Prostate Cancer Risk Factors Do Not Predict Hormone Therapy Benefit

An individual patient data (IPD) meta-analysis published in European Urology has revealed that traditional adverse pathological features used to classify aggressive prostate cancer after surgery do not predict which patients will benefit from adding hormone therapy (HT) to postoperative radiation.

The study, led by Amar Kishan, MD, professor and executive vice chair of radiation oncology at the David Geffen School of Medicine at UCLA, analyzed data from 4,781 patients across five phase 3 randomized clinical trials. The findings highlight a critical clinical paradigm shift: while traditional post-surgical pathological risk factors are strongly prognostic of overall patient outcomes, they cannot serve as predictive biomarkers to guide the use of hormone therapy after surgery.

The Adverse Feature Count (AFC) Score

In clinical practice, physicians commonly utilize surgical pathology from a radical prostatectomy to estimate how aggressive a patient's disease is and to guide subsequent salvage or adjuvant treatment. To evaluate whether these post-surgical risk factors modify the relative benefit of postoperative HT, Kishan and his team developed an Adverse Feature Count (AFC) score based on four high-risk pathological findings:

• High-grade disease (Grade Group 4 or 5)
• Seminal vesicle invasion (SVI)
• Positive surgical margins (PSM)
• Extracapsular extension (ECE)

A patient is assigned a score from 0 to 4 based on the number of these features present. In their analysis, the investigators confirmed that the AFC is a robust prognostic marker for long-term clinical outcomes. A higher AFC score was independently associated with a significant increase in the risk of death and cancer metastasis. Specifically, each additional adverse feature increased the risk for worse overall survival (OS) by 25% (hazard ratio [HR], 1.25; 95% confidence interval [CI], 1.16–1.36; p < 0.001) and worse metastasis-free survival (MFS) by 29% (HR, 1.29; 95% CI, 1.20–1.38; p < 0.001).

The Crucial Prognostic vs. Predictive Distinction

While the AFC proved to be a powerful prognostic tool, it failed to demonstrate predictive capability for hormone therapy benefit. That is, patients with multiple adverse features did not experience a greater relative improvement in survival outcomes from adding HT compared to those with fewer or no adverse features.

The AFC did not significantly modify the OS benefit of HT (interaction HR, 0.93; 95% CI, 0.79–1.10; p = 0.4) or the MFS benefit of HT (interaction HR, 0.88; 95% CI, 0.77–1.01; p = 0.08). These results remained consistent across sub-analyses, including categorical groupings and a pre-specified sensitivity analysis restricted only to Grade Group 4–5 and SVI.

The authors concluded: "Overall, the data suggest that although adverse pathological features are prognostic, they do not appear predictive of HT benefit after postoperative radiotherapy, which is relevant for patients with a PSA ≤0.5 ng/ml before radiation."

Kishan emphasized this core clinical challenge in a UCLA Health news release: “These findings highlight an important distinction between predicting risk and predicting treatment benefit. While pathology helps us understand which cancers are more aggressive, we need biomarkers that can tell us which patients are most likely to benefit from a specific therapy.”

To construct this robust IPD meta-analysis, the researchers compiled individual patient-level data from five major phase 3 randomized clinical trials comparing postoperative radiation therapy alone with radiation therapy plus HT.

Actionable Nursing Implications

For oncology nurses, these findings have major clinical and supportive care implications. Because hormone therapy reduces hormone levels that fuel prostate cancer growth, it is associated with systemic, quality-of-life-altering toxicities. As reported in the UCLA Health news release, hormone therapy can cause side effects, including fatigue, hot flashes, sexual dysfunction, bone loss and metabolic changes. Oncology nurses are at the frontline of educating patients about these adverse effects and managing them clinically.

Knowing that patients with a pre-radiotherapy PSA level of 0.5 ng/mL or less derive limited overall survival benefit from HT, and that high-risk pathology does not increase this benefit, nurses can support crucial shared decision-making and potential treatment de-escalation discussions.

This spares low-PSA patients from experiencing substantial treatment-related toxicities when no therapeutic benefit is predicted.

This study points to a future centered on tumor biology rather than physical pathology. The study’s authors note that these results "further underscore the need for predictive biomarkers, such as PAM50 shown in NRG GU006, to personalize treatment based on biology rather than prognosis." Oncology nurses must remain informed about emerging genomic and molecular assays, as they will define the next generation of personalized oncology care.

References

1. Kishan AU, Sun Y, Parker CC, et al. Influence of Adverse Pathologic Features on Potential Benefit of Hormone Therapy Use with Postoperative Radiotherapy in Recurrent Prostate Cancer: An Individual Patient Data Meta-analysis. Eur Urol. Published online July 28, 2026. doi:10.1016/j.eururo.2026.07.013.
2. Heady D. Study finds traditional prostate cancer risk factors do not predict hormone therapy benefit. UCLA Health News. August 10, 2026.