FDA Grants Breakthrough Device Designation to Artificial Intelligence-Powered Companion Diagnostic for Lung Cancer

The FDA has awarded breakthrough device status to the Artificial Intelligence-powered Ventana TROP2 RxDx Device, a first-of-its-kind computational pathology diagnostic for non–small cell lung cancer.

The U.S. Food and Drug Administration (FDA) has granted breakthrough device designation to the Ventana TROP2 (EPR20043) RxDx Device, marking the first time a computational pathology companion diagnostic (CDx) has received this certification in the field of non–small cell lung cancer (NSCLC). Developed by Roche Diagnostics, the immunohistochemistry assay integrates digital pathology algorithms and advanced artificial intelligence-based image analysis, significantly surpassing the accuracy of traditional manual scoring methods used in cancer diagnostics.

The device analyzes digital images of stained tissue slides, leveraging Roche Digital Pathology scanners to assess patient samples with enhanced precision for TROP2 protein expression. By incorporating AstraZeneca’s proprietary Quantitative Continuous Scoring computational pathology platform, the technology enables a new standard of diagnostically relevant depth. This improved precision is expected to help clinicians more accurately identify NSCLC patients who may benefit from targeted therapies, particularly Datroway (datopotamab deruxtecan-dlnk), a TROP2-directed antibody drug conjugate specifically engineered for advanced or metastatic cases lacking actionable genomic alterations.

The FDA breakthrough designation, as outlined by the agency and the National Institutes of Health, is reserved for devices that present transformative clinical potential based on preliminary evidence, triggering accelerated review and additional agency support. Roche and AstraZeneca highlight that this regulatory milestone should speed up the device’s availability to patients, streamlining personalized treatment decisions and potentially improving patient outcomes. Importantly, while the device automates much of the image analysis, pathologists remain pivotal in reviewing results within a proper clinical and histological context, ensuring robust validation for use in guiding oncology treatment.

Company leaders from both Roche Diagnostics and AstraZeneca emphasized the innovation’s significance in advancing precision oncology. The Ventana TROP2 RxDx Device is poised to assist healthcare professionals in delivering more personalized care through computational pathology and artificial intelligence, potentially setting a precedent for future diagnostic technologies in cancer care.

78

Impact Score

UK mps open inquiry into artificial intelligence and edtech in education

UK mps have launched a cross party inquiry into how artificial intelligence and education technology are reshaping learning across early years, schools, colleges and universities, and how government should balance innovation with safeguards. The education committee will examine opportunities to improve teaching and workload alongside risks around inequality, privacy, safeguarding and assessment.

Most UK firms see Artificial Intelligence training gap as shadow tool use grows

New research finds that 6 in 10 UK businesses say employees lack comprehensive Artificial Intelligence training, even as shadow use of unapproved tools becomes widespread and investment surges. Executives warn that without stronger skills, governance and strategy, many organisations risk missing out on expected Artificial Intelligence returns.

COSO issues internal control roadmap for governing generative artificial intelligence

COSO has released governance guidance that applies its Internal Control-Integrated Framework to generative artificial intelligence, offering audit-ready control structures and implementation tools for organizations. The publication details capability-based risk mapping, aligned controls, and practical templates to help institutions manage emerging technology risks.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.