Effects of previous therapy An increased proportion of patients now have breast conserving therapy rather than mastectomy for first line surgical treatment of breast carcinoma. This is frequently followed by radiotherapy to the breast. Both pre- and post operative chemotherapy may also be utilised in different treatment regimes of patients with primary breast cancer. Fine needle aspiration cytology (FNAC) specimens or biopsies of clinical or mammographic abnormalities in the residual breast may thus show changes due to irradiation or systemic therapies which may mimic residual, recurrent or second malignancies. It is therefore vital that adequate clinical information is given about previous treatments when any specimen is sent to the laboratory. Second malignancies, both mammary and non-mammary in origin are also seen in patients treated for breast cancer and these must be considered when new clinical or radiological lesions are identified. Within the breast parenchyma both the stromal component and the epithelium may show radiation changes. Stromal fibrosis with reactive 'radiation fibroblasts', as seen elsewhere in the body, can be identified. The nuclei of these fibroblasts may be variable in size and hyperchromatic or hypochromatic with 'smudged' outlines; nucleoli may also be visible. 35 The fibrosis may be extensive and a severe reaction is a particular complication of radiotherapy in patients with underlying connective tissue diseases. Elastosis is also seen in association with radiotherapy but this is a non-specific feature which may be present in benign breast conditions and after surgery alone and is thus of little diagnostic help. The most characteristic radiation changes, however, involve the terminal duct lobular units where atypical epithelial cells are present within the surrounding lobular stroma which is usually hyalinized and sclerotic. 36 The epithelial changes are seen in all irradiated patients but show considerable variation in severity, apparently unrelated to radiation dose. Nuclei are hyperchromatic and enlarged but nucleoli are inconspicuous and chromatin is homogeneous without the 'clumping' seen in malignancy. The cytoplasm of the cells is often vacuolated. Myoepithelial cells are reported to be unaffected by radiotherapy. The epithelial atypia described may mimic residual or recurrent carcinoma. However, although the epithelial cells are plump and may appear to fill the terminal duct lobular units, the latter are not significantly distended and a lack of stratification and mitotic activity may be helpful in determining that the features are due to irradiation. Neither the monotony of the regular cells of lobular carcinoma in situ nor the pleomorphism, conspicuous nucleoli, mitoses and loss of polarity of ductal carcinoma in situ (DCIS) within lobules ('cancerization of lobules') is present. Although epithelial atypia may be seen in larger ducts, as reported with chemotherapy, this is always accompanied by the more characteristic changes in the terminal duct lobular unit. Necrosis is not usually identified. The epithelial features associated with radiotherapy may also be seen in foci of sclerosing adenosis thus making histological interpretation more difficult. The overall architectural pattern on low power examination is then not that of carcinomatous infiltration but remains lobulo-centric and this is a useful discriminatory feature. To diagnose residual or recurrent carcinoma in these circumstances may be a particular problem in frozen section material and a definite diagnosis may not be possible on such preparations. In general frozen section requests should be discouraged when small foci of recurrent carcinomas are suspected in patients who have had radiotherapy. It is worth noting however that recurrent breast carcinomas are usually similar in appearance to the original tumour and review of the previous reports and histological slides may be invaluable. The lack of the architectural assistance gained from histological examination of an excision biopsy specimen may make the assessment of needle core and cytological preparations from the irradiated breast very difficult. It has been reported that the cytological features in particular may be indistinguishable from malignancy. Caution is required in the interpretation of cytological atypia in FNAC from irradiated tissues 37 and the cellularity of the sample is more useful diagnostically; paucicellular samples are uncommon from foci of recurrent or residual malignancy but are usual from the non-neoplastic irradiated breast. If scanty highly atypical cells are seen in a smear it may be necessary to issue a C4 - 'suspicious' rather than a C5 - 'malignant' report and the same is true for needle core biopsies. The tissue changes present in residual breast carcinoma cells after radiotherapy are essentially similar in nature to those seen in non-neoplastic breast tissue. Cytoplasmic vacuolation may be identified and carcinoma cell nuclei may show marked pleomorphism with the presence of bizarre forms and extensive tumour necrosis may be encountered. As described above the fibrosis of the stroma may lead to difficulties in distinguishing small foci of carcinoma cells which have become entrapped from radiation changes in sclerosing adenosis and non-neoplastic conditions. In these circumstances immunohistochemistry with epithelial markers and smooth muscle actin may be invaluable; the epithelial nature of the cells can be confirmed but smooth muscle actin immunoreactivity will be seen surrounding entrapped benign tubules. Chemotherapy in the residual non-neoplastic breast causes changes similar to those in the irradiated organ. Atrophy of epithelial structures is seen in lobules with a variable degree of intralobular fibrosis. 38 Indeed lobular atrophy occurs in up to 65% of cases and is seen in both pre- and post-menopausal women. 39 Bland fibrous obliteration of acini may occur and fibrous or fatty involution of the stroma elsewhere in the breast may be identified, possibly related to suppression of ovarian function. Epithelial atypia is present in the lobules in approximately 30% to 50% of cases but, unlike irradiation change, is equally common in ducts where this feature may be seen in 40-50% of specimens. 38, 39 Epithelial cells may be enlarged with prominent nuclei, small nucleoli, thick nuclear membranes and pale vacuolated cytoplasm, essentially similar to the features of irradiation in breast tissue. The nuclear membrane remains well-defined and hyperchromasia is less prominent than in residual or recurrent carcinoma cells. Uncommonly, necrosis and consequent calcification may occur within the duct lumen and can thus produce a suspicious mammographic appearance necessitating FNAC or biopsy. Residual carcinomatous foci may show similar histopathological changes as a result of chemotherapy to those described following irradiation. These features may be degenerative and transient. 39 It has been reported, however, that malignant cells can be mistaken for histiocytes with increased quantities of vacuolated eosinophilic cytoplasm and eccentric, sometimes hyperchromatic, vesicular nuclei. 38 Vacuolization of the cytoplasm is seen in approximately 60% of cases and may also be seen in metastatic deposits within lymph nodes. Chromatin clumping and prominent nucleoli may be seen and multinucleated forms identified. Significant difficulties in diagnosis can occur when small foci of residual or recurrent carcinoma are admixed with genuine histiocytes and foreign body type giant cells adjacent to the site of a previous biopsy. Although some authors have reported that carcinoma cells retain an overtly malignant appearance after chemotherapy we believe that in difficult cases immunohistochemical profiles may be invaluable; carcinoma cells retain cytokeratin and epithelial membrane antigen (EMA) expression post chemotherapy despite the alteration in morphological appearance. It is of interest to note that infiltrating lobular carcinomas are said not to show the cytological effects of chemotherapy seen in other types of breast carcinoma. 40 Fibrous scarring after surgery is seen in about one quarter of women and may mimic residual or recurrent carcinoma clinically and mammographically. Biopsy may be required to exclude malignancy although some authors suggest that fine needle aspiration cytology is cost effective and reliable in distinguishing scar tissue from carcinomatous foci within scar lesions. 41 As with radiotherapy, FNAC samples from residual carcinomatous foci are usually cellular with overtly malignant cells compared with smears from areas of fibrous scarring which are almost acellular, usually with a predominance of macrophages. Mammographic calcifications may be identified in patients who have had wide local excision of a primary breast carcinoma with subsequent radiotherapy. A high index of suspicion should be retained for new calcifications in the post-irradiated breast which are more often associated with residual or recurrent carcinoma than non-neoplastic irradiation changes. 42 |
Copyright 2000, The Author(s) and/or The Publisher(s)
| Organisation: FORPATH asbl  |
Coordination: Dr Bernard Van den Heule |
Host: Labo CMP |