ASCO GU 2022: Development and Validation of a Prognostic AI Biomarker Using Multi-Modal Deep Learning with Digital Histopathology in Localized Prostate Cancer on NRG Oncology Phase III Clinical Trials

( The 2022 GU ASCO Annual meeting included a session on the optimization management of localized prostate cancer, specifically looking at artificial intelligence (AI), active surveillance, and intervention, featuring a presentation by Dr. Osama Mohamad discussing the development and validation of a prognostic AI biomarker using multi-modal deep learning with digital histopathology in localized prostate cancer on NRG Oncology phase III clinical trials. Prognostication in localized prostate cancer is reliant on non-specific tools, an issue that leads to the over-and under-treatment of patients:

Osama Mohamad-0.jpg

Various tissue-based molecular biomarkers have attempted to fill this unmet need, but most lack prospective randomized trial validation. Advantages of AI tools utilizing digital histopathology are:

  • Performance: prognostic and predictive of treatment response
  • Short turnaround time: seconds of cloud computing
  • Robustness: tool is developed from thousands of patients
  • Widespread adoption: no consumption of tumor tissue

As such, Dr. Mohamad and investigators trained and validated prognostic biomarkers in localized prostate cancer using five phase III randomized trials, by leveraging multi-modal deep learning on digital histopathology. 

Histopathology image data was generated from pre-treatment biopsy slides in five NRG Oncology phase III randomized radiotherapy prostate cancer trials (RTOG 9202, 9408, 9413, 9910, and 0126). The trials were randomly split into training (80%) and validation (20%) cohorts. A multi-modal artificial intelligence architecture was developed to take clinicopathologic and image-based (histopathology) data as input and predict binary outcomes. Using this architecture, various models were trained to predict relevant clinical endpoints: biochemical recurrence (BCR), distant metastasis (DM), prostate cancer-specific survival, and overall survival (OS). These models were then validated for measures of prognostic discrimination using the time-based area under the curve (AUC) method:

Osama Mohamad-1.jpg

Clinicopathologic and histopathology image data were available for 5,654 of 7,957 eligible patients (71.1%), yielding 16.1 TB of data from 16,204 histopathology slides of pretreatment biopsy samples. After training the models, locking them, and evaluating them on the validation cohort, Dr. Mohamad and colleagues found that the multi-modal artificial intelligence prognostic model had superior discrimination compared to the NCCN model (PSA, T-stage, and Gleason score) for 5-year DM (AUC of 0.84 vs 0.73), 5-year BCR (AUC of 0.69 vs 0.58), 10-year prostate cancer-specific survival (AUC of 0.79 vs 0.66), and 10-year OS (AUC of 0.65 vs 0.58): 

Osama Mohamad-2.jpg

Within each of the individual trials in the validation cohort, the multi-modal artificial intelligence-model had superior performance compared to NCCN risk groups for all clinical endpoints.

Dr. Mohamad concluded his presentation of a prognostic AI biomarker using multi-modal deep learning with digital histopathology in localized prostate cancer on NRG Oncology phase III clinical trials with the following take-home messages:

  • This represents the first ever development and validation of prognostic biomarkers in localized prostate cancer using multiple large phase III clinical trials
  • Successfully validated that the multi-modal artificial intelligence-prognostic biomarkers are superior to standard clinical and pathologic variables in identifying future BCR, DM, prostate cancer-specific survival, and OS
  • This massively scalable technology is feasible and can help personalize the management of prostate cancer patients

Presented by: Osama Mohamad, MD, PhD, Radiation Oncologist, UCSF, San Francisco, CA

Co-Authors: Andre Esteva, Jean Feng, Shih-Cheng Huang, Douwe Van der Wal, Jeffry Simko, Sandy DeVries, Emmalyn Chen, Edward M. Schaeffer, Todd Matthew Morgan, Jedidiah Mercer Monson, Farah Naz, James Wallace, Michelle J. Ferguson, Jean-Paul Bahary, Howard M. Sandler, Phuoc T. Tran, Daniel Eidelberg Spratt, Stephanie L. Pugh, Felix Feng, Osama Mohamad 

Written by: Zachary Klaassen, MD, MSc – Urologic Oncologist, Assistant Professor of Urology, Georgia Cancer Center, Augusta University/Medical College of Georgia, @zklaassen_md on Twitter during the 2022 American Society of Clinical Oncology Genitourinary (ASCO GU) Cancers Symposium, Thursday, Feb 17 – Saturday, Feb 19, 2022 

email news signup