When a PSA test indicates an elevated level, determining whether a biopsy is necessary to confirm or rule out prostate cancer can be complex. A recent retrospective study conducted by researchers at the German Cancer Research Center (DKFZ) and the Department of Urology at Heidelberg University Hospital has revealed that integrating risk markers, thorough MRI evaluation, and artificial intelligence (AI) enhances the accuracy of prostate cancer risk assessment. Consequently, men with a low risk may avoid undergoing a biopsy.
Image Credit: MattL_Images/Shutterstock.com
However, final certainty can only be obtained by taking tissue samples from the prostate.
Biopsies are invasive and in rare cases can lead to infections or bleeding, sometimes even requiring hospitalization.
David Bonekamp, Radiologist, German Cancer Research Center
“Our aim is to filter out those men who only have a minimal risk of cancer. They could be spared tissue removal or postpone it for a certain period of time. Men with a high probability of prostate cancer, on the other hand, benefit from the biopsy, as the cancer can be detected early,” said Bonekamp.
Currently, researchers use a calculator to estimate the risk of prostate cancer by incorporating various parameters such as PSA levels, age, prostate volume, and MRI findings. This process involves the PI-RADS system, which standardizes MRI image evaluation to provide a probability score for prostate cancer presence.
To explore whether deep learning AI could enhance or potentially replace PI-RADS, Bonekamp’s team conducted a retrospective study using data from 1,627 men who underwent multi-parametric MRI of the prostate at Heidelberg between 2014 and 2021, followed by biopsy.
An algorithm developed by the DKFZ for analyzing MRI images was trained on data from over 1,000 of these men. The remaining data was used to assess whether combining this AI with the existing risk calculator could improve the accuracy of prostate cancer predictions.
The study found that substituting PI-RADS with the AI method alone did not significantly change diagnostic accuracy. However, integrating AI with PI-RADS markedly improved results, identifying 49 percent of men as minimal risk who had originally undergone biopsy.
This means that the combination of deep learning and radiological findings could theoretically have avoided almost half of these biopsies without overlooking a relevant number of tumors.
Adrian Schrader, Study First Author, German Cancer Research Center
The radiologists concluded that deep learning-based AI and PI-RADS assessments by experienced radiologists offer complementary diagnostic insights. Together, they enhance the precision of patient risk stratification.
For patients with an elevated PSA value, it could be a great advantage in the future to integrate AI analysis into further diagnostics. However, prospective studies must confirm the benefits of the procedure and clarify that it has no disadvantages for patients.
David Bonekamp, Radiologist, German Cancer Research Center
Journal Reference:
Schrader, A., et al. (2024) Prostate cancer risk assessment and avoidance of prostate biopsies using fully automatic deep learning in prostate MRI: comparison to PI-RADS and integration with clinical data in nomograms. European Radiology. doi.org/10.1007/s00330-024-10818-0