Abstract RF3-04: Tumor-informed circulating tumor DNA analysis to assess molecular residual disease for prognosis and prediction of benefit from palbociclib in the PALLAS trial

Abstract Background: PALLAS (PALbociclib CoLlaborative Adjuvant Study, NCT02513394) is a randomized phase III trial comparing two years of the CDK4/6 inhibitor palbociclib combined with the physician’s choice of adjuvant endocrine therapy, versus endocrine therapy alone for patients with Stage II-III hormone receptor-positive, HER2-negative (HR+/HER2-) breast cancer. Molecular residual disease (MRD) detection via circulating tumor DNA (ctDNA) testing was a predefined biomarker analysis to identify patients at highest risk of recurrence. Methods: PALLAS participants with available tumor block, at least a baseline (pre-study treatment) plasma sample, and either whole blood or buffy coat were randomly selected from the overall study population. Plasma samples on cycle 1 day 1 (C1D1), cycle 6 day 1 (C6D1), and at the end of treatment (EOT) were collected, processed, and stored. Nucleic acids were extracted from primary tumor tissue or, when primary tissue was insufficient, from post-neoadjuvant tumor tissue, provided the sample had sufficient material and tumor cellularity (>25 mm2 tissue area with ≥20% tumor nuclei). Data from next generation sequencing of tumor tissue and matched normal DNA were used to create a personalized ctDNA SignateraTM assay (Natera, Inc.) for MRD assessment in the peripheral blood. ctDNA was reported as mean tumor molecules/mL (MTM/mL) of plasma. The primary endpoint was distant recurrence-free interval (DRFI) based on ctDNA status. Cox Proportional Hazards models were used to evaluate prognosis and predictive interactions. To minimize bias, the biorepository and laboratory teams were blinded to participant identity. Results: In total, 1280 participants were randomly selected for inclusion in this preplanned ctDNA sub-study. WES and WGS are being conducted on tissue, and Signatera results will be generated. Results for association of MRD status (positive/negative) and MTM/mL with DRFI in the context of 7 year median follow up of this study cohort will be reported. Conclusions: The planned analysis is ongoing and the presented work will provide data from the first randomized phase 3 adjuvant study in HR+/HER2- breast cancer to report MRD status by treatment arm. Citation Format: H. A. Parsons, K. Ballman, E. Heitzer, M. Watson, M. Balic, D. Hlauschek, D. Renner, E. Kalashnikova, G. Steger, J. M. Balko, Y. Novik, M. Martin, A. A. Rodriguez, Z. Dayao, A. Chan, E. Nili Gal-Yam, M. C. Liu, C. Isaacs, M. Los, M. Gil Gil, B. Felder, C. Denkert, P. Fasching, F. Liu, T. O'Donnell, E. L. Mayer, M. Gnant, W. F. Symmans, A. DeMichele. Tumor-informed circulating tumor DNA analysis to assess molecular residual disease for prognosis and prediction of benefit from palbociclib in the PALLAS trial [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr RF3-04.