Epithelial-mesenchymal transition has been reported to be an etiological factor in metaplastic carcinoma [7
]. The overt carcinoma cells of almost all of the MPC breast cancer cases reported in Japan were positive for both epithelial cell markers and mesenchymal cell markers (Table ). Electron microscope findings and the results of immunohistochemical studies were reported to indicate that MPC is of myoepithelial cell origin [1
]. On the other hand, Okuyama et al. examined specimens from 8 patients and reported that the overt carcinoma cells of all of those cases were negative for myoepithelial cell markers [3
]. Moreover, Only 4 of the 27 patients in Japan was positive for myoepithelial cell markers. In the patient we have described, as well, the overt carcinoma cells were positive for vimentin, S-100 protein and cytokeratins (AE1/AE3, CK7, CK8 and CK19). They showed negative staining for α-SMA, p63, CK5/6 and GFAP, which are myoepithelial cell markers. p63 has been reported to be useful as diagnostic marker for metaplastic carcinoma [9
]. It was reported that the carcinoma cells with spindle and/or squamous differentiation in metaplastic carcinoma showed positive staining for p63. The malignant component with no squamous or sarcomatous differentiation in MPC of our patient might cause negative staining for p63.
Sources, dilution and pretreatment of antibodies used
Our patient's MPC exhibited the same cell marker profile as that reported by Okuyama et al
., showing the properties of both epithelial cells and mesenchymal cells. It was reported that the results of immunohistochemistry differed between the peripheral epithelial area and the central myxoedematous area. In the central myxoedematous area, which can be thought to be causing metaplasia, down-regulation of epithelial markers and up-regulation of mesenchymal markers were observed [1
]. On the contrary, for our patient, the peripheral epithelial area and the central my edematous area showed no differences in their staining profiles. Recent molecular studies have shown the monoclonal origin of carcinosarcoma of the breast, as the carcinomatous and sarcomatous elements share common genetic alterations [12
]. These observations support the hypothesis that the matrix-producing carcinoma may be derived from a single totipotent stem cell.
Our patient had triple-negative breast cancer with regard to ER, PgR and Her2. In addition, it is interesting that almost all of the reported Japanese cases of MPC of the breast were triple-negative. It is said that most cases of metaplastic carcinoma are also triple-negative breast cancer [14
], and this is important in terms of elucidating the etiology of the metaplastic change. Ninomiya et al
., reported that one of their two cases of MPC of the breast was the basal phenotype [5
]. McCarthy et al
., generated a conditional mouse model of BRCA1
deficiency. Mammary tumors that developed in these mice had basal and metaplastic characteristics in the form of spindle cell and squamous cell differentiation. Most of the tumors were negative for ER, PgR and Her2 [6
]. Additionally, a recent report has shown that epithelial mesenchymal transition-like changes occurred preferentially in the basal-like subtype of breast carcinomas [16
]. Furthermore, subpopulations of cancer cells with stem cell properties are especially frequent within basal-like breast cell lines [17
]. Stem cell-like breast cell lines are also able to undergo epithelial mesenchymal transition [18
]. These data suggest that basal-like cancer cells may undergo epithelial mesenchymal transition with intrinsic phenotype of cancer stem cells. Although most cases of triple-negative breast cancer have the basal-like phenotype [6
], MPC of our patient had lack of any markers for myoepithelial cell type and basal-like cell type. BRCA1
has been shown to play an important role in mammary differentiation and the loss of BRCA1
function resulted in the accumulation of cells expressing the stem/progenitor marker ALDH-1 [19
]. Although the BRCA1
status of our patient has not been identified, it was suggested the possibility that the tumor cells of our MPC might be blocked differentiation with expansion of undifferentiated cell compartment.
Okuyama et al
., reported that the incidence of MPC of the breast was 0.05% in Japan [4
]. Our search of the main Japanese medical journals found a total of 27 cases of MPC of the breast reported in Japan to date, including our present patient [3
]. Imaging diagnosis by contrast-enhanced CT and contrast-enhanced MRI have revealed that this disease is characterized by a ring structure in its periphery. For that reason, it was concluded that it is necessary to consider the possibility of MPC of the breast when such image findings are obtained [3
]. In our present patient, as well, contrast-enhanced CT revealed an irregularly shaped, 2.5 cm tumor showing peripheral ring-shaped contrast enhancement.
Most MPC of the breast are triple-negative, and postoperative adjuvant chemotherapy is often administered [13
]. However, some studies have shown this therapy to have been ineffective, and further research on this issue is warranted.
The prognosis of MPC of the breast is said to be better than that of other carcinomas that are accompanied by osteocartilaginous metaplasia [20
]. Wargotz et al
., reported a 5-year survival rate of 68% for MPC of the breast [1
], but the number of reported cases has been small and the prognosis thus remains unclear. Our patient refused postoperative adjuvant chemotherapy, and distant metastasis was detected at 10 months after the partial mastectomy. In the future it will be necessary to study a larger number of patients with MPC of the breast and further elucidate the clinicopathological characteristics of this malignancy.