A new artificial intelligence (AI) tool can detect early signs of prostate cancer in biopsy samples before the disease becomes visible to pathologists, according to a study published in Nature.
The AI model identified subtle tissue changes in over 80% of patients later diagnosed with clinically significant prostate cancer, despite their initial biopsies being classified as benign.
The research, led by Carolina Wählby, PhD, of Uppsala University, analyzed biopsies from 213 men with elevated PSA levels. Of these, 88 developed aggressive prostate cancer within 30 months.
The AI model successfully flagged early warning signs—such as changes in stromal collagen and glandular cells—with high sensitivity (92%) and an area under the curve (AUC) of 0.82, outperforming standard pathological review.
“This shows that AI analysis of routine biopsies can detect subtle signs indicating clinically significant prostate cancer before it becomes obvious to a pathologist,” said Wählby.
“The study has been nicknamed the ‘missed study,’ as the goal of finding the cancer was ‘missed’ by the pathologists,” explained Wählby. “We have now shown that with the help of AI, it is possible to find signs of prostate cancer that were not observed by pathologists in more than 80% of samples from men who later developed cancer.”
Prostate cancer is the most common cancer in men, with 1 in 8 diagnosed in their lifetime.
Early detection is critical, as the disease is highly treatable when caught early. This AI tool could help identify high-risk patients earlier, enabling closer monitoring and timely intervention.
The study suggests that AI could soon play a key role in complementing traditional screening methods, improving early diagnosis and patient outcomes.