Summary: PreciseDx will present findings on their AI-driven breast cancer recurrence prediction test at the 2024 ASCO Annual Meeting.

Takeaways:

  1. PreciseDx’s AI-enabled PreciseBreast test aims to provide an objective risk score for early-stage breast cancer recurrence, improving upon traditional subjective grading methods.
  2. The study evaluates the test’s performance in predicting recurrence within a MammaPrint low-risk cohort from the Netherlands, using a comprehensive dataset of digitized pathology images and clinical data.
  3. The collaboration with Dordrecht Albert Schweitzer Hospital since 2022 has advanced the validation of the PreciseBreast test, demonstrating promising results to be shared at the ASCO meeting.

PreciseDx, a provider in oncology diagnostics leveraging Artificial Intelligence (AI) for morphology-driven disease analysis, announced its attendance and participation at the 2024 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago, IL.

PreciseDx Co-Founder and Chief Medical Officer, Michael J. Donovan, PhD, MD, will present the findings from the company’s ongoing collaboration with the Laboratory of Pathology, Dordrecht Albert Schweitzer Hospital in the Netherlands (NTH). The poster presentation, “AI-Enabled Digital Test to Predict Disease Recurrence for Patients with Early-Stage Invasive Breast Cancer Demonstrates Good Performance in a MammaPrint Low Risk Cohort from the Netherlands with a Median 6-Year Follow-Up,” will be held on June 1, 2024.

Improved Breast Cancer Grading

Traditional invasive breast cancer (IBC) grading remains limited due to diagnostic subjectivity. PreciseDx has developed PreciseBreast, a test validated to predict the risk of recurrence in early-stage breast cancer patients. The test leverages Artificial Intelligence (AI), morphological features, and clinical data to provide an objective risk score. This study evaluates the performance of PreciseBreast as compared to MammaPrint in a cohort of patients with outcome data.

About the New Study

In this study, a MammaPrint cohort with a median 6-year follow-up was examined. Hematoxylin and Eosin (H&E) stained images, digitized at Philips in Eindhoven, NTH, were coupled with clinical data from pathology and Netherlands cancer registries. This comprehensive dataset encompassed demographics, pathology results, MammaPrint risk classification, treatment details, and recurrence events. The validated PreciseBreast (PDxBr) model was then assessed on this MammaPrint cohort using key performance indicators such as AUC (area under the curve)/concordance index, NPV (negative predictive value), PPV (positive predictive value), hazards ratio (HR), sensitivity, and specificity to evaluate its efficacy.

“We initiated this collaboration with the Dordrecht Albert Schweitzer Hospital in 2022 and have since made considerable progress in further validating the efficacy of our PreciseBreast test and overall phenotyping platform,” says Donovan. “We look forward to sharing the results of our ongoing study with our colleagues and peers at the 2024 ASCO Annual Meeting.”