FDA grants 510(k) clearance to AI-powered prostate cancer diagnostic

The FDA has granted a 510(k) clearance to the Ibex Prostate Detect software (formerly Galen Second Read), an in vitro AI-powered digital pathology solution that can assist in identify small and rare prostatic cancers via tissue biopsies, Ibex Medical Analytics announced in a news release.¹

The system achieved a PPV of 99.6% for heatmap accuracy.

According to the company, Ibex Prostate Detect “analyzes scanned histopathology whole slide images (WSIs) from prostate core needle biopsies prepared from hematoxylin & eosin (H&E) stained formalin-fixed paraffin embedded tissue.”

The software uses AI to generate heatmaps that can identify prostatic tumors that may have been missed by pathologists on the initial diagnostic assessment. If a tissue is determined to be suspicious of cancer, the software sends case- and slide-level alerts in addition to a heatmap that guides pathologists on the areas likely to be cancerous.

"Prostate cancer is one of the most prevalent cancers worldwide, with an estimated 1 in 8 men diagnosed in their lifetime,” said Ibex scientific advisor Mahul B. Amin, MD, vice president and medical director of Labcorp’s Hospital Systems Operating Division and clinical professor of pathology and laboratory medicine at the University of Tennessee Health Science Center, in the news release.¹ “With the global incidence expected to double by 2040, accurate and timely diagnoses are more critical than ever. Advancements in AI-powered pathology can support pathologists in delivering precise and efficient diagnoses, ultimately improving patient outcomes.”

The FDA clearance is supported by data from precision and clinical validation studies of the software conducted across laboratories in the US and Europe.

Overall, the system achieved a positive predictive value (PPV) of 99.6% for heatmap accuracy. The tool was also able to identify 13% of cancer cases that were missed by pathologists diagnosing without the AI tool. These missed cases were subsequently confirmed by pathologists during the studies.

Specifically, in a validation study published in Lancet Digital Health,² the AI tool achieved an area under the receiver operating characteristic curve (AUC) of 0.991 (0.979 to 1.00) for detecting cancer in an external validation set and an AUC of 0.957 (0.930 to 0.985) for perineural invasion. The system also achieved an AUC of 0.941 (0.905 to 0.977) for distinguishing between low-grade (Gleason grade 6 of atypical small acinar proliferation) and high-grade tumors (Gleason Grade 7-10) and an AUC of 0.971 (0.943-0.998) for detecting Gleason pattern 5.

In the study, the AI diagnostic provided 560 cancer alerts, of which 9% (n = 51) led to additional cuts or stains being ordered.

The clearance for Ibex Prostate Detect marks the first FDA clearance for Ibex Medical Analytics. Their Ibex platform includes solutions across prostate cancer, breast cancer, and gastric cancer.

“The FDA clearance marks a significant milestone in Ibex’s journey and exemplifies our dedication to developing clinically validated solutions that help improve patient health outcomes,” said Joseph Mossel, CEO at Ibex Medical Analytics, in the news release.¹ “This clearance affirms our commitment to the safety and efficacy of our AI solutions and strengthens our ability to provide cutting-edge innovation to pathologists, which ultimately benefit patients. We hope this accomplishment will bolster industry-wide confidence that AI-powered digital pathology is ready for widespread clinical adoption.” 

Reference

1. Ibex Medical Analytics receives first FDA 510(k) clearance. News release. Ibex Medical Analytics. Published online and accessed February 10, 2025. https://ibex-ai.com/fda-510k-clearance/

2. Pantanowitz L, Quiroga-Garza GM, Bien L, et al. An artificial intelligence algorithm for prostate cancer diagnosis in whole slide images of core needle biopsies: a blinded clinical validation and deployment study. Lancet Digit Health. 2020;2(8):e407-e416. doi:10.1016/S2589-7500(20)30159-X