FDA Grants Breakthrough Status to Roche’s AI-Based Lung Cancer Diagnostic
BASEL, 29TH APRIL, 2025 – Roche today announced that the U.S. Food and Drug Administration (FDA) has granted Breakthrough Device Designation to its VENTANA® TROP2 (EPR20043) RxDx Device, the first AI-powered companion diagnostic of its kind for non-small cell lung cancer (NSCLC).
The device combines an immunohistochemistry (IHC) assay with an advanced digital pathology algorithm to provide unprecedented diagnostic precision.
Whole slide images of NSCLC tissue are analyzed by artificial intelligence to compute a TROP2 score quantitatively, which was previously done by less consistent manual scoring.
“This FDA Breakthrough Device Designation is another example of our commitment to deliver innovation that enables more precise diagnosis in oncology,”
said Matt Sause, CEO of Roche Diagnostics.
“This solution, which leverages our industry-leading expertise in companion diagnostics development, uses artificial intelligence for a greater depth of sample analysis, helping to deliver truly personalized treatment.”
This diagnostic solution includes AstraZeneca’s Quantitative Continuous Scoring (QCS) platform, enabling reliable and more reproducible classification of TROP2 expression.
This digital pathology system includes Roche’s image management platform (navify®), pathology scanners (DP 200, DP 600), and the BenchMark ULTRA IHC/ISH instrument for tissue staining.
“This FDA Breakthrough Device Designation underscores the potential of our computational pathology platform to enable more personalised treatment decisions for people with cancer,”
Said Susan Galbraith, Executive Vice President, Oncology Haematology R&D, AstraZeneca.
This designation could help accelerate access to the TROP2 CDx and support their treating clinicians in identifying patients likely to respond to DATROWAY® (datopotamab deruxtecan-dlnk), a novel TROP2-directed antibody drug conjugate being developed by Daiichi Sankyo and AstraZeneca.
The VENTANA TROP2 (EPR20043) RxDx Device is used to treat patients with advanced or metastatic non-squamous NSCLC who do not have actionable genomic alterations.
After being reviewed by a pathologist, the AI algorithm computes a Normalised Membrane Ratio (NMR) to define TROP2 expression as either positive or negative and implicates its therapeutic accuracy for targeting.
CrowdStrike Unveils Agentic AI Innovations for Next-Gen SOC
Sonata Software Wins $73M AI-Led Digital Deal with US TMT Giant
