ArteraAI Prostate Test

Histopathology prostate

AI-derived multimodal prognostic and predictive test for localized prostate cancer to guide therapy intensification. Originally offered as a laboratory-developed test; FDA De Novo authorized 2025-07-31 (DEN240068).

Regulatory

Regulatory status	De Novo authorized
FDA submission pathway	De Novo
FDA decision date	2025-07-31
LDT	Yes
NY CLEP	cleared
CAP-accredited lab	Yes
CLIA-certified lab	Yes

Validation summary

Study design	Retrospective
Cohort size	886 patients — patients (≥1 H&E biopsy slide containing the highest Gleason grade core per patient)
Number of validation sites	3
Site geography	Multi-center (US)
Comparator	Clinical outcomes
Primary endpoint	Time-to-event by risk category
Primary result	10-year risk of distant metastasis: ArteraAI High 28.1% (95% CI 19.4–37.5%), Intermediate 6.6% (3.6–10.8%), Low 3.3% (1.8–5.6%); overall 8.1% (6.1–10.4%). 10-year risk of prostate cancer-specific mortality: High 10.2% (4.7–18.2%), Intermediate 1.1% (0.2–3.7%), Low 0.6% (0.1–2.0%); overall 2.3% (1.2–3.8%). N=886 across 3 US sites.
External validation	Yes — Pivotal clinical validation cohort N=886 across 3 US sites, independent of the training data (NRG/RTOG phase 3 trials, N=10,009 across Canary-PASS, RTOG 0126, 0415, 0521, 9202, 9408, 9413, 9902, 9910, STAMPEDE, and Contemporary Biopsy Cohort A). · Subgroup-consistent prognostic separation across treatment groups (Active Surveillance n=314; Radiation Therapy n=203; Radical Prostatectomy n=354) and across African-American (n=72) vs. non-African-American (n=814) subgroups.
Peer-reviewed	Yes
FDA decision summary	View on accessdata.fda.gov

Key publications

Prostate cancer therapy personalization via multi-modal deep learning on randomized phase III clinical trials pivotal
npj Digital Medicine · 2022
Artificial Intelligence Predictive Model for Hormone Therapy Use in Prostate Cancer
NEJM Evidence · 2023

Limitations noted

FDA Decision Summary notes (i) the African-American subgroup was small (N=72) and 'should be taken in consideration when using the risk estimates'; (ii) erroneous results may lead to over- or under-treatment; (iii) the device 'should be used in conjunction with a complete standard of care evaluation' and is not a primary diagnosis; (iv) clinical validation was at three US sites only and is retrospective.

Deployment

Available

Yes

Sources

https://artera.ai/test/ · company · 2026-04-20

How to cite this entry

Elbakri A. OncologyAI Registry: ArteraAI Prostate Test. Available at: https://oncologyairegistry.org/tools/artera-prostate.html. Accessed [date].