Melvin L.K. Chua, FRCR, PhD, FASCO, on validation of an MMAI model in Asian patients

Data presented at the American Society for Radiation Oncology 67th Annual Meeting validated the performance of the ArteraAI Prostate test, a digital pathology-based multimodal AI (MMAI) prognostic model, in an Asian cohort of patients with prostate cancer.¹ Overall, the test was able to accurately predict outcomes in these patients, validating it an independent prognostic tool across racial subgroups.

The test has been previously validated in Caucasian and African American patients.

In a recent interview with Urology Times®, presenting author Melvin L.K. Chua, FRCR, PhD, FASCO, outlined key findings and notable takeaways from the study. Chua is a clinician-scientist at the National Cancer Centre Singapore.

Data showed that MMAI score was significantly associated with distant metastasis (sHR, 2.59; 1.40 to 4.78; P = .003), biochemical failure (sHR, 2.09; 1.28 to 3.42; P = .003), and distant metastasis-free survival (HR, 2.00; 1.19 to 3.36; P = .009). These findings remained consistent when the investigators adjusted for clinical risk group. However, MMAI score was not found to be associated with overall survival (HR, 1.37; 0.63 to 3.00; P = .43).

Based on these results, Chua concluded, “Those are very robust findings, and it shows that the test is valid, [across demographics], because now the Artera test has been validated in the Western cohort, the African American cohort, as well as the Asian cohort. At the same time, it is also able to perform in a cohort that largely consisted of high-risk patients. Taken together, these are very compelling findings and the key results of our work.”

REFERENCES

1. Ong EHW, Hong BH, Joun S, et al. Validation of a prognostic multimodal artificial intelligence (MMAI) model in Asian prostate cancer (PCa) patients (pts) from Singapore. Presented at: American Society of Radiation Oncology 67th Annual Meeting. September 27-October 1, 2025. San Francisco, California. Abstract 1117