Abstract PS2-08-30: Circulating microRNAs in liquid biopsy for the molecular stratification of HER2+breast cancer subtype/title

Abstract Introduction: The heterogeneity of breast tumor tissue cells, when examined through gene expression-based molecular profiling, revealed the coexistence of multiple subtypes displaying receptor discordance, therapeutic resistance, and risk of relapse. An integrated phenogenomic screening, small RNA sequencing, and mutation-based clustering of samples collectively highlighted the presence of 10 tumor subtype combinations. The regulation of genomic alterations responsible for the emergence of these diverse subtypes is controlled by small non-coding RNAs known as microRNAs (miRNAs), which were detectable in circulation as well. The detection of miRNAs in minimally invasive liquid biopsy samples is advantageous as a predictor of long-term disease progression and in stratifying responses to neoadjuvant chemotherapy NAC, thereby enabling more effective management of therapeutic strategies. Aim: This study seeks to identify microRNA signatures from patient-derived blood samples (plasma) for the detection and precise stratification of HER2+ breast cancer subtype. Materials and Methods: In silico profiling of miRNA-sequencing dataset from The Sweden Cancerome Analysis Network-Breast (SCAN-B) study led to the selection of a panel of 35 tumor-specific miRNAs that were significantly upregulated p-value 0.05; fold change 2.0. Patient-derived primary n = 30 and normal plasma samples as healthy controls n = 5, retrospectively stored at the Biobank based at the Lambe Institute for Translational Research, Galway, were used for this study. The study received Ethics Committee approval from the CREC at University Hospital Galway (Galway C.A. 2073). Total RNA was isolated and miRNA enriched using Qiagen kits, reverse transcribed with the TaqMan™ advanced cDNA synthesis reagents, and expression quantified with TaqMan probes using the real-time PCR technique. Data normalization was done using the 2-ΔΔCT method, and the expression significance of miRNAs was analysed using XLSTAT software (version 2023.3.1). Binomial logistic regression models were generated using HER2 status Hercept score (3%) set as outcome and the average gene score of significant miRNAs as the potential determinant of HER2-enriched samples luminal B and HER2+ from the HER2− basal and Luminal A samples. Results and Discussion: From an initial panel of 35 miRNAs, 7 miRNAs proved significant and were exclusively expressed in primary HER2+ samples. The best fitting model was an average of three miRNAs hsa-miR-205-5p, hsa-miR-195-5p, and hsa-miR-4446-3p that had a specificity of 93.33%, a sensitivity of 85.71% and recorded an AUC of 0.971 (p0.05; CI- 0.370 – 3.852), indicating a high HER2+ subtype-specificity. This generated a probability (prob) prediction score, the median value used to group samples into probhigh and problow groups for predicting prognosis and disease-free survival outcome. Conclusions: Liquid biopsy microRNA-based testing will enable stratification and potentially create a method for therapeutic screening of breast cancer subtypes, through multiple longitudinal minimally invasive samplings before the need for a core needle biopsy. The miRNA signature outlined here provides a good diagnostic starting point for use in HER2+ cancers. Citation Format: V. Richard, K. Bruce, M. Kerin. Circulating microRNAs in liquid biopsy for the molecular stratification of HER2+breast cancer subtype 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 PS2-08-30.