The U.S. Food and Drug Administration (FDA) has granted Breakthrough Device designation to Cleo Skin, an AI pathology tool for diagnosing skin cancers using digitized tissue slides.  

Cleo Skin analyzes histologic whole-slide images from skin biopsies and excised tissue to identify suspicious areas, classify lesions, and provide key measurements. Lesions can be classified as melanoma, nevus, squamous cell carcinoma, basal cell carcinoma, or other skin cancer. The tool can also measure lesion dimensions and margins and identify areas of mitotic activity. The AI solution can be integrated into laboratories’ existing image management systems.  

The developer, Primaa, noted that the device is intended to support rather than replace pathologists’ review and judgment.  

Supporting Evidence 

In the multicenter SkinPerf2026 study, 16 French pathologists from 11 French institutions reviewed 313 whole-slide images with Cleo Skin. The device increased sensitivity for melanoma detection from 83.8% to 88.6%. Among junior pathologists, sensitivity rose from 79.7% to 88.2%, reaching a level comparable to that of their senior colleagues. The median time to detect mitosis decreased by 24%.  

“Mitotic counting remains an important but time-consuming part of melanoma assessment,” said Wonwoo Shon, DO, Director of Dermatopathology Services at Cedars-Sinai Medical Center and Founder of Aimes Diagnostics. “What is particularly useful about Cleo Skin is that its gallery-based mitosis detection can transform this repetitive search into a more focused and efficient review. This is a practical example of how AI can support, not replace, the pathologist by streamlining routine tasks and allowing greater attention to findings that require expert judgment. I believe this type of human-AI collaboration represents the future of pathology.”   

The device received CE-IVDR certification in Europe in March 2026 and has since been used by European pathologists.  

An exploratory U.S. study evaluated Cleo Skin’s use with PathPresenter, a digital pathology image management system. Seven pathologists used the integrated system to review 75 whole-slide images. The study showed a 20% reduction in review time and a 17% improvement in sensitivity among junior pathologists.  

“The best AI in pathology works in the background and leaves the judgment in the foreground. Cleo Skin acts like a colleague who has already been through the slide, highlighting where to look and proposing key measurements, so the evidence is ready when the pathologist makes the final decision. In our study, junior pathologists gained the most in accuracy,” said Rajendra Singh, MD, Associate Vice Chair of Pathology at the University of Pennsylvania and Co-founder of PathPresenter. 

The FDA designation provides the company with prioritized interaction with the agency ahead of a U.S. regulatory submission. It does not constitute marketing authorization.  

“We look forward to working closely with the FDA through the Breakthrough Devices Program to bring this support to U.S. pathologists, who face the same growing pressure on skin biopsy volumes,” said Hugo Watel, Chief Operating Officer of Primaa.  

ASCO AI in Oncology is published by Conexiant under a license arrangement with the American Society of Clinical Oncology, Inc. (ASCO^®). The ideas and opinions expressed in ASCO AI in Oncology do not necessarily reflect those of Conexiant or ASCO. For more information, see Policies.