Abstract P3-11-16: Personalized Neoadjuvant Therapy Guided by Drug Screening of Patient-Derived Tumor-Like Cell Clusters in HER2-Positive Breast Cancer: A Prospective Phase II Study and Exploratory Analysis

Abstract Background: HER2-positive breast cancer exhibits unique characteristics that can significantly influence patients’ response to neoadjuvant therapy. There is a critical need for a platform that facilitates in vitro drug screening and tailors personalized treatment regimens for such patients. Previously, we introduced an in vitro model named ‘patient-derived tumor-like cell clusters' (PTCs), a cell cluster including tumor cells, mesenchymal cells, and lymphocytes, which accurately replicates the structural, functional, and micro-environmental features of the original tumors. Our retrospective research has shown that PTCs could serve as a reliable preclinical model for drug screening, and thereby, to guide neoadjuvant therapy strategies for patients with breast cancer. This prospective, phase II, open-label study (ClinicalTrials.gov No. NCT04750122) aims to prospectively investigate efficacy of personalized neoadjuvant therapy guided by drug sensitivity profiles of PTCs in HER2-positive breast cancer and to reveal biomarkers associated with resistance to anti-HER2 targeted therapies. Methods: Tissues were obtained by core needle biopsy of candidate patients and cultured to generate PTCs, which were subsequently subjected to drug sensitivity testing and bulk RNA sequencing. Patients with HER2-positive breast cancer were assigned to neoadjuvant therapy according to their individual drug sensitivity profiles. The treatment regimens included THP (taxanes, trastuzumab, and pertuzumab), TCbHP (taxanes, carboplatin, trastuzumab, and pertuzumab), or treatment of physician’s choice (TPC). After completing neoadjuvant therapy, surgery was performed to evaluate the pCR rates of PTC-guided treatment (PGT) and TPC. For the PTCs subjected to RNA-seq, drug sensitivity testing was conducted using H, P, trastuzumab emtansine (T-DM1) and trastuzumab deruxtecan (T-Dxd), respectively. The gene expression data was then analyzed to identify potential biomarkers by differentially-expressed-gene analysis and to establish a predictive model for therapeutic efficacy by logistic regression analysis. Results: From April 2021 to July 2023, a total of 729 samples were collected by core needle biopsy from candidate patients, among whom 592 patients were diagnosed as invasive breast cancer. The success rate of generating PTCs through invasive breast cancer was 83.4%. Overall, 24 patients with HER2+ breast cancer that successfully constructed PTCs and drug sensitivity tests were enrolled in this study. Nineteen patients (79.2%) had T2-3 tumors and nineteen (79.2%) had node-positive disease. The in vitro drug sensitivity profiles suggested that 6 patients (25%) were sensitive to THP, 14 (58.3%) were sensitive to TCbHP, and 4 (16.7%) were resistant to both. Following neoadjuvant treatment, the patients that received THP treatment achieved a pCR rate of 83.3%, while the patients that were sensitive to TCbHP rather than THP achieved a pCR rate of 78.6%. Compared to only 1 out of the 4 patients with resistant PTCs achieved pCR, 16 of the 20 patients receiving PGT achieved pCR (80.0% vs 25.0%, p=0.027). In subgroup analyses of the PGT counterparts, the pCR rates were 62.5% for HR+/HER2+ patients and 91.7% for HR-/HER2+ patients. Besides, 14 PTC samples carried on bulk RNA sequencing. As a result, 5 co-upregulated genes (PLBD1, BMP3, HLA-DRB5, IGLV2-14, TIAF1) and 5 co-downregulated genes (EPN3, KCNK15, ACSF2, AC025048.2, PIP4K2B) were identified in the resistant group compared with the sensitive group. The multidrug resistance signature was utilized to establish a predictive model, of which the area under curves were 100.0% for predicting response to H, P, T-DM1, and T-Dxd. Conclusion: The in vitro drug screening PTC platform could serve as a promising approach for tailoring personalized neoadjuvant treatment strategies for HER2-positive breast cancer. The identification of multidrug resistance signature emerges as a practical and effective biomarker for forecasting the response to anti-HER2 targeted therapies. Citation Format: Yaqian Xu, Chaobin Wang, Xiangui Zhang, Shenyi Yin, Houpu Yang, Fei Xie, Yuan Peng, Yang Yang, Wei Du, Jianzhong Xi, Shu Wang. Personalized Neoadjuvant Therapy Guided by Drug Screening of Patient-Derived Tumor-Like Cell Clusters in HER2-Positive Breast Cancer: A Prospective Phase II Study and Exploratory Analysis [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr P3-11-16.