(UroToday.com) The 2024 American Society of Clinical Oncology (ASCO) annual meeting featured a session on prostate cancer, and a presentation by Magdalena Fay discussing an independent blinded validation of an artificial intelligence-based digital histology classifier for prostate cancer recurrence and metastasis risk prediction. Artificial intelligence tools which identify pathology features from digitized whole slide images of prostate cancer generate data to predict risk of disease recurrence and metastasis. PathomIQ and ISMMS have developed an artificial intelligence-enabled prognostic test, PATHOMIQ_PRAD, which predicts risk of biochemical recurrence and distant metastasis using whole slide images. The objective of this study presented at the ASCO 2024 annual meeting was to evaluate the clinical validity of PATHOMIQ_PRAD using a retrospective clinical cohort at Cleveland Clinic. Secondly the performance of the test was also compared to Decipher, an established genomic risk classifier.
This was a retrospective PATHOMIQ_PRAD analysis of prostate cancer whole slide images of patients who underwent prostatectomy at Cleveland Clinic from 2009-2022 and did not receive any adjuvant therapy before biochemical recurrence. Overall, 263 patients in the cohort received definitive treatment with radical prostatectomy and had a median follow-up of 50 months. Of these patients, 65 patients had biochemical recurrence, and 14 patients developed distant metastasis as of last follow up. Whole slide images were de-identified, anonymized, and patient outcomes were blinded during the study. Patients were stratified into high-risk and low-risk categories based on pre-determined thresholds for PATHOMIQ_PRAD scores (0.45 for biochemical recurrence and 0.55 for distant metastasis):
The Kaplan-Meier method with log-rank was used to compare biochemical recurrence-free survival and metastasis-free survival. Multivariable Cox proportional hazards regression was used to identify factors associated with biochemical recurrence. The cohort clinical characteristics for these patients is as follows:
The rate of biochemical recurrence-free survival (>0.45 vs. <0.45, p < 0.0001) and metastasis-free survival (>0.55 vs. <0.55, p < 0.0001) were associated with PATHOMIQ_PRAD score:
The rate of biochemical recurrence-free survival (>0.6 vs. <0.6, p = 0.0009) and metastasis-free survival (>0.6 vs. <0.6, p = 0.0095) were also associated with Decipher score:
All 14 patients who had distant metastasis during the follow up time had a high PATHOMIQ_PRAD score. Univariable analysis shows that PATHOMIQ_PRAD reliably identifies patients at risk of biochemical recurrence (HR 4.19, p < 0.0001), and had comparable prognostic performance to Decipher (HR 2.83, p = 0.0013). Multivariable analysis showed that there was an increased risk of biochemical recurrence in both the high-risk PATHOMIQ_PRAD (HR 3.58, 95% CI 1.75-7.31) and Decipher (HR 2.20, 95% CI 1.16-4.17) groups relative to the low-risk groups, which suggests that combining the two tests may further improve risk stratification.
Magdalena Fay concluded his presentation by discussing an independent blinded validation of an artificial intelligence-based digital histology classifier for prostate cancer recurrence and metastasis risk prediction with the following take home messages:
- These results show that PATHOMIQ_PRAD continues to demonstrate clinical validity in predicting risk of biochemical recurrence and distant metastasis with favorable performance compared to a commonly used genomic classifier
- PATHOMIQ_PRAD may identify patients for early treatment intensification, as well as inform clinical trial patient selection
Presented by: Magdalena Fay, MD, MS, Cleveland Clinic Lerner College of Medicine, Cleveland, OH
Written by: Zachary Klaassen, MD, MSc – Urologic Oncologist, Associate Professor of Urology, Georgia Cancer Center, Wellstar MCG Health, @zklaassen_md on Twitter during the 2024 American Society of Clinical Oncology (ASCO) Annual Meeting, Chicago, IL, Fri, May 31 – Tues, June 4, 2024.