
RlapsRisk ™ BC
Assess the risk of breast cancer relapse
Clinical context
Revolutionize breast cancer care management by democratizing access to actionable medical insights
Clinicians lack the tools to identify these high-risk patients at an early stage. Current methods either lack consistency in accuracy 4,5,6 or they are expensive and non-routine, 7 such as genomic sequencing.
Therapeutic and pathologist workflow
Integrating AI-powered diagnostics seamlessly
Subtype confirmed
ER+/HER2- invasive carcinoma of the breast
Surgery
Tumor excision
Pathologist
Selects slide of indication
WSI Review
Analysis of digitized slide
Owkin diagnostic
RlapsRisk BC
PDF report
Pathologist
Delivers final pathology report
Multidisciplinary board
Further testing / treatment decision
Testimonial
Incoming President of ESMO
Research and development
Timeline of milestones
Clinical variables C-index = .72
AI model C-index = .75
AI model integrating clinical variables C-index = .78
Image AI analysis outperforms clinical scores
RlapsRisk BC achieves 78% sensitivity and 80% specificity for post-treatment, time-dependent accuracy at 5 years, outperforming current clinical scores in practice. 1

Cumulative Sensitivity/Dynamic Specificity are natural extensions of sensitivity/specificity to the setting of time-to-event outcomes, such as the metastasis-free interval (MFI), understood as the time to distant relapse occurrence from initial surgery. In use here, as they easily accommodate time-dependent outcome status as well as right-censoring.
Validating product performance
When run on our training cohort, RlapsRisk BC’s performance demonstrates strong discrimination between risk groups, better informing oncologists on the risk classification of their patients to aid in treatment decisions.
Through blind validation on a separate cohort, RlapsRisk BC demonstrates that it is well calibrated with similar performances on data it was not trained on.
P-value < 0.01. Sample size N = 676
Citations
- Garberis IJ, Gaury V, Drubay D, et al. Blind validation of an AI-based tool for predicting distant relapse from breast cancer HES stained slides. Poster presented at: European Society for Medical Oncology (ESMO); May 9th - 13th 2022; Paris France.
- https://www.bcrf.org/breast-cancer-statistics-and-resources/
- Long-term hazard of recurrence in HER2+ breast cancer patients untreated with anti-HER2 therapy, Strasser-Weippl et al. 2015, BMC
- Gown AM. Current issues in ER and HER2 testing by IHC in breast cancer. Mod Pathol. 2008 May;21 Suppl 2:S8-S15. doi: 10.1038/modpathol.2008.34. PMID: 18437174.
- Casterá C, Bernet L. HER2 immunohistochemistry inter-observer reproducibility in 205 cases of invasive breast carcinoma additionally tested by ISH. Ann Diagn Pathol. 2020 Apr;45:151451. doi: 10.1016/j.anndiagpath.2019.151451. Epub 2019 Dec 17. PMID: 31955049.
- Polley MY, Leung SC, McShane LM, et al. An international Ki67 reproducibility study. J Natl Cancer Inst. 2013 Dec 18;105(24):1897-906. doi: 10.1093/jnci/djt306.
- Blok EJ, Bastiaannet E, van den Hout WB, et al. Systematic review of the clinical and economic value of gene expression profiles for invasive early breast cancer available in Europe. Cancer Treat Rev. 2018 Jan;62:74-90. doi: 10.1016/j.ctrv.2017.10.012.