December 7, 2023

AI-Driven Digital Pathology Colorectal Cancer Diagnostic Tool Validated in Study

A groundbreaking AI-driven digital pathology diagnostic tool for colorectal cancer, MSIntuitâ„¢ CRC, has been validated in a blind study published in Nature Communications. Developed by techbio company Owkin in collaboration with pathology labs in France, the tool aims to enhance the precision of diagnosis and treatment for colorectal cancer. The disease, which caused nearly one million deaths worldwide in 2020, is the third most common cancer globally and the second leading cause of cancer mortality.

Microsatellite Instability (MSI) is a crucial genomic biomarker in colorectal cancer and accounts for approximately 15% of all cases. Recent clinical trials have established the prognostic and therapeutic significance of the MSI phenotype, particularly with the approval of immune checkpoint inhibitor (ICI) therapies. Patients with MSI-positive tumors are more likely to respond to ICI therapy, while those with microsatellite stable (MSS) tumors are not. Consequently, MSI testing is globally recommended as a guide for selecting the best therapy.

The study published in Nature Communications demonstrates that MSIntuit CRC can accurately rule out nearly 50% of MSS patients, while correctly classifying over 96% of MSI patients, achieving results comparable to the current gold standard methods. This innovative pre-screening tool has the potential to streamline the diagnostic process, reduce pressure on laboratory resources, and enable more patients to be screened faster.

Magali Svrcek, an international expert in GI pathology and co-last author of the publication, highlights the direct impact this approach will have on oncologist decision-making and improving patient access to the most effective treatment. The diagnostic tool also has the potential to optimize costs and organization of MSI testing in pathology labs, especially in countries that implement universal MSI screening.

Meriem Sefta, Chief Diagnostic Officer at Owkin, emphasizes the need for tools that can ease bottlenecks and resource pressures while increasing the efficiency of biomarker testing in clinical practice. The ultimate goal is to develop an AI diagnostic that can identify actionable biomarkers from a single routine clinical slide, bringing precision medicine closer to reality.

A key strength of the study is the blind validation of the model using 600 consecutive colorectal cancer cases diagnosed across nine different pathology labs over two years. This approach minimizes the risk of selection bias and ensures the robustness and generalizability of the AI model across labs and geographies.

To learn more about MSIntuit CRC and its availability, visit Owkin’s website at www. The full publication can be accessed here. This pioneering achievement by Owkin is driven by patient-derived data and aims to benefit patients by improving the accuracy and efficiency of colorectal cancer diagnosis.

1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it