If ctDNA had social media, its first post would be: "Still here after treatment. Not great news. Please stop ignoring my red flags."
That is the emergency siren in this new Nature Cancer briefing. The paper highlights a prospective study called PARADIGM, which asked a very practical question: when men with high-volume metastatic prostate cancer start first-line treatment, can blood tests tell us early who is still in trouble?
The short answer: yes. PSA helps. ctDNA helps more. Together, they look less like a horoscope and more like a triage board.
PSA Is Useful. PSA Is Also Dramatic.
Prostate-specific antigen, or PSA, is the old workhorse. It is a protein made by prostate cells, and doctors have used it for decades to track prostate cancer. When treatment works, PSA often drops. When cancer grows, PSA often rises.
Often.
That word is doing a lot of unpaid labor.
PSA is tied to androgen receptor signaling, meaning hormone-blocking treatments can make PSA behave beautifully while some cancer cells quietly keep doing crime in the alley. PSA can be helpful, but it is not a tumor livestream. It is more like a neighbor texting, "Something seems weird next door," then going back to watering the plants.
ctDNA is different. Circulating tumor DNA is tiny scraps of genetic material shed by cancer cells into the bloodstream. A liquid biopsy looks for those scraps. No scalpel. No spelunking into bone metastases with a needle, which everyone agrees is a hobby best avoided.
The PARADIGM Study: Blood Draws With Consequences
The underlying PARADIGM cohort followed 114 biologically male patients with high-volume metastatic prostate cancer across 14 UK centers. These patients were starting androgen deprivation therapy plus either docetaxel chemotherapy or an androgen receptor pathway inhibitor, such as abiraterone, enzalutamide, or apalutamide.
Researchers checked blood repeatedly during the first six treatment cycles. They focused on cycles 3 or 4, about 6 to 12 weeks into combination treatment. That timing matters. It is early enough that doctors might still change course before the cancer has held a staff meeting and reorganized.
Before treatment, ctDNA appeared in 70% of a small sequential subgroup. By cycles 3 or 4, it was still detectable in 29% of the main evaluable group. That persistence was the problem.
Patients with detectable ctDNA at cycles 3 or 4 had a 12-month overall survival of 73%, compared with 99% for those without detectable ctDNA. At 24 months, survival was 50% versus 85%. That is not a subtle eyebrow raise. That is the lab result equivalent of someone walking into the emergency room holding their own chart and saying, "We should probably move."
The Plot Twist: ctDNA Speaks Earlier
PSA still mattered. Patients with higher PSA after treatment generally did worse. But ctDNA added information that PSA missed.
In multivariable models, ctDNA and PSA each carried independent prognostic value after combination treatment started. The poorest-risk group, defined by both markers looking bad, had a much higher hazard of death than the best-risk group.
Even sharper: ctDNA seemed to flag danger earlier than PSA. In samples taken after androgen deprivation therapy but before the second drug or chemotherapy began, ctDNA predicted shorter survival. PSA did not. PSA was still checking its clipboard. ctDNA had already called the code.
That is the appeal here. Not just prediction. Timing.
Why This Could Matter In Real Life
Metastatic prostate cancer is no longer one predictable road. Some patients do very well for years. Others progress fast despite modern treatment. Doctors now have more options: hormone intensification, chemotherapy, PARP inhibitors for selected DNA repair mutations, PSMA-targeted radioligand therapy, and clinical trials. The menu is expanding. Great. Also, slightly terrifying. Like being handed a 14-page diner menu during a fire drill.
A blood test that sorts patients early could help match intensity to risk. If ctDNA disappears, maybe treatment is working and extra toxicity can wait outside. If ctDNA persists, maybe that patient needs faster imaging, treatment escalation, trial enrollment, or closer follow-up.
This does not mean ctDNA should automatically dictate care tomorrow morning. The study was observational. It showed prognosis, not that changing treatment based on ctDNA improves survival. That next step needs clinical trials.
But the signal fits with other work. Recent reviews and metastatic prostate cancer studies consistently show that detectable or higher ctDNA tracks with worse progression-free and overall survival. In other words, when tumor DNA keeps showing up in blood, cancer is usually not just leaving confetti. It is leaving evidence.
The Catch, Because Biology Charges Hidden Fees
ctDNA testing is more expensive than PSA. It needs careful assays, good timing, and validation across treatment settings. Prostate cancer can also be technically annoying because many tumors shed limited DNA after therapy starts, and prostate cancer often carries copy-number changes rather than easy recurring point mutations.
The PARADIGM team used a prostate-adapted approach that detects tumor-specific copy-number imbalance. That is not glamorous cocktail-party material, unless your cocktail party is unusually intense, but it matters. The test has to find the kind of genomic mess prostate cancer actually makes.
Bottom Line
PSA is still useful. Keep it on the team.
But ctDNA may be the faster alarm. In high-volume metastatic prostate cancer starting first-line treatment, persistent ctDNA after a few weeks identified patients at much higher risk of death, and combining ctDNA with PSA improved survival prediction.
If future trials show that acting on this information improves outcomes, prostate cancer care could move from "wait and see" toward "see early and move." That is a better sentence. Also a better clinic day.
References
-
Added value of ctDNA testing with primary treatment for metastatic prostate cancer prognosis. Nature Cancer (2026). https://doi.org/10.1038/s43018-026-01176-5
-
Jayaram A, Rashid M, Reid AHM, et al. Combined ctDNA and serum PSA for dynamic monitoring of metastatic prostate cancer starting first-line treatment: a prospective national cohort study. Nature Cancer (2026). https://doi.org/10.1038/s43018-026-01172-9
-
Ye D, Zhao Y, Yuan M, et al. Prognostic significance of circulating tumor DNA in metastatic prostate cancer: a systematic review and meta-analysis. Clinical Genitourinary Cancer (2026). https://doi.org/10.1016/j.clgc.2026.102545
-
Sweeney CJ, et al. Circulating tumor DNA assessment for treatment monitoring adds value to PSA in metastatic castration-resistant prostate cancer. Clinical Cancer Research 30, 4115-4122 (2024).
-
Fonseca NM, Maurice-Dror C, Herberts C, et al. Prediction of plasma ctDNA fraction and prognostic implications of liquid biopsy in advanced prostate cancer. Nature Communications 15, 1828 (2024). https://doi.org/10.1038/s41467-024-45475-w
-
Kwan EM, Ng SWS, Tolmeijer SH, et al. Lutetium-177-PSMA-617 or cabazitaxel in metastatic prostate cancer: circulating tumor DNA analysis of the randomized phase 2 TheraP trial. Nature Medicine (2025). https://doi.org/10.1038/s41591-025-03704-9
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.