Abstract PS3-12-09: Spatial immune profiling and pathological response after neoadjuvant therapy in early HER2-positive breast cancer

Abstract Background: HER2-positive (HER2+) breast cancer is a heterogeneous disease, with variability in tumor architecture, immune infiltration, and protein expression within both the tumor (TU) and the surrounding tumor microenvironment (TME). A better understanding of how the TME influences treatment response is critical. This study investigates whether protein expression levels and intratumoral heterogeneity are associated with pathological complete response (pCR) in patients with HER2+ early breast cancer (EBC) treated with neoadjuvant therapy. Methods: A total of 54 HER2+ EBC core biopsies from patients treated with neoadjuvant HER2-targeted chemotherapy at Hospital Clinic of Barcelona (April 2022 – September 2023) were analyzed using GeoMx Digital Spatial Profiling. Protein expression was measured in the immune (CD45+) and tumor (PanCK+) compartments. For each sample, a minimum of three TU and TME regions of interest (ROIs) were selected to ensure adequate spatial coverage and capture intratumoral heterogeneity. Protein expression data were batch-corrected; for each ROI we computed the mean, standard deviation (SD), and coefficient of variation (CV). CV, as a relative metric, served as a proxy for intraregional heterogeneity, while the mean reflected protein expression. In parallel, the HER2DX genomic test was performed on bulk RNA from the same biopsy to evaluate transcriptomic correlates; PAM50 was also assessed. Stromal tumor-infiltrating lymphocytes (TILs) were assessed according to international guidelines. Associations between protein metrics, HER2DX scores, TILs, and pCR were tested using Wilcoxon, Kruskal–Wallis, and logistic regression in RStudio. Results: Among the 54 HER2+ cases, the median age was 58 years, and the pCR rate was 51.9%. Clinical and biological features showed general concordance with those reported in a HER2+ neoadjuvant cohort, including PAM50 subtypes (11.3% Basal-like, 43.4% HER2-Enriched, 18.9% Luminal A, 20.8% Luminal B, 5.7% Normal-like), menopausal status (66.7% postmenopausal), hormone receptor (HR) status (72.2% positive), and HER2DX pCR score (31.4% high, 27.5% medium, 41.2% low). Higher HER2 mean expression in TU ROIs (Odds ratio [OR] = 1.54; p = 0.008), but not HER2 CV (p = 0.327), was associated with pCR. In TME ROIs, increased CVs of T cell markers, CD3 (OR = 1.30; p = 0.020), CD4 (OR = 1.40; p = 0.010), CD8 (OR = 1.24; p = 0.040), and the plasma cell marker CD27 (OR = 1.24; p = 0.030), were also significantly associated with a higher probability of pCR, whereas their mean expression levels were not; while T cell exhaustion TIM3 marker mean expression in TME ROIs was inversely associated with pCR (OR = 0.23; p = 0.042). TILs levels, as quantitative and qualitative variable, were not significantly associated with pCR in this cohort (OR = 1.01, p = 0.300). Finally, the HER2DX pCR score was significantly associated with pCR status, both when analyzed as a continuous variable (OR 1.90 per 10-unit increase; 95% CI = 1.34–2.77; p