A routine blood test offers new insight into survival in advanced prostate cancer
30 Apr 2026
New analysis from the STAMPEDE trial show that a simple, routine blood test can help predict how long a patient with advanced prostate cancer is likely to live. The findings were recently published in The Lancet Oncology.
Prostate cancer is the most common cancer in men. Over the past two decades, treatments for advanced prostate cancer have improved significantly, and many men now live for years after diagnosis. However, prostate cancer does not behave the same way in every patient. Doctors have long faced a difficult balancing act; identifying who will benefit from more intensive treatment and who may do just as well with less. Getting that balance right matters because treating someone more aggressively than necessary is associated with increased side effects and poorer quality of life.
The new study takes an important step towards addressing that challenge. It shows that a routine blood test, taken as early as six weeks after starting treatment, can provide meaningful information about how a patient is likely to fare in the long term.
The test measures prostate-specific antigen (PSA), a protein produced by prostate cancer cells that can be detected in the blood. When men start hormone therapy for advanced prostate cancer, PSA levels typically fall, and lower PSA levels are generally associated with better outcomes. Researchers found that men whose PSA dropped to 0.2 ng/mL or below within 24 weeks of starting treatment had substantially better survival than those who did not. Crucially, men who reached this level as early as six or twelve weeks did just as well in the long run. This is an encouraging finding that suggests early PSA response could help guide treatment decisions sooner.
One of the most important results of the study was that a good PSA response does not tell the whole story on its own. Even among men who reached the same low PSA level by 24 weeks, survival outcomes varied considerably depending on how much cancer was present at the time of diagnosis. Among men treated with hormone therapy combined with abiraterone (a drug that blocks male hormones more powerfully than standard hormone therapy alone), those with a smaller burden of cancer who reached a PSA of 0.2 ng/mL or below had an eight-year survival rate of 64%. This compared with 45% of survival rate among men with more extensive spread. In other words, two patients with the same PSA reading at 24 weeks may have very different long-term prospects depending on how widely their cancer had already spread when treatment began.
This is where the findings become particularly useful for personalising care. By combining the PSA result at 24 weeks with information about disease burden at the start of treatment, doctors and patients can gain a much clearer picture of likely long-term outcomes. To support this approach in practice, researchers have developed a simple flowchart that uses these two pieces of information to estimate survival probabilities and help guide clinical decision-making.
For men with a smaller amount of cancer spread who achieve a very low PSA, there is a real question of whether lifelong hormone therapy is always necessary, or whether treatment could be safely reduced to spare unnecessary side effects. Trials such as DE-ESCALATE and LIBERTAS are now exploring this possibility. At the other end of the spectrum, men with more widespread disease who do not achieve a sufficiently low PSA response are more likely to need more intensive treatment. The TRIPLE SWITCH and PEACE6 trials are exploring that approach. In both cases, the aim is the same; to ensure that each patient receives the treatment that is right for them, rather than one-size-fits-all approach.
The analysis included 7,129 men who took part in the STAMPEDE clinical trial between 2005 and 2016, across the UK and Switzerland. Of these, 4,438 had metastatic prostate cancer, meaning the cancer had spread to other parts of the body such as the bones, and 2,691 had very high-risk non-metastatic disease, meaning the cancer had not yet spread but was considered at high risk of doing so. Participants were followed for up to 19 years, making this one of the most comprehensive long-term analyses of its kind.
Further information: