Nuclear EGFR in ductal invasive breast cancer: correlation with cyclin-D1 and prognosis

The epidermal growth factor receptor (EGFR)-family and cyclin-D1 have been extensively studied in breast cancer; however systematic studies that examine protein expression and gene status in the same cohort of patients are lacking. Also emerging evidences suggest existence of a direct EGFR-signaling pathway, which involves cellular transport of EGFR from cell membrane to the nucleus, and transcriptional regulation of the target genes. Thus, we examined the protein expression of membrane EGFR, nuclear EGFR, cyclin-D1 and the corresponding gene status in 113 breast carcinomas by immunohistochemistry and fluorescence in situ hybridization using tissue microarrays. Membrane EGFR overexpression and EGFR gene amplification were detected in 2% cases, while nuclear EGFR was detected in 40% of cases, with 12% having high nuclear EGFR staining. Nuclear EGFR correlated with tumor size (P=0.0005), lymph node metastasis (P=0.0288), Nottingham prognostic index (P=0.0011) and estrogen receptor (ER) expression (P=0.0258) but the letter correlation was observed only in premenopausal group of patients. Strong cyclin-D1 expression and cyclin-D1 gene (CCND1) amplification were found in 64 and 13% of the cases, respectively. Cyclin-D1 expression showed positive correlation with ER (P=0.0113) and inverse correlation with Nottingham prognostic index (P=0.0309) and membrane EGFR (P=0.0201). CCND1 amplification also showed inverse correlation with membrane EGFR (P=0.0420). A strong correlation between membrane EGFR expression and gene amplification (P=0.0035), as well as cyclin-D1 overexpression and gene amplification (P=0.0362), was demonstrated. On univariate analysis cyclin-D1 expression showed a correlation with longer overall survival in the premenopausal group and nuclear EGFR correlated with shorter overall survival in whole cohort as well in the premenopausal group of patients. Multivariate analysis revealed nuclear EGFR to be an independent prognostic factor and showed 3.4 times greater mortality risk for nuclear EGFR+++ patients as compared with nuclear EGFR negative patients (hazard ratio =3.402; P=0.0026).