In Vivo Insights into EGFR-Targeted Breast Cancer Therapy: From Mechanism to Efficacy
Keywords:
EGFR inhibition, breast cancer models, xenograft tumors, DMBA induction, AKT/MAPK signaling, VEGF expression, AI histopathology, ImageXpertAbstract
Background Breast cancer is a heterogeneous malignancy with epidermal growth factor receptor (EGFR) overexpression implicated in up to 70% of cases, particularly triple-negative subtypes, driving proliferation, invasion, and angiogenesis through downstream AKT, MAPK, and VEGF pathways. While molecular studies underscore EGFR as a viable target, discrepancies in preclinical validation have hindered clinical translation of inhibitors like gefitinib.
Objective This study translates molecular mechanisms of EGFR signalling into in vivo efficacy assessments using complementary preclinical models, evaluating tumour regression, biomarker modulation, and histopathological changes augmented by artificial intelligence (AI)-based analysis.
Methods Athymic nude mice bearing MDA-MB-468 xenografts and Sprague-Dawley rats with 7,12-dimethylbenz[a]anthracene (DMBA)-induced mammary tumours (n=10/group) were treated with gefitinib (50 mg/kg, oral, 4 weeks) or vehicle. Tumor volumes were measured biweekly via calipers. Post-treatment tissues underwent immunohistochemistry (IHC) and Western blot for phosphorylated AKT (p-AKT), MAPK (p-MAPK), and VEGF. ImageXpert AI software quantified necrosis, proliferation (Ki-67), and biomarker intensity from H&E and IHC slides, validated against manual pathologist scoring.
Results Gefitinib induced significant tumor regression: 62% volume reduction in xenografts (p<0.001 vs. control) and 48% in DMBA models (p<0.01). Pathway inhibition was evident with 68% decrease in p-AKT, 52% in p-MAPK, and 42% in VEGF expression (all p<0.05). AI analysis demonstrated high concordance with manual reads (Pearson's r=0.91 for necrosis; r=0.88 for Ki-67), identifying 35% greater necrosis in treated groups and subtype-specific vascular remodelling.
