Clinical Validation of Image-Based AI Predictive Biomarkers for Precision Neoadjuvant Triple-Negative Breast Cancer Treatment (PEAR-TNBC).

TPS627 Background: Triple-negative breast cancer (TNBC) has the worst prognosis of all breast cancer subtypes, with a 5-year recurrence rate of 30% in patients with localized disease. Pathological Complete Response (pCR) strongly predicts the risk of recurrence and death, which are reduced by approximately 80% in patients who achieve a pCR following neoadjuvant treatment. There are multiple treatment regimens available for TNBC patients in the neoadjuvant setting but no biomarkers to match patients to treatments. Giving ineffective treatment leads to worse outcomes for patients and additional burden on healthcare systems. Pear Bio has developed a functional precision medicine test coupling an organ-on-a-chip with a computer vision (CV) pipeline through which an individual patient’s response to different treatments can be monitored simultaneously ex vivo. Each microtumor’s response is monitored using time-lapse 3D microscopy. A proprietary CV pipeline quantifies the responses (e.g. tumor cell viability, culture size). Lab test responses are correlated to the clinical outcomes of the patients to identify biomarkers with high predictive accuracy. This generates multiple CV-based biomarkers with potential predictive value. PEAR-TNBC (NCT05435352) is a prospective, observational UK-wide trial aimed at assessing the test’s accuracy in predicting pCR in TNBC patients in the neoadjuvant setting. The study recruits patients with early TNBC (N=54) who are willing to undergo an additional core needle biopsy prior to starting neoadjuvant therapy. The assay is run in parallel with the patient’s treatment and the treating oncologist is blind to the results of the Pear Bio test. Methods: The study has 2 cohorts (A and B), each recruiting at least 20 evaluable patients. To be eligible, patients need to have ≥10mm tumor, stage I-III, TNBC, with planned neoadjuvant treatment followed by surgery and be willing to undergo a study-specific core needle biopsy procedure. Cohort B also requires a 40mL blood sample. To be evaluable, eligible patients’ samples must be successfully cultured in the 3D ex vivo assay and they must complete at least 4 cycles of neoadjuvant chemotherapy prior to surgery. Cohort A recruitment began on 05/27/22 and has recruited 20 evaluable patients, through 5 sites. Cohort B accrual will begin upon completion of Cohort A. The primary endpoint is the accuracy of the Pear test in predicting pCR, defined as specificity, with secondary outcomes as sensitivity, and positive and negative predictive values. Novel aspects include a focus on specificity (i.e. accurately predicting patients who will not achieve pCR) combined with the use of Receiver-operating characteristic curves to define optimal cutoffs for the CV biomarkers. Future trials will aim to run the test prior to therapy selection, guiding the choice of regimen to assess the test’s ability to increase pCR rates. Clinical trial information: NCT05435352 .