Utility of immunohistochemical markers in irradiated breast tissue: an analysis of the role of myoepithelial markers, p53, and Ki-67.

Abstract

Radiation therapy is an important adjunct to breast-conserving surgery, but the diagnosis of recurrent/de novo carcinoma in a background of radiation atypia can be difficult, especially on small biopsies. Immunostaining for myoepithelial cell proteins is often used to assess invasion in nonirradiated breast tissue, yet these stains have not been investigated specifically in irradiated breast. We studied 29 irradiated breast resection specimens, some with carcinoma in situ (CIS, n=13) and/or invasive carcinoma (n=13). Representative blocks were stained for the myoepithelial proteins p63, smooth muscle myosin heavy chain (SMM), calponin, CK5/6, the proliferative marker Ki-67, and the tumor-suppressor p53. Nonirradiated control tissue was also stained with Ki-67 and p53 (CIS, normal, contralateral). Areas of radiation atypia/atrophy and nearly all CIS in irradiated breast tissue had abundant myoepithelial cells as evidenced by SMM, calponin, and p63 stains, with focal staining attenuation or gaps with SMM and calponin and frequently absent CK5/6 staining. As predicted, myoepithelial cell staining was absent in invasive carcinoma. p63 staining revealed postradiation myoepithelial nuclear morphologic changes. p53 staining was increased, although weak, in irradiated non-neoplastic breast (12% irradiated; 4% nonirradiated); however, irradiated CIS had less p53 staining when compared with control CIS (3% irradiated; 38% nonirradiated). As expected, Ki-67 was increased in carcinoma as compared with non-neoplastic irradiated tissue. Thus, myoepithelial immunostaining is a useful diagnostic adjunct in irradiated breast, with caveats similar to nonirradiated breast. Ki-67 may be helpful in some postradiation specimens; however, p53 staining is not reliable in this setting.