use androgen deprivation therapy [ADT] because androgen is a key driver of proliferation. However, we’ve learned over the years that long-term suppression of androgen signaling, whether it be with GnRH antagonists/agonists, or next-generation agents, can push the cancer to change phenotype. This is called lineage plasticity.

When we first started treating prostate cancer, there weren’t many options. Medical oncologists specializing in prostate cancer are a relatively new group; and these patients were primarily treated by urologists. Back then, when we saw transformed tumors—like neuroendocrine tumors—we would say [these were] extremely rare.”

However, approximately 10 years ago, a pathology study [examined biopsies of] patients with mCRPC. They found that 15% to 17% of patients who had received novel androgen signaling inhibitors experienced some type of transformation, such as neuroendocrine transformation, aggressive variant prostate cancer, or other phenotypic changes, such as sarcomatoid transformation. Molecularly, these cases showed loss of tumor suppressor genes and other features.

[We] can see that with increasing and more intensive use of androgen signaling inhibitors, more of these transformed cancers [are] emerging. It’s similar to how excessive use of antibiotics leads to resistant strains [of bacteria]. In cancer, we now talk every week in a tumor board about patients with atypical prostate-specific antigen [PSA] patterns, with disproportionate disease burden without the PSA warning. These cases are radiographically progressing, even when PSA remains undetectable.

During the State of the Science Summit, you highlighted a pair of case studies. How could these types of examples help clinicians further understand lineage plasticity?

With these two clinical cases, I wanted to show that [although] intensive therapy helps control cancer, it may also create new, evolved cancer phenotypes. We need to pay attention to this. In the second case, I discussed a 54-year-old gentleman with high-risk, de novo metastatic hormone-sensitive prostate cancer. He received triplet therapy and achieved complete remission, both radiographically and by PSA. However, less than 1 year later, cervical lymph node [lesions] appeared. CT scans revealed additional lymphadenopathy. PSA remained undetectable, but a biopsy showed poorly differentiated disease. These transformations are increasing in frequency.

On one hand, the good news is we have treatments to control cancer. On the other hand, we must recognize how our treatments influence the trajectory and natural history of metastatic prostate cancer.

In 2025, the landscape of mCRPC has already changed. These patients have usually undergone multiple therapies by the time they are diagnosed [with mCRPC]. Androgen signaling inhibitors now have indications [in multiple prostate cancer settings]. These shifting treatment patterns impact disease evolution. Cancer is a moving target, largely due to treatment availability. We’re treating a different disease now.

What are some of the key factors contributing to treatment drop-off in older patients with mCRPC, and how can clinicians better support these patients to improve real-world outcomes?

mCRPC is a disease [typically seen in older patients], so we’re treating both cancer and geriatric syndromes. Bridging the gap between the advances from clinical trials and the sobering reality of real-world outcomes requires more than just talking about drugs and protocols.

We have to ask: Why do patients [outcomes] drop off between first-line and second-line treatment? There are many reasons. One is disease progression; if the second drug is mechanistically similar, it won’t be effective due to cross-resistance. Another is performance status decline. These patients have already experienced years of ADT and treatment intensification. Androgen signaling inhibitors squeeze out every drop of testosterone, and that impacts performance. Patients become more vulnerable [as they move through lines of treatment].

If we want to improve real-world outcomes, we must understand biology, heterogeneity, and resistance, [and we must] proactively support patients. We know treatment can make them weaker, so we need to orchestrate care like a symphony. Don’t just kill the cancer; support the host. Engage and empower patients. Help them understand the trade-offs so they can make informed decisions.

In addition to physician education, at UT Southwestern we advocate for including advanced nurse practitioners and physician assistants in cancer care. They’re now an important force in treatment planning, symptom and toxicity management, and survivorship navigation.

All these systems must work together. [Clinical practice] is not just about implementing clinical trial protocols. Guidelines give us a map, but good, biologically informed decision-making gives us a compass to improve patient outcomes.