AI Test Identifies Prostate Cancer Patients Who Benefit Most from Abiraterone Treatment

Breakthrough in Prostate Cancer Treatment

Researchers from University College London (UCL) and the Institute of Cancer Research have developed a groundbreaking artificial intelligence (AI) test that can identify which men with prostate cancer will benefit most from the life-extending drug abiraterone. The findings, presented at the American Society of Clinical Oncology (ASCO 2025) Annual Meeting, could revolutionize prostate cancer treatment and potentially save lives 1

1

2

AI-Powered Precision Medicine

Source: News-Medical

The new AI test, developed by Artera Inc., analyzes prostate cancer tumor samples and identifies features invisible to the human eye. By studying biopsy images from over 1,000 men who participated in the STAMPEDE trial, researchers found that abiraterone, when given alongside standard hormone therapy, can almost halve the risk of death for approximately 25% of men with high-risk prostate cancer that has not spread 1

1

3

Professor Gert Attard from UCL Cancer Institute explained, "This study shows, in a very large cohort of patients, that novel AI algorithms can be used to extract information from routinely available pathology slides to tailor these treatments to specific patients and minimize overtreatment whilst maximizing the chance of cure" 1

1

Significant Impact on Patient Outcomes

The AI test assigns patients a score - either biomarker-positive or biomarker-negative - which correlates with their treatment outcomes:

1. For biomarker-positive tumors (25% of cases):

   • Abiraterone reduced the risk of death after five years from 17% to 9% 2

     2

     4

     4

     .

2. For biomarker-negative tumors:

   • The risk of death without abiraterone was 7%, reduced to 4% with the drug.
   • This difference was not statistically or clinically significant 2

     2

     4

     4

     .

These results suggest that patients with biomarker-negative tumors could benefit from standard therapy alone, avoiding unnecessary treatment and potential side effects associated with abiraterone 3

3

Cost-Effectiveness and Accessibility

Source: Euronews

Professor Nick James, co-lead of the trial, emphasized the potential cost-effectiveness of this approach. Since the patent for abiraterone expired in 2022, the drug now costs just £77 per pack, significantly less than newer cancer treatments 2

2

5

However, access to abiraterone varies across the UK. While it is available for men with newly diagnosed, high-risk prostate cancer that has not spread in Scotland and Wales, NHS England has not approved its use for this group, citing cost concerns 2

2

5

Related Stories

NHS Trials AI-Powered Same-Day Prostate Cancer Diagnosis

NHS Trials AI Tool to Cut Prostate Cancer Diagnosis Time from Weeks to Hours

AI Tool Revolutionizes Prostate Cancer Treatment: Predicts Success of Focal Therapy

AI Model Shows Promise in Assessing Prostate Cancer Aggressiveness and Guiding Treatment

Calls for NHS England to Review Decision

Experts are urging NHS England to reconsider its decision not to fund abiraterone for the 8,400 men per year with newly diagnosed, high-risk prostate cancer that has not yet spread. Dr. Matthew Hobbs, Director of Research at Prostate Cancer UK, stated, "These exciting results suggest a way to make this an even more cost-effective approach. We therefore echo the researchers' urgent call for abiraterone to be made available to those men whose lives it can save" 1

1

4

Future Implications

Source: The Telegraph

This AI-powered approach to prostate cancer treatment represents a significant step towards personalized medicine. By identifying patients most likely to benefit from specific treatments, healthcare systems can optimize resource allocation and improve patient outcomes 3

3

5

As AI continues to advance in medical applications, similar approaches could be developed for other types of cancer and diseases, potentially transforming healthcare delivery and improving patient care across various medical fields.