Intracystic Papillary Carcinomas of the Breast A Reevaluation

Intracystic Papillary Carcinomas of the Breast: A Reevaluation Using a Panel of Myoepithelial Cell Markers Laura C. Collins, MD*, Victor P. Carlo, MD*, Harry Hwang, MD﹟ , Todd MD S. Barry, MD﹟ , Allen M. Gown, MD﹟ , and Stuart J. Schnitt, MD* MD MD From the *Department of Pathology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA; and ﹟ Pheno. Path Laboratories, Seattle, WA. Am J Surg Pathol 2006; 30: 1002– 1007 TMUH, Intern, Wang Chieh Nan

Introduction (1) • Histologic features useful for distinguishing • • between benign and malignant papillary lesions of the breast. However, the correct categorization of papillary lesions continues to be one of the most challenging areas in breast pathology. benign intraductal papilloma, papilloma with atypia (atypical papilloma), and papillary ductal carcinoma in situ (DCIS)

Introduction (2) • Immunohistochemical identification of • myoepithelial cells (MECs) and their distribution Prior studies have shown that benign papillomas consistently exhibit a basal layer of MEC throughout the papillary fronds, whereas in papillary carcinomas such a MEC layer is either completely absent or markedly reduced.

Introduction (3) • Intracystic papillary carcinomas (IPC) of the breast have traditionally been considered to be variants of DCIS characterized by the presence of papillary carcinoma within a cystically dilated duct. • Recent evidence has raised the possibility that not all lesions categorized histologically as IPC are in situ carcinomas.

Introdction (4) • We evaluated the presence and distribution of MEC at the periphery of the nodules in a series of IPC and, for comparison, a series of benign intraductal papillomas, using immunostaining for 5 highly sensitive MEC markers.

Materials and Methods (1) • 22 cases of IPC from the surgical pathology • • archives and consultation files of the Pathology Department of Beth Israel Deaconess Medical Center (BIDMC), Boston, MA and Pheno. Path Laboratories, Seattle, WA. Age from 46 to 91 years (mean=69 y; median=66 y) the tumors varied in size from 0. 5 to 2. 5 cm (mean 1. 3 cm; median 1. 2 cm).

Materials and Methods (2) • Each of these lesions was composed of a solitary, discrete, circumscribed tumor composed microscopically of aborizing papillary fronds covered by a population of monomorphic cuboidal or columnar epithelial cells and surrounded by a fibrotic rim.

Materials and Methods (3) • 15 benign intraductal papillomas from the • • surgical pathology archives of BIDMC. Age from 33 to 61 years (mean=44 y; median=41 y) Size from 0. 3 to 1. 2 cm (mean 0. 7 cm; median 0. 6 cm).

Results (1) • Foci of DCIS were present in ducts adjacent to the IPC in 10 cases and 1 or more areas of unequivocal invasive carcinoma were present in 3 (ductal, NOS type in all 3 of these cases).

Results (2) • MEC were not identified within the papillary fronds of any of the 22 IPC using any of the 5 MEC markers studied. • In contrast, MEC were detected around normal ductal-lobular structures and around foci of conventional DCIS present adjacent to the nodules of IPC.

Results (3) • In some cases, a variable number of smooth • muscle myosin heavy chain (SMMHC) positive and/or calponin positive cells were noted at the outer edge of the lesion. However, their spindle cell morphology, contiguity with similar-appearing cells in the stroma further away from the nodule, and lack of staining for p 63 is consistent with these cells being stromal myofibroblasts in close proximity to the lesion rather than MEC.

Results (4) • All 15 benign IP showed a MEC layer around virtually the entire periphery of the lesion with all of the MEC markers

Result (5) • In 13 of the IPC a small proportion of neoplastic • epithelial cells present near the outer edge of the nodules (ie, closest to the fibrotic rim) showed immunoreactivity with antibodies to p 63 (n=7), CK 5/6 (n=2), or both (n=4). These cells were morphologically identical to the epithelial cells comprising the rest of the IPC and did not have the cytomorphology of MECs

Discussion (1) • It has long been recognized that some breast • carcinomas are composed of well-circumscribed nodules of papillary carcinoma that appear to be confined to a cystically dilated duct. These lesions most commonly occur in older women, have been termed intracystic or ‘‘encysted’’ papillary carcinomas, and have been considered to represent a form of DCIS.

Discussion (2) • Many studies have clearly shown that the tumor cell nests of invasive breast cancers consistently lack a surrounding layer of MEC whereas the ductal-lobular spaces involved by carcinomas in situ, even when distorted, exhibit a delimiting layer of MEC at their periphery, although this layer may sometimes be (or appear to be) discontinuous.

In Fact…

Discussion (3) • It could be argued that the absence of a MEC layer around nests of tumor cell nests is the defining feature of invasive breast carcinomas.

Discussion (4) • In this study, we were unable to demonstrate MEC at the • periphery of the circumscribed tumor nodules in any of 22 lesions categorized as IPC by conventional histopathologic features using any of 5 sensitive MEC markers that recognize different components of MECs. Consistent with our observations, Hill and Yeh recently reported that 5 of 9 lesions classified as ‘‘intracystic/intraductal papillary carcinoma’’ showed complete absence of MEC at the periphery of the neoplastic cell nests using antibodies to calponin, SMMHC, and p 63.

WHY?

Discussion (5) • One possible explanation is that these are in situ lesions in which the delimiting MEC layer has become markedly attenuated or altered with regard to expression of these proteins, perhaps due to their compression by the expansile growth of these lesions within a cystically dilated duct.

But…

Discussion (6) • This seems unlikely as all intraductal papillomas in our study, even those of sizes comparable to those of some of the IPC, exhibited a peripheral layer of MEC with all of these antibodies.

It is also possible that…

Discussion (7) • MEC are in fact present but that the particular proteins we evaluated by immunohistochemistry are not expressed.

However…

Discussion (8) • This too seems unlikely as there was no histologic evidence of MECs at the periphery of the tumor nodules in any of these IPC cases and given the broad range of categories included among the proteins we evaluated.

Alternatively…

Discussion (9) • It may be that at least some lesions categorized as IPC • using conventional histologic criteria actually represent circumscribed nodules of invasive papillary carcinoma rather than in situ lesions. Hill and Yeh: There may be a spectrum of progression from in situ to invasive papillary carcinomas and further suggested that large, expansile, well- circumscribed papillary carcinomas with a prominent fibrotic rim may represent an encapsulated variant of invasive papillary carcinoma rather than a variant of DCIS.

And…

Discussion (10) • A review of the clinical outcome of patients given • the diagnosis of IPC sheds little additional light on the nature of these lesions. At least some patients reported in the older literature as having IPC had very large tumors, some sufficiently large to have required mastectomy. In addition, some of these patients had axillary lymph node metastases and/or developed distant metastases and died of their disease.

Thus…

Discussion (11) • Some lesions previously reported as IPC clearly • represented invasive breast cancers with secondary cystic degeneration rather than variants of DCIS. More contemporary studies have indicated that patients with IPC have an excellent outcome with adequate local therapy alone, a course more consistent with that of DCIS.

Discussion (12) • The favorable clinical outcome reported in recent series of IPC does not provide direct evidence that these are in situ rather than invasive lesions as some low grade invasive carcinomas (particularly, small tubular carcinomas, invasive cribriform carcinomas, and adenoid cystic carcinomas) are associated with an equally favorable outcome.

In our view…

Discussion (13) • The critical issue in defining the nature of these lesions is whether one considers the absence of a delimiting MEC layer sufficient evidence to conclude that these lesions are invasive carcinomas rather than variants of DCIS.

It is of interest that…

Discussion (14) • A small proportion of epithelial cells at the periphery of the tumor nodules in some of our cases of IPC exhibited expression of p 63 and/or CK 5/6.

Discussion (15) • Stefanou similarly noted that 10% to 15% of epithelial • • cells in some ‘‘in situ papillary-type ductal carcinomas’’ and invasive papillary carcinomas showed staining for p 63 and that the p 63 -positive epithelial cells in these cases were located at the epithelial-stromal interface. The explanation for this observation is at this time a matter of speculation. However, Stefanou et al raised the interesting possibility that these p 63 -positive epithelial cells have acquired a ‘‘myoepithelial differentiation program. ’’ Given that MEC possess important tumor suppressor properties, it is possible that the epithelial cells with a partial myoepithelial phenotype at the periphery of the nodules may have a role in modulating the invasiveness of lesions classified as IPCs.

In conclusion…

Discussion (16) • at least some (and perhaps most) lesions classified as IPC on histologic grounds alone represent encapsulated nodules of invasive papillary carcinoma.

Discussion (17) • Based upon these observations, we concur with the view of Hill and Yeh that the term ‘‘encapsulated’’ papillary carcinoma is preferable to intracystic papillary carcinoma for circumscribed nodules of papillary carcinoma surrounded by a fibrous capsule and in which a peripheral delimiting layer of MEC is not demonstrable.

Discussion (18) • Regardless of whether these lesions are truly in • situ or invasive carcinomas, available outcome data indicate that they have an excellent prognosis with adequate local therapy alone. Therefore, we believe it is most prudent to continue to manage patients with these lesions as they are currently managed (ie, similar to patients with DCIS) and to avoid categorization of such lesions as frankly invasive papillary carcinomas.

Thanks for your attention!