Radiological Features/Serum PSA May Predict Prostate Cancer Survival

Serum prostate-specific antigen (PSA) scores and radiological features could be combined to optimize the prediction of survival outcomes among patients with metastatic or very high-risk non-metastatic prostate adenocarcinoma, according to a post hoc analysis of 5 phase 3 trials using the STAMPEDE platform protocol (NCT00268476) published in The Lancet Oncology.¹ Moreover, metastatic volume or nodal status was found to influence survival rates associated with on-treatment PSA categories.

Key Data

The analysis demonstrated that an on-treatment PSA concentration of 0.2 ng/mL or less at any measured timepoint predicted favorable survival outcomes. While PSA at 24 weeks showed the strongest association with overall survival (OS), reaching a concentration of 0.2 ng/mL or less as early as 6 or 12 weeks was associated with equivalent survival rates. Specifically, in the metastatic cohort, the 96-month OS rate for patients with a PSA of 0.2 ng/mL or less was 47.9% (95% CI 43.2%-52.4%) at 6 weeks, 50.2% (95% CI, 46.6%-53.8%) at 12 weeks, and 50.3% (95% CI, 47.3%-53.2%) at 24 weeks.

Investigators found that survival rates within PSA subcategories differed significantly based on metastatic volume and nodal status. Among patients with low-volume metastatic disease who reached a PSA of 0.2 ng/mL or less at 24 weeks, the 96-month OS rate was 57.1% (95% CI, 53.0%-61.0%), which was higher than the rate of 39.8% (95% CI, 35.0%-44.5%) observed in patients with high-volume metastatic disease. For those receiving abiraterone acetate (Zytiga) with or without enzalutamide (Xtandi), survival was longest. Within this abiraterone-treated group, patients with metastatic low-volume disease and a PSA of 0.2 ng/mL or less at 24 weeks experienced a 96-month OS rate of 64.1% (95% CI, 57.8%-69.8%) compared to 44.6% (95% CI, 37.1%-51.9%) for those with high-volume disease.

In the non-metastatic cohort, the 96-month OS rate was highest for patients with node-negative disease at 82.8% (95% CI, 78.7%-86.1%). For patients treated with abiraterone who had node-positive disease and a PSA of 0.2 ng/mL or less at 24 weeks, the 96-month OS rate was 79.4% (95% CI, 73.8%-83.9%). Multivariable Cox models revealed that increasing PSA concentrations were associated with steep rises in the hazard of death, particularly at values above 1.5 ng/mL in the non-metastatic cohort and 3 ng/mL in the metastatic cohort.

Trial Details

Researchers assessed 7129 patients with metastatic or very high-risk non-metastatic prostate adenocarcinoma across 126 clinical sites in the UK and Switzerland to establish the prognostic value of serum PSA nadir values for OS between October 5, 2005, and September 2, 2016. This post hoc analysis utilized data from 5 individually powered phase 3 trials conducted at 126 hospitals or oncology centers in Switzerland and the UK.

The investigators performed this post hoc analysis protocol to determine if combining on-treatment PSA concentrations with radiological features like metastatic volume or nodal status could improve survival predictions. This study aimed to provide robust evidence to inform prognosis and treatment selection, particularly evaluating whether earlier PSA measurements at 6 or 12 weeks could predict outcomes as effectively as the established 24-week landmark.

Patients were randomly assigned to either standard of care (androgen deprivation therapy [ADT] alone or ADT plus docetaxel [Taxotere]) or experimental groups including ADT plus docetaxel with or without zoledronic acid (Zometa), ADT plus abiraterone acetate with or without enzalutamide, or ADT plus prostate radiotherapy.

Of eligible patients with a PSA value at 1 or more landmarks, 4438 of patients had metastatic disease, and 2691 had very high-risk non-metastatic disease.

• Among metastatic patients with available imaging, 2211 (55.9%) of 3956 had high-volume metastases according to the CHAARTED definition.
• In the non-metastatic cohort, 1033 (38.4%) of 2691 patients were lymph node positive.
• The median follow-up for the study was 9.6 years (IQR 8.3-11.4).
• Inclusion required a WHO performance score of 0 to 2, a neutrophil count higher than 1.5 x 10⁹ cells per L, and a platelet count higher than 100 x 10⁹ per L.
• The median time from starting ADT to randomization was 50 days (IQR 30-69) for the metastatic cohort and 43 days (IQR, 21-64) for the non-metastatic cohort.

End Points

The primary outcome for this analysis was OS, defined as the time from the specified landmark timepoints (6, 12, or 24 weeks after randomization) to death from any cause. Researchers identified that 3327 (75.0%) of 4438 patients with metastatic disease and 1033 (38.4%) of 2691 patients with non-metastatic disease died during the study period. To improve data completeness, survival data for patients in England and Wales were supplemented through linkage with the Civil Registration of Death dataset. A secondary post-hoc exploratory endpoint was prostate cancer-specific survival, which was evaluated using Fine-Gray competing-risks regression.

Safety

The study authors noted that while prostate cancer is a heterogeneous disease, some patients might achieve similar outcomes with less intense regimens that are less toxic, which supports the exploration of treatment de-intensification for certain subgroups. Conversely, the data indicated that treatment intensification should be considered for patients with high-volume metastatic disease or those who do not reach a PSA of 0.2 ng/mL or less. Palliative radiotherapy was permitted throughout the trial for symptom control in patients with metastatic disease. The analysis did not report new specific adverse effects but highlighted that 121 (1.7%) of 7129 eligible patients withdrew consent and were censored.

Reference

Kayani M, Murphy L, Dutey-Magni P, et al. On-treatment serum prostate-specific antigen and overall survival in prostate cancer (STAMPEDE platform protocol): a post-hoc analysis of data from five phase 3 trials. Lancet Oncol. 2026;27(5):625-636. doi:10.1016/S1470-2045(26)00066-5