Owkin Dx vs Stratipath
Both read routine H and E slides to stratify breast cancer risk, both are CE marked with no FDA clearance, and both pitch themselves against expensive molecular multigene assays, so the decision turns on evidence quality and regulatory cleanliness. If your failure mode is adopting a prognostic whose validation does not generalize, start with Stratipath, which has the stronger peer reviewed evidence, a 2,719 patient validation across two independent cohorts. If you want a tool whose value is easy to model against the molecular test it displaces, or you also need colorectal MSI pre screening, start with Owkin Dx, but read its enrichment caveat and its older IVDD status. Owkin covers two indications and reports an honest adjunct effect size; Stratipath is single indication but better validated.
- Two indications, not one: breast relapse risk with RlapsRisk BC plus colorectal MSI pre screening with MSIntuit CRC, both first in class CE-IVD in Europe, versus Stratipath's single breast indication.
- The most legible commercial case in the prognostic group: both products define themselves against a priced comparator, gene expression tests and confirmatory MSI testing, so value models against a known line item.
- Honest adjunct reporting: the breast model's contribution is published as a concordance index of 0.81 combined with clinico pathological factors against 0.76 for those factors alone, and the company discloses its cohort enrichment openly.
- The stronger evidence base: a peer reviewed retrospective validation across 2,719 primary breast cancer patients in two independent Swedish cohorts, testing the generalization prognostic models most often fail.
- Cleaner regulatory footing: described in the literature as the first CE-IVD AI for primary breast cancer risk stratification in routine use, marked under IVDR rather than the older IVDD directive Owkin's breast product carries.
- Lower adoption friction: distributed inside PathAI's AISight since 2025, so for laboratories on that platform adoption is algorithm selection.
Side-by-Side
| Axis | O Owkin Dx |
S Stratipath |
|---|---|---|
| AI Centrality | ||
| Autonomy and Oversight Model | ||
| Model and Technology Transparency | ||
| Clinical and Operational Evidence | ||
| AI Safety and PHI Stewardship | ||
| HIPAA and BAA Posture | ||
| Security Certifications and Trust Center | ||
| FDA and Regulatory Status | ||
| AI Governance and Bias Disclosure | ||
| EHR and Interoperability Depth | ||
| Deployment Model and Data Residency | ||
| Commercial Transparency | ||
| Setting and Specialty Coverage |
Both are CE-IVD only with no FDA clearance, both position against molecular multigene assays, and neither publishes pricing, a security attestation, or demographic subgroup analysis, which carries extra weight for a prognostic because a risk score that behaves differently across populations translates directly into differences in treatment intensity. Owkin's breast validation cohort was enriched with relapse patients and confined to endocrine therapy only cases, which the company discloses and which inflates apparent discrimination relative to an unselected population; independent validation with Cerba Path is underway, and its RlapsRisk BC is indexed under the older IVDD General mark rather than IVDR. Stratipath's evidence, though peer reviewed, is retrospective and drawn only from Swedish cohorts, so generalization remains open. One design nuance on Owkin's colorectal product: MSIntuit is a rule out placed upstream of confirmatory MSI testing, so a false negative removes a patient from that testing entirely.