Cystosarcoma phylloides of the breast and its mimics. An immunohistochemical and ultrastructural study

Cystosarcoma phylloides of the breast is a tumor composed of breast ducts and a cellular stromal component that can be benign or malignant. The origin of the stromal cells is controversial. We undertook an immunohistochemical and ultrastructural study of 11 cases of cystosarcoma phylloides to assess the histogenesis of the stromal component. By light microscopy, 4 were diagnosed as benign, and 7 were diagnosed as malignant. Antibodies to vimentin, desmin, actin, high- and low-molecular-weight keratins, and S100 protein were used for immunohistochemical staining. In the 4 benign cases of cystosarcoma phylloides, the stromal cells stained positively only for vimentin. In the malignant tumors, the spindle cell component stained for vimentin in all the cases. In addition, the malignant stromal cells coexpressed desmin in two cases and keratin and S 100 protein in another case. By electron microscopy the stromal component in the benign case and in two of five malignant cases was composed of a mixture of fibroblasts and myofibroblasts. The entire neoplastic stroma in two other malignant cases showed features of smooth-muscle differentiation, whereas in another case all the stromal cells showed myoepithelial differentiation. Thus, in benign and malignant cystosarcoma phylloides, the stromal component consists of a mixture of fibroblasts and myofibroblasts. Leiomyosarcomas and myoepitheliomas can mimic malignant cystosarcoma phylloides, but immunohistochemistry and electron microscopy can differentiate these entities. This is important since their biologic behavior is different.

Stromal proliferations of the breast: an ultrastructural and immunohistochemical evaluation of cystosarcoma phyllodes, juvenile fibroadenoma, and fibroadenoma

The stromal cells of three cystosarcoma phyllodes, five typical fibroadenomas, and one juvenile fibroadenoma were studied by light and electron microscopy. Immunohistochemical staining for the S-100 protein also was performed on tissues from each of the three categories. On ultrastructural examination, cells comprising the three varieties of lesions were similar. Cells with fibroblastic features predominated in all cases. Myoid differentiation was present in two cases, one of cystosarcoma and one of fibroadenoma. Junctional complexes were present in the cystosarcomas but not in the fibroadenomas. Basal lamina was focally present around stromal cells in the cystosarcoma phyllodes but was not evident around cells of the typical fibroadenomas or the juvenile fibroadenoma. Stromal cells of the fibroadenomas and the cystosarcoma phyllodes did not stain for S-100 protein. The results support the hypothesis that the proliferating cells in all three tumor categories are similar and have features of fibroblasts. The lack of staining for S-100 protein would suggest an origin different from the myoepithelia. The latter conclusion, however, must be interpreted with a degree of reservation as we have shown that not all myoepithelial cells stain with certain monoclonal antibodies directed against the alpha and beta chain of S-100 protein.

Cystosarcoma phyllodes of the prostate. A pathologic and immunohistochemical study

We treated two patients who had lesions in the prostate with histologic features similar to those of cystosarcoma phyllodes of the breast. In one case, the stroma progressed to a clearly sarcomatous appearance, whereas the other tumor had a cellular stroma that was mitotically inactive. This element was immunoreactive for vimentin and desmin in both cases but was negative for epithelial markers. In contrast, the epithelial component was immunoreactive for prostate-specific antigen and epithelial membrane antigen. Following surgical resection, both patients were well two and three years later, without local recurrence or distant metastasis. The histogenesis of these tumors is unknown.

Estrogen and progesterone receptors in benign breast tumors and lesions: relationship with histological and cytological features

The pattern of estrogen (ER) and progesterone receptors (PR) and their relationship to histo- and cyto-pathological parameters has been studied in 97 cases of benign breast disease and benign phyllode tumors (95 women, of whom 76 were premenopausal, and 2 men). Total (cytosolic + nuclear) ER and PR were assayed by a single-saturating dose method using a tris-KCl buffer. The cut-off between positive and negative ER and PR assay was 100 femtomoles/g tissue. All specimens were processed for histological examination: epithelial and fibroblastic proliferation, epithelial/stromal ratio and presence of focal or diffuse hyalinosis. In 33% of the 46 cases of fibrocystic disease one receptor at least was present (13% ER+, 31% PR+). All the 8 cases in which infiltrating epitheliosis was present were PR+ and 4 of them were also ER+. In 72% of the 31 fibroadenomas one receptor at least was present (19% ER+, 71% PR+). In all these cases levels of receptors were lower than in malignant tumors. An inverse correlation between PR + prevalence and fibrohyalinosis was observed; on the other hand a positive relationship between PR + and fibroblastic (p less than 0.001) or epithelial (p less than 0.01) proliferation was found. In all 5 benign phyllode tumors examined PR + were present at a very high level, almost as high as in malignant tumors. Of the 15 other benign breast lesions, all but one (1 hamartoma) were ER- and PR-.

A malignant cystosarcoma phyllodes with positive estrogen receptor and its heterotransplantation into nude mice

We detected cytosol estrogen receptors in a malignant cystosarcoma phyllodes and succeeded in serial transplantations into athymic nude mice. The tumors of this transplantable strain (MC-3-JCK) have the same histologic features as the original tumors, and retain significant amounts of cytosol estrogen receptors. This strain appears to provide a useful experimental model for the study of biologic and therapeutic aspects of human cystosarcoma phyllodes of the breast.

Differentiation between metaplastic carcinomas and sarcomas of the human female breast by fibronectin

The distribution pattern of fibronectin in metaplastic carcinomas, stromal sarcomas, malignant cystosarcoma phyllodes tumours and histiocytic type lymphomas of the human female breast has been studied using the indirect immunoperoxidase technique on formalin fixed paraffin embedded tissue. Fibronectin was demonstrated as intensely stained strands between tumour cells forming an irregular network in metaplastic carcinomas and lymphomas. Stromal sarcomas and the malignant stromal component of the phyllodes tumours exhibited, in contrast, a uniform staining throughout tumour cells and stroma which was weaker than in adjacent normal-looking connective tissue. We suggest that the intense staining reaction of metaplastic carcinomas is due to the scirrhous reaction generally associated with invasive human breast carcinomas. The advantage of using fibronectin as a diagnostic tool in the differentiation of carcinoma/lymphoma versus sarcoma is the fact that the antigen is a stromal marker and its staining intensity is not influenced by the morphology or degree of differentiation of non-mesenchymal tumours.

The basement membrane component laminin in breast carcinomas and axillary lymph node metastases

The occurrence of the basement membrane (BM) glycoprotein laminin was investigated in 91 malignant breast tumors and in 359 axillary lymph nodes of which 70 contained metastases. The indirect double-layer immunoperoxidase technique was applied using rabbit anti-rat laminin. Positive cytoplasmic staining for laminin was found in all breast carcinomas. BM could be demonstrated around in situ carcinomas and in the higher differentiated carcinomas around islands of infiltrating tumor cells. Lymph node metastases resembled the primary tumors in the cytoplasmic staining reaction for laminin, but no BM staining was found. Tissue from 50 benign breast lesions showed cytoplasmic staining for laminin in both the secretory and the myoepithelial cells, but consistently weaker than in the carcinomas. BM was always present.

Histochemically demonstrable enzyme activities and their independence of the hormone receptor content in female breast carcinoma

Acid phosphatase, leucine aminopeptidase, monoamine oxidase and non specific esterase activities were histochemically demonstrated in specimens derived from 15 infiltrating ductal carcinomas of female breast. The relative areas occupied by the enzyme-positive carcinoma cells were visually estimated and, in the cases of leucine aminopeptidase, assessed morphometrically. All enzyme activities were found to be subject to major variations within a single carcinoma and between individual carcinomas, and the activity of any single enzyme was independent of that of three others. None of the enzyme activities correlated with the estrogen and progesterone receptor values, nor the histological grade of malignancy of the tumour. Thus, histochemically demonstrable enzyme activities seem to be of no use in predicting the hormone receptor content in infiltrating ductal carcinomas of the female breast.

Cystosarcoma phylloides. A steroid receptor and ultrastructure analysis

Six cases of cystosarcoma phylloides were evaluated by ultrastructure and steroid receptor analysis. Electron microscopy of the lesions supported previous reports of a heterogeneous tumor consisting of pleomorphic mesenchyme and normal or proliferative epithelium. In each case estrogen and progesterone receptor analysis indicated the presence of a nonsaturable estrogen and progesterone 4S binding protein rather than a specific steroid receptor as suggested by previous studies.

Steroid receptors in sarcomatous lesions

Clinical observations showed that the natural history of sarcomatous lesions may be different among males and females, and it may be influenced by hormonal factors. Estrogen receptors (ER) were measured on biopsy specimens of melanoma (two patients), soft-tissue sarcoma (four patients), cystosarcoma phylloides (five patients), benign breast tissues (27 patients), and breast carcinoma (109 patients). Thirty-four specimens also had progesterone receptors (PR) analyzed. One of the five cystosarcoma phylloides and five of the six nonmammary sarcoma tissues contained ER (mainly of the 4 Svedburg (S) variety) of more than 7 femtomoles (fmoles)/mg cytosol proteins (6/11 = 54%). For comparison three of the 14 fibroadenoma specimens and two of the 13 patients with other benign lesions had positive ERs (5/27 = 19%), whereas 56% of the breast carcinomas were ER positive. Since the amount of 8S ER found in sarcomatous tissues is relatively low, hormonal treatment would not be effective.

Reproducibility of measurements of oestrogen-receptor concentration in breast cancer

The reproducibility of measurements of oestrogen-receptor activity has been examined in multiple specimens from a rabbit uterus, a rat mammary tumour and human breast tumours. The relationship between receptor concentration and tumour histology has also been investigated in 11 large primary tumours. In the animal tissues, receptor measurements were relatively reproducible (coefficient of variance: wet wt. basis 16-17%, protein basis 16-21%) but in human breast tumours receptor activity varied considerably (c.v.: wet wt. basis, 22-125%; protein basis, 28-72%). In addition to these variations in receptor activity within tumours, there was a difference between tumours, as demonstrated by an analysis of variance (P less than 0.01). In the 11 primary breast cancers selected for study, the level of receptor activity was related to menopausal status and the tumour content of the specimen. We conclude that the receptor activity detected varies within a tumour and depends upon the tumour content of the biopsy specimen. Predictions based on precise quantitation of receptor concentrations may therefore necessitate replicate tumour sampling and correction for the fraction of non-tumour tissue in each sample.