AI model developed by SOPHiA GENETICS and UroCCR predicts post-operative outcomes in kidney cancer study

SOPHiA GENETICS (Nasdaq: SOPH), a cloud-native software company and a leader in data-driven medicine, and the French Kidney Cancer Research Network (UroCCR) collaborated on a study using a multimodal algorithm to help predict post-operative outcomes for those facing renal cell carcinoma (RCC), with the results recently published in npj Precision Oncology. The findings of the study showed the artificial intelligence (AI) model co-constructed by SOPHiA GENETICS and UroCCR provided a strong prediction for post-operative outcomes when compared to usual prognostic scores. This publication follows a prior collaboration that demonstrated the value of multimodal analysis on pre-operative kidney cancer upstaging. 

UroCCR is one of the world’s largest collaborative kidney cancer network with 51 multidisciplinary clinical teams across France. In close relationship with the French Association of Urology (AFU), the goal of UroCCR is to connect a national, multidisciplinary network of medical and scientific professionals who focus on therapeutic management and applied research into kidney cancer. In 2021, UroCCR partnered with SOPHiA GENETICS to develop an AI-based model to predict whether kidney cancer will progress from a localized tumor after surgery.

The UroCCR database provided multimodal, real-world data, including radiological, clinical, and biological data from more than 3,300 patients across France operated between May 2000 and January 2020. Researchers applied SOPHiA GENETICS’ proprietary AI offering to analyze the data into easily visualized, reliable predictions that were measured against, and outperformed, most common risk scores.

The volume and complexity of available health data continues to increase, and while this can assist in personalized diagnostics and treatment, it is most effective when paired with AI. Our work with UroCCR over the last three years has significantly advanced its research of RCC and shown the power of AI to provide insights from real-world multimodal data. We are extremely pleased with the findings from this study and look forward to continued collaboration with UroCCR.”  

Thierry Colin, VP, Multimodal Research and Development, SOPHiA GENETICS

SOPHiA GENETICS’ technology and global decentralized network are designed to break data silos and empower researchers with data-driven insights to drive the use of precision medicine. The study from UroCCR and SOPHiA GENETICS shows that an AI-based prediction model has the potential to support clinical treatment decision-making and provide an indication of which patients may potentially benefit from adjuvant systemic therapy, versus those that can sustain with surveillance.

“At UroCCR, our research is centered on the idea that a shared database helps facilitate and expand the use of precision medicine for patients facing kidney cancer,” said Pr Jean-Christophe Bernhard, M.D., PhD., Urologic Surgeon at CHU Bordeaux and head of the UroCCR. “Our work with SOPHiA GENETICS and the results of our latest study demonstrated the benefits of using AI to analyze real-world multimodal data. SOPHiA GENETICS’ technology and network have been key to the success of our research and will be imperative as we continue progressing with increased data inputs.”

UroCCR was created in 2011 and was funded and labeled by the French National Cancer Institute (INCa) as one of the fourteen official nationwide clinical and biological databases (BCB). The network makes it possible to identify and document clinical, biological, and radiological data – in a shared database – for all newly diagnosed patients in participating centers. The network has been referenced in 2023 by the High Authority on Health (HAS) as real-life data provider of interest. For more information connect on X and LinkedIn.

The SOPHiA DDM™ Platform has the ability to analyze multimodal data due to its cloud-based AI-driven environment that integrates and standardizes diverse data modalities to fuel the development of predictive models capable of addressing research questions.