Researchers have trained and validated an artificial intelligence (AI) model based on magnetic resonance imaging (MRI) scans to develop a consistent method of estimating prostate cancer lesion size, according to a recent study published by Yang et al in Radiology. The findings could aid physicians in making more accurate and informed treatment decisions.
Background
Prostate cancer is the second most common cancer type in U.S. men, with nearly 300,000 new diagnoses per year.
MRI has improved physicians’ ability to diagnose prostate cancer and is a routine part of diagnosis and treatment. Although physicians can estimate tumor size based on MRI images, these estimates are somewhat subjective and can vary from person-to-person.
Study Methods and Results
In the recent study, the researchers used the MRI scans of over 732 patients with prostate cancer undergoing treatment at a single center to train an AI model—with the goal of developing a more consistent method of estimating tumor size. They then investigated whether the AI model’s tumor size estimates were associated with treatment success in the 5 to 10 years following diagnosis.
The researchers found that the AI model was able to identify and demarcate the edges of 85% of the most radiologically aggressive prostate cancer lesions: those with a Prostate Imaging Reporting and Data System score of 5, indicating a very high risk of clinically significant prostate cancer. The AI model’s size estimates also showed potential as a prognostic marker. For instance, tumors with a larger volume were associated with a higher risk of recurrence, treatment failure, and metastasis—independent of other factors typically used to estimate this risk—among those treated surgically or with radiation therapy. Further, among the patients who received radiation therapy, AI-determined tumor volume performed better than traditional risk stratification in predicting metastasis.
“The AI measurement itself can tell us something additional in terms of patient outcomes,” explained senior study author Martin King, MD, PhD, of the Department of Radiation Oncology at Brigham and Women’s Hospital. “For patients, this can really tell them something about … the chances of cure and the likelihood that their cancer will reoccur or metastasize in the future,” he added.
Conclusions
The researchers suggested that the AI tool could be used to help physicians better understand a tumor’s aggressiveness, to inform more personalized treatment plans, and to guide radiation therapy. In addition to helping physicians and patients understand their cancer’s aggressiveness, the AI model could help guide radiation oncologists by pinpointing the tumor’s focal region for more targeted treatment. The new strategy was found to be quicker compared with currently used methods of predicting prostate cancer aggressiveness, which usually take 2 weeks or longer to yield results. AI-informed testing could provide an opportunity for earlier treatment.
“Al-determined tumor volume has the potential to advance precision medicine [in] patients with prostate cancer by improving our ability to understand the aggressiveness of a patient's cancer and therefore recommend the most optimal treatment,” highlighted lead study author David D. Yang, MD, of the Department of Radiation Oncology at Brigham and Women’s Hospital and a founding member of the Mass General Brigham health-care system.
The researchers plan to assess their model with a larger, multi-institutional dataset.
“We want to validate our findings, using other institutions and patient cohorts with different disease characteristics, to make sure that this approach is generalizable to all patients,” concluded Dr. Yang.
Disclosure For full disclosures of the study authors, visit pubs.rsna.org.
