Related Experiment Video
Updated: Jun 24, 2025
A Syngeneic Mouse Model of Metastatic Renal Cell Carcinoma for Quantitative and Longitudinal Assessment of Preclinical Therapies
Published on: April 12, 2017
Histopathology Based AI Model Predicts Anti-Angiogenic Therapy Response in Renal Cancer Clinical Trial
Jay Jasti1, Hua Zhong1,2, Vandana Panwar2
1Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, TX, USA.
Background:
Predictive biomarkers of treatment response are lacking for metastatic clearcell renal cell carcinoma (ccRCC), a tumor type that is treated with angiogenesis inhibitors, immune checkpoint inhibitors, mTOR inhibitors and a HIF2 inhibitor. The Angioscore, an RNA-based quantification of angiogenesis, is arguably the best candidate to predict anti-angiogenic (AA) response. However, the clinical adoption of transcriptomic assays faces several challenges including standardization, time delay, and high cost. Further, ccRCC tumors are highly heterogenous, and sampling multiple areas for sequencing is impractical.
Approach:
Here we present a novel deep learning (DL) approach to predict the Angioscore from ubiquitous histopathology slides. In order to overcome the lack of interpretability, one of the biggest limitations of typical DL models, our model produces a visual vascular network which is the basis of the model's prediction. To test its reliability, we applied this model to multiple cohorts including a clinical trial dataset.
Results:
Our model accurately predicts the RNA-based Angioscore on multiple independent cohorts (spearman correlations of 0.77 and 0.73). Further, the predictions help unravel meaningful biology such as association of angiogenesis with grade, stage, and driver mutation status. Finally, we find our model is able to predict response to AA therapy, in both a real-world cohort and the IMmotion150 clinical trial. The predictive power of our model vastly exceeds that of CD31, a marker of vasculature, and nearly rivals the performance (c-index 0.66 vs 0.67) of the ground truth RNA-based Angioscore at a fraction of the cost.
Conclusion:
By providing a robust yet interpretable prediction of the Angioscore from histopathology slides alone, our approach offers insights into angiogenesis biology and AA treatment response.
Insights
A new deep learning model predicts angiogenesis (Angioscore) from kidney cancer histopathology slides, offering a cost-effective alternative to RNA assays for predicting anti-angiogenic therapy response.
Area of Science:
- Oncology
- Computational Pathology
- Biomarker Discovery
Background:
- Metastatic clear-cell renal cell carcinoma (ccRCC) lacks predictive biomarkers for treatment response.
- Angiogenesis is a key target, with the RNA-based Angioscore predicting anti-angiogenic (AA) therapy response.
- Current transcriptomic assays for Angioscore are limited by cost, time, and heterogeneity challenges.
Purpose of the Study:
- To develop a deep learning (DL) model for predicting the Angioscore directly from histopathology slides.
- To overcome the interpretability limitations of traditional DL models by generating a visual vascular network.
- To validate the model's reliability across multiple independent cohorts, including a clinical trial.
Main Methods:
- A novel deep learning approach was employed to analyze histopathology images.
- The model was trained to predict the RNA-based Angioscore.
- Interpretability was enhanced by generating a visual vascular network as the basis for predictions.
- The model was validated on diverse patient cohorts, including a clinical trial dataset.
Main Results:
- The DL model accurately predicted the RNA-based Angioscore in independent cohorts (Spearman correlations of 0.77 and 0.73).
- Model predictions revealed biological associations between angiogenesis, tumor grade, stage, and driver mutations.
- The model successfully predicted response to AA therapy in both real-world and clinical trial data.
- Its predictive performance nearly rivaled the ground truth RNA-based Angioscore at a significantly lower cost.
Conclusions:
- This interpretable DL approach provides robust Angioscore prediction from histopathology slides.
- The method offers valuable insights into angiogenesis and anti-angiogenic treatment response in ccRCC.
- This technique presents a practical and cost-effective alternative to current genomic assays for clinical application.
More Related Videos
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
11:27Modeling Spontaneous Metastatic Renal Cell Carcinoma mRCC in Mice Following Nephrectomy
Published on: April 29, 2014