DCIS-MI is a rare histological subtype of breast carcinoma. The cells deemed to be invasive must be distributed either singly or as small groups in a non-organoid pattern having irregular shapes that are reminiscent of the conventional invasive carcinoma with no particular orientation [4
]. In the fourth edition of the European guidelines for quality assurance in breast cancer screening and diagnosis
published in 2006 [13
] and in the 2010 edition of the American Joint Committee on Cancer staging system [14
], the diagnosis of microinvasive carcinoma of the breast (T1mic) is applied to those invasive carcinomas with no focus measuring > 1 mm. The current staging manual states that microinvasive carcinoma is nearly always encountered in a setting of DCIS [or, less often, lobular carcinoma in situ (LCIS)] in which small foci of tumor cells have invaded through the basement membrane into the surrounding stroma [15
]. Although some authors have required that such invasive foci extend beyond terminal ductal lobular unit (TDLU) stroma into the interlobular tissue [16
], this is not supported by the ultrastructural confirmation of intralobular capillaries in close proximity to the delimiting fibroblastic layer of terminal ducts [17
]. Furthermore, DCIS often unfolds the TDLU externalizing the once intralobular stroma. Any distinctive feature of the TDLU stroma is often masked by an inflammatory infiltrate and/or stromal fibrosis. In the presence of DCIS, the TDLU stroma may not be distinguishable from the interlobular stroma [19
In this study, all the cases were classified within the context of a study and were classified by the same pathologists repeatedly. Pathologists made the pathological diagnosis mainly based on morphological characteristics and the immunohistochemical staining according to that diagnostic criteria and the pathological diagnosis conclusion was unaffected by the archived diagnoses which had already reported to the clinician. In order to minimize the diagnostic subjectivity, we asked five experienced pathologists who had received the same training and mastered the diagnostic criteria, to review all the available slides and asked them to classify double-blindly. If their diagnoses are inconsistent, they would discuss together and make the final diagnosis. In this way, we kept our diagnostic results reliable.
But various patterns in DCIS may be mis-interpreted as stromal invasion, including the cases where there is degenerative appearance of the dislodged tumour cells, chronic inflammatory reaction, crush artefacts, cautery effects, and distortion or entrapment of involved ducts or acini by fibrosis [6
]. As reported earlier [4
], it is difficult to determine whether in-situ breast carcinoma is associated with microinvasion, even with the help of immunohistochemistry techniques.
The absence of basement membrane material around nests of tumor cells defines the process as being invasive. Immunohistochemistry for basement membrane components (laminin and type IV collagen) is helpful in detecting the presence or absence of basement membrane. However, cells of invasive cancer can still synthesize components of basement membrane around invasive nests. Therefore, the use of basement membrane markers for the detection of stromal invasion is not recommended [4
The presence of MECs around nests of carcinoma cells defines the process as being in situ. Immunohistochemistry for P63, S100 and smooth-muscle myosin heavy chain (SMM-HC) for MECs has been used in determining whether a process represents in situ carcinoma or stromal invasion. But, those antibodies occasionally form an apparently discontinuous myoepithelial layer around nests of in situ lesions, and they also react with a small but significant subset of breast carcinoma tumor cells [4
]. While MECs are retained around ductal-lobular spaces containing DCIS, molecular studies have indicated that MECs surrounding with mammary ducts and lobular acini have important roles in the development and physiology of normal mammary glands, including maintenance of the basement membrane around ductal-lobular structures, providing a physical barrier between epithelial cells and the surrounding stroma, and maintaining epithelial cell polarity. Furthermore, experimental evidence has indicated that MECs produce factors that, through paracrine effects, to inhibit tumor growth, invasion, and angiogenesis [23
]. Recently, more and more attention has been paid to the potential role of the MECs in the progression of DCIS to invasive breast carcinoma. Though MECs that surround spaces involved by DCIS differ substantially from normal MECs in several respects [25
], Calponin is a contractile element that expressed in differentiated smooth muscle cells and is highly sensitive to normal noninvasive MECs and breast MECs [30
In this study, our diagnosis mainly based on morphological characteristics, while refering the expression of SMA, CD34 and FAP-α of stromal fibroblasts besides integrity of MECs. CD34 is a transmembrane glycoprotein expressed by haematopoietic stem cells, endothelial cells and mesenchymal cells in different tissues including breast that is thought to be involved in the modulation of cell adhesion and signal transduction. CD34+
fibrocytes/fibroblasts derive from myeloid precursors, besides its function as a matrix-production cell, it is a potent antigen-presenting cell and therefore it has been claimed that CD34+
may play a role in host response to tissue damage [31
]. Different studies have shown that the presence or absence of this population of cells might be useful in distinguishing benign from malignant lesions of the skin [34
] and gastrointestinal tract [35
]. The stroma of normal mammary gland contains many CD34 positive fibroblasts/fibrocytes and the presence of stromal positive CD34 fibroblasts has been shown to be associated with benign lesions [31
]. In malignant tumors it was noticed a loss of CD34 positive cells and gain of smooth muscle cell actin positive myofibroblasts [33
] and this is in keeping with our findings. In our study, stromal fibroblasts in the normal mammary tissues and UDH showed mainly immunophenotype of SMA-
. In the tumor-host interface of DCIS and IDC, stromal fibroblasts exhibited mainly immunophenotype of SMA+
. In the tumor-host interface of DCIS-MI, stromal fibroblasts exhibited SMA-
, however, in the tumor-host interface at the invasive front of DCIS-MI lesions, stromal fibroblasts exhibited mainly immunophenotype of SMA+
To the best of our knowledge, there is no study evaluating the role of FAP-α with reference to microinvasion of DCIS.
FAP-α is a cell surface glycoprotein belonging to the serine protease family, is expressed by the CAFs in over 90% of human epithelial cancers including breast, ovarian, bladder, colorectal, and lung cancers, but it is not expressed in epithelial cancer cells, normal fibroblasts, and other normal tissues except the transient expression in healing wounds [40
]. In this study, at the invasion front of all the DCIS-MI, stromal fibroblasts expressed FAP-α and there was significant statistical difference in the expression of FAP-α protein in groups 3 and 4, but no statistical difference was reported in the FAP-α protein expression for groups 4 and 5. FAP-α plays an important role in tumor growth and metastasis, as its expression on CAFs may create an environment permissive for cancer growth and invasion via collagenase and dipeptidyl peptidase activities [10
]. We suggested the possibility of FAP-α promoting the formation of microemboli that facilitates the metastasis of breast cancer. Thus, FAP-α can serve as a novel marker for pathologically determining whether DCIS has microinvasion.
Based on immunohistochemical staining, we found that some initally diagnosed DCIS-MI had discontinuously MECs around the so-called invasive foci without stromal reaction (SMA-CD34+FAP-α-). In such condition, we changed the original diagnosis of DCIS-MI to DCIS. In the same way, some of the DCIS cases were rediagnosed as DCIS-MI. In the study group 3, at the invasive front of the 18 re-examined DCIS-MI, 12 cases showed SMA positive, while 5 cases showed CD34 positive. 17 cases showed Calponin negative, and one Calponin negative case was still diagnosed as DCIS, but all of the 18 re-examined cases showed negative expression of FAP-α in the stromal fibroblasts. The sensitivity of SMA, CD34, Calponin and FAP-α in diagnosis of DCIS is 75%, 78%, 98% and 98%, respectively. In the study group 4, in the tumor-host interface of the 13 re-examined DCIS cases, 8 cases showed SMA positive, while 7 cases showed CD34 negative. 4 cases showed Calponin negative, and 12 cases showed FAP-α negative. The sensitivity of SMA, CD34, Calponin and FAP-α in diagnosis of DCIS-MI is 77%, 71%, 97% and 98%, respectively. In our study, using Calponin and FAP-α adjunctively improved the sensitivity of pathological diagnosis of DCIS by 13.6%, whereas the adjunctive use of Calponin and FAP-α improved the sensitivity of pathological diagnosis of DCIS-MI by 11.29%.
In conclusion, this study provides the first evidence that immunostaining with FAP-α and Calponin can serve as a novel marker for pathologically diagnosing whether DCIS has microinvasion. We also suggested the possibility of FAP-α promoting the formation of microemboli, which facilitate the metastasis of breast cancer.