In a peer-reviewed study, a group of scientists from French-American tech-bio company Owkin and pathology laboratories in France have conducted a blind validation of MSIntuit™ CRC.
This is a pioneering AI-driven digital pathology diagnostic tool developed by Owkin, designed as a pre-screening tool with the goal of enhancing the accuracy of diagnosing and treating colorectal cancer.
Colorectal cancer is a significant global health concern, with nearly two million new cases and one million deaths recorded worldwide in 2020. It ranks as the third most prevalent cancer worldwide and is the second leading cause of cancer-related deaths.
Microsatellite Instability (MSI) is a crucial genomic biomarker in colorectal cancer (CRC), accounting for approximately 15% of the overall CRC population. Recent clinical trials have demonstrated the prognostic and therapeutic significance of the MSI phenotype, especially with the approval of immune checkpoint inhibitor (ICI) therapies.
Patients with MSI-positive tumors are likely to respond to ICI therapy and are recommended for this treatment, while ICI is typically not advised for those with microsatellite-stable (MSS) tumors. As a result, MSI testing is now globally recommended by consensus guidelines to determine the most suitable therapy.
The development of prescreening tools offers the potential to streamline the process and alleviate the burden on laboratory personnel and resources by identifying patients who are most likely to benefit from MSI testing.
Today's study published in Nature Communications reveals that MSIntuit CRC can precisely rule out nearly 50% of MSS patients while correctly classifying more than 96% of MSI patients, on par with current gold standard methods (92-95%). Such novel solutions give the way for an optimized screening workflow to screen more patients quickly.
“This new approach will have a direct impact on oncologist decision-making and help bring the best treatment to patients sooner. It could also optimize costs and organization of MSI testing in pathology labs, especially for countries applying universal MSI screening,” notes Magali Svrcek, International expert in GI pathology, Professor at Saint Antoine Hospital, Sorbonne Université, AP-HP, France, and co-last author of this publication.
With the increasing number of biomarkers to be routinely tested in clinical practice, the need for tools that can both ease bottleneck and resource pressures while ramping up biomarker testing is paramount. Our solution represents the first step towards the development of an AI diagnostic that can identify actionable biomarkers from a single H&E slide used in clinical routine, pushing us closer to realizing a precision medicine future.
Meriem Sefta, Chief Diagnostic Officer, Owkin
A notable strength of the study lies in the blind validation of the AI model, which was conducted on a dataset of 600 consecutive colorectal cancer (CRC) cases diagnosed at nine different pathology labs over a two-year period.
This approach reduces the risk of selection bias and benefits from the participation of Medipath, the largest pathology lab network in France. The validation process included the use of two different pathology slide scanners and consistently achieved high sensitivity rates of 96% and 98%, respectively.
Furthermore, the study employed various methodological approaches, with a strong focus on sensitivity and specificity as performance indicators. These measures underscore the commitment to ensuring the robustness of the AI model and its suitability for clinical use while also striving for generalizability across various laboratories and geographic regions.
Saillard, C., et al. (2023). Validation of MSIntuit as an AI-based pre-screening tool for MSI detection from colorectal cancer histology slides. Nature communications. doi.org/10.1038/s41467-023-42453-6