Epigenetic silencing in non-neoplastic epithelia identifies E-cadherin (CDH1) as a target for chemoprevention of lobular neoplasia

Abstract

Invasive lobular carcinoma (ILC) of the breast is believed to develop from in situ lesions, atypical lobular hyperplasia (ALH), and lobular carcinoma in situ (LCIS). Down-regulation of the cell-cell adhesion protein E-cadherin is a defining feature of lobular breast cancer (LBC) and already occurs in ALH and LCIS. Apart from mutational mechanisms, epigenetic silencing of the E-cadherin gene (CDH1) is thought to be involved in E-cadherin down-regulation and has been observed at a high frequency in ILC. Whether CDH1 promoter methylation is already present in in situ lesions and thus contributes to the initiation of LBC is not established. We thus examined microdissected archived tissue from 20 LBCs by methylation-specific PCR to determine the CDH1 methylation status of lobular lesions. Nineteen of the 20 LBCs had a hypermethylated CDH1 promoter, including 13/14 ILCs and 13/13 ALHs or LCIS. Bisulphite sequencing indicated that methylation was complete within the investigated promoter fragment. Intriguingly, CDH1 methylation was likewise present in 8/8 adjacent non-neoplastic epithelia, but not in 6/6 mammary epithelia from healthy subjects. E-cadherin protein and mRNA were down-regulated in in situ lesions relative to adjacent epithelia. Together, these results indicate that CDH1 promoter methylation occurs in LBC prior to E-cadherin down-regulation and neoplastic formation. We thus propose that epigenetic silencing represents the first of the two hits required to silence both CDH1 alleles for LBC to develop. Because promoter methylation is in principle reversible, our findings suggest that chemoprevention of LBC by epigenetic drugs should be feasible. Furthermore, the presence of CDH1 methylation in pre-neoplastic epithelia suggests the existence of mammary regions with increased disease susceptibility, providing an explanation for the often multifocal presentation of LBC.