Biomarkers play an integral role in prognostication, treatment selection, and response monitoring in metastatic castration-resistant prostate cancer (mCRPC). Andrew J. Armstrong, MD, MSc, discusses emerging biomarker data presented at the recent 2026 ASCO Annual Meeting and how these data could be used to help personalize treatment for patients with mCRPC.

Following these presentations, featured expert Andrew J. Armstrong, MD, MSc, was interviewed by Conference Reporter Editor-in-Chief Tom Iarocci, MD. Clinical perspectives from Dr Armstrong on these findings are presented here.

Biomarkers serve many purposes in the management of prostate cancer. Prognostic biomarkers can tell us about survival and natural history, predictive biomarkers can help personalize treatment for patients, and response biomarkers can help determine whether a patient is benefiting from treatment.

If we zero in on the information we can get from circulating tumor DNA (ctDNA), it is pretty vast. For prognosis, the more ctDNA a patient has, the worse they do. And undetectable ctDNA is one of the strongest prognostic factors. At ASCO 2026, Susan Halabi, PhD, FASCO, presented the phase 3 Alliance A031201 study, which validated a ctDNA-based prognostic model for radiographic progression-free survival (rPFS) in patients with chemotherapy-naive mCRPC who were treated with enzalutamide with or without abiraterone acetate and prednisone (abstract 5018). For this analysis, ctDNA was used to assess tumor content and genetic alterations. These were combined with clinical variables such as pattern spread and symptoms, as well as various biomarkers such as hemoglobin and alkaline phosphatase. This resulted in a clinical genetic model that was strongly prognostic for ARPI-associated rPFS.

ctDNA can also help predict treatment response, particularly for patients with HRR mutations. For example, if we find a BRCA2 mutation, this can predict a survival benefit with olaparib or one of the ARPI/PARP inhibitor combinations. At ASCO 2026, Lorena Incorvaia, MD, PhD, presented results from a longitudinal cohort study that found that the type of BRCA2 mutation and location (mutation vs indel/deletion) may be critical determinants of survival and treatment response in metastatic prostate cancer (abstract 10503). Single nucleotide polymorphisms conveyed a better prognosis than indels, including with PARP inhibitors, which may be more susceptible to reversion resistance mutations.

Other HRR mutations are predictive as well. However, not all HRR mutations carry the same treatment response. In fact, a US Food and Drug Administration (FDA) meta-analysis of ARPI/PARP inhibitor combinations showed great benefit for BRCA2 carriers; modest benefit for CDK12, PALB2, and BRCA1 carriers; and essentially zero benefit for CHEK2 and ATM carriers in mCRPC (or androgen pathway modulator–resistant prostate cancer per the new Prostate Cancer Working Group 4 [PCWG4] terminology).

And then we have surprising new results from the phase 3 TALAPRO-3 study, which were discussed by Neeraj Agarwal, MD, FASCO, at ASCO 2026 (abstract LBA5007). This study showed improved rPFS outcomes in patients with metastatic hormone-sensitive prostate cancer (or androgen pathway modulator–sensitive prostate cancer per PCWG4 terminology) with HRR mutations. Patients with BRCA2 mutations had the greatest benefit from talazoparib plus enzalutamide, but patients with non–BRCA-mutated, HRR-mutated disease also benefited. Interestingly, if we look at the patients with ATM and CDK12 mutations in this study, they also derived clear benefit from talazoparib-plus-enzalutamide treatment in the earlier setting. While overall survival is immature, this suggests that patients with HRR+ metastatic hormone-sensitive/androgen pathway modulator–sensitive prostate cancer benefit from ADT and ARPI/PARP inhibitor triplet therapy.

The degree of prostate-specific membrane antigen (PSMA) avidity on PSMA positron emission tomography/computed tomography scans is also a useful biomarker. A couple of years back, we published the phase 3 VISION study, which investigated the prognostic value of declining prostate-specific antigen levels in patients with PSMA+ mCRPC receiving radioligand therapy (RLT). This study showed a significant association between declining prostate-specific antigen levels and improved outcomes, including rPFS, health-related quality of life, and response to ¹⁷⁷Lu-PSMA-617 treatment. This makes sense mechanistically; patients with brighter tumors have more target, and with more target, there is more radiation delivered to that target. RLT is not the end of the road. In fact, we are seeing RLT move into earlier settings, providing an alternative to chemotherapy.

However, the question remains: What if the tumor diverges and evolves away from AR or PSMA into aggressive variant disease (eg, neuroendocrine prostate cancer) or even double-negative disease (eg, AR- and neuroendocrine marker–negative cancer)? Both tumor types have a very poor prognosis and are treated with platinum-doublet chemotherapy.

At ASCO 2026, I gave a presentation on the multicenter phase 2 CHAMP study, which is the first study to evaluate the use of chemoimmunotherapy in patients with aggressive variant disease or small-cell histology (abstract 5016). In this 40-patient study, our goal was to see if adding dual immune checkpoint blockade (ie, the CTLA-4 blocker ipilimumab and the PD-1 blocker nivolumab) to chemotherapy (ie, cabazitaxel plus carboplatin) could extend PFS in this patient population. Overall, the study was positive. At 6 months, 74% of men in CHAMP were free of progression or death, which was significantly improved compared with approximately 55% based on historic data. The median rPFS was 12 months. This is an ongoing study in long-term follow-up right now, but I would say that it was a success from a phase 2 perspective. We would like to move this to a randomized phase 3 study to validate these results.

Looking to the future, there is strong interest in the cell surface targeting of KLK2, ACP3, B7H3, STEAP1, CD46, and neuroendocrine markers such as DLL3 and GPC3, among others, as well as in developing companion diagnostics for those markers, including positron emission tomography/computed tomography scans. I think that over the next 10 years, we are going to see a wealth of new biomarkers that will enable us to identify patients based on their cell surface antigen and then target that with a suite of targeted cell surface therapies.