Galen Prostate

Histopathology prostate

AI second-read for prostate cancer detection and grading in core needle biopsies.

Regulatory

Regulatory status	510(k) cleared
FDA submission pathway	510(k)
FDA decision date	2025-01-24
FDA 510(k) number	K241232
CE marked	Yes

Validation summary

Study design	Retrospective
Cohort size	347 patients; 772 samples — FDA-cited Study 1 (missed-cancer detection): 347 cases initially diagnosed benign by SoC at 3 sites (2 US + 1 OUS), retrospective. FDA-cited Study 2 (assisted reader study): 772 cases/slides (376 negative + 396 positive) at 4 sites (3 US + 1 OUS) with 12 reader pathologists, retrospective with washout.
Number of validation sites	4
Site geography	Multi-center (international)
Comparator	Pathologist consensus
Primary endpoint	Sensitivity / specificity
Primary result	Study 1 (slide-level): Sensitivity 81.0% (95% CI 69.2%, 92.9%), Specificity 91.6% (95% CI 90.9%, 92.3%). Study 1 (case-level): Sensitivity 80.8% (95% CI 74.1%, 87.6%), Specificity 46.9% (95% CI 39.5%, 54.3%). Study 2 combined (12 pathologists): Sensitivity 93.9% with Galen vs 90.5% SoC, improvement +3.5% (95% CI 2.3%, 4.5%); Specificity 87.9% with Galen vs 91.1% SoC, difference -3.2% (95% CI -4.3%, -1.9%). Slides initially diagnosed benign by SoC (intended-use population): with Galen sensitivity 36.3% (95% CI 28.0%, 45.5%) vs 0% SoC; specificity 96.5% (95% CI 95.2%, 97.5%) vs 100% SoC.
External validation	Yes — Both K241232 clinical studies are independent of the algorithm's training data per the Decision Summary; cohorts were retrospectively collected and de-identified at 3 (Study 1) and 4 (Study 2) US/OUS sites. · Per-pathologist improvement in sensitivity ranged 0.0% to +11.6% across all 12 readers (Table 7).
Peer-reviewed	Yes
FDA decision summary	View on accessdata.fda.gov

Key publications

An artificial intelligence algorithm for prostate cancer diagnosis in whole slide images of core needle biopsies: a blinded clinical validation and deployment study
The Lancet Digital Health · 2020

Limitations noted

FDA Decision Summary notes (i) Study 1 demonstrated 'decrease in specificity of the Galen Second Read compared to the specificity of SoC … and it can be managed by mitigation measures such as use of additional stains'; (ii) Galen 'outputs are not intended to be used on a standalone basis for diagnosis, to rule out prostatic AdC or to preclude pathological assessment of WSIs according to the standard of care' (p. 2); (iii) only authorized with Philips IntelliSite Pathology Solution Ultra-Fast Scanner (PIPS UFS); (iv) intended for cases initially diagnosed as benign — not for general prostate biopsy reading.

Deployment

Available

Yes

Sources

https://ibex-ai.com/fda-510k-clearance/ · company · 2026-05-31

How to cite this entry

Elbakri A. OncologyAI Registry: Galen Prostate. Available at: https://oncologyairegistry.org/tools/ibex-galen-prostate.html. Accessed [date].