New trends of immunohistochemistry for making differential diagnosis of breast lesions

Immunohistochemistry is widely used for pathological diagnosis of breast lesions. Other than hormone receptors and HER2/neu analysis for primary breast carcinomas, several markers may be useful for differential diagnoses, although in limited situations. To decide the malignant potential of intraductal proliferative lesions, analysis for the staining pattern of cytokeratins may be a good reference. Most ductal carcinoma in situ cases are diffusely positive for luminal cell markers (CK8, CK18, CK19), but negative for basal cell markers (CK5/6 and CK14). However, usual ductal hyperplasia may show the mosaic staining patterns for any of these markers, which may indicate a heterogeneous cell population in benign lesions. Myoepithelial markers (α-SMA, myosin, calponin, p63, CD10) are almost consistently positive for benign papillomas but they do not completely distinguish intraductal papillary carcinomas. Preservation of myoepithelial layer is the diagnostic key when looking at benign sclerosing lesions, including carcinoma with pseudoinvasive structures. E-cadherin is mostly positive for ductal carcinomas but negative for lobular carcinomas. Some of the lobular carcinomas are positive for 34βE12, but they are consistently negative for CK5/6. Comparison with histopathological findings of hematoxylin and eosin is essential to make proper diagnosis in the individual case.     

Value of p63 and cytokeratin 5/6 as immunohistochemical markers for the differential diagnosis of poorly differentiated and undifferentiated carcinomas.

To facilitate the differential diagnosis of poorly differentiated metastatic carcinomas of unknown primary site, we evaluated p63 and cytokeratin (CK) 5/6 as immunohistochemical markers for squamous cell carcinomas. The study cases were as follows: squamous cell carcinoma of the lungs, head/neck, esophagus, cervix uteri, or anal canal, 73; non-squamous cell carcinomas of various primary sites, 141; and urothelial carcinoma, 20. We also tested 14 malignant mesotheliomas. Immunoreactivity for p63 was as follows: squamous cell carcinomas, 59 (81%); urothelial carcinoma, 14 (70%), most often with diffuse staining patterns; non-squamous cell carcinomas, 20 (14.2%), resulting in a specificity of 0.86 of p63 for squamous cell carcinomas. Coexpression of p63 and CK5/6 had a sensitivity of 0.77 and a specificity of 0.96 for squamous cell carcinomas. Increasing the minimal criterion of positive immunostaining for both markers to more than 50% of immunoreactive tumor cells resulted in a specificity of 0.99, although the sensitivity diminished to 0.66. All malignant mesotheliomas were negative for p63. Our data suggest that positive immunostaining for both p63 and CK5/6 in poorly differentiated metastatic carcinomas is highly predictive of a primary tumor of squamous epithelial origin.     

Metaplastic breast carcinomas are basal-like tumours

AIMS: Recently, an immunohistochemical panel comprising antibodies against HER2, oestrogen receptor (ER), epidermal growth factor receptor (EGFR) and cytokeratin (CK) 5/6 was reported to identify basal-like breast carcinomas, as defined by cDNA microarrays. Our aim was to analyse a series of metaplastic breast carcinomas (MBCs) using this panel plus two other basal markers (CK14 and p63) and progesterone receptor (PR), to define how frequently MBCs show a basal-like immunophenotype. METHODS AND RESULTS: Sixty-five cases were retrieved from the pathology archives of the authors' institutions and reviewed by three of the authors. Immunohistochemistry with antibodies for HER2, ER, EGFR, CK5/6, CK14 and p63 was performed according to standard methods. All but six cases (91%) showed the typical immunoprofile of basal-like tumours (ER- and HER2-, EGFR+ and/or CK5/6+). When CK14 and p63 were added to the panel, two additional cases could be classified as basal-like. The majority of MBCs lacked PR, except 4/19 (21%) carcinomas with squamous metaplasia. CONCLUSIONS: Our results demonstrate that MBCs show a basal-like phenotype, regardless of the type of metaplastic elements. Moreover, as these neoplasms frequently overexpress EGFR (57%), patients with MBC may benefit from treatment with anti-EGFR drugs.     

乳腺导管内乳头状肿瘤187例临床病理及免疫组织化学特征

目的 探讨乳腺导管内乳头状肿瘤(IDPN)的诊断方法和标准.方法 收集187例IDPN患者的临床和病理资料,结合目前认可的2003年WHO乳腺和女性生殖系统肿瘤病理学和遗传学分类标准、Page等和Tavassoli的诊断标准,对其形态学特点进行分析,并对其中53例行CD10、p63、CK14、CK5/6、CK7、乳珠蛋白-1(MGB1)及p53免疫组织化学EnVision法染色分析.结果 187例IDPN患者中导管内乳头状瘤(IDPMa) 128例,不典型导管内乳头状瘤(A-IDPMa) 16例,导管内乳头状癌(IDPCa) 43例.IDPN在形态学上表现为不同程度的上皮细胞和间质增生,以及继发病变等,这些使病灶呈现异常复杂的多样性.免疫组织化学肌上皮标记(CD10和p63)染色在IDPMa、A-IDPMa及IDPCa的表达依次减少,组间比较差异均有统计学意义(均P＜0.001).基底型角蛋白(CK5/6和CK14)染色显示良性病变的表达呈镶嵌状阳性表达,在A-IDPMa的不典型区和IDPCa中表达明显减少或缺如,两者相比差异有统计学意义(P＜0.001).腺腔上皮标志物CK7染色各组间比较差异无统计学意义(P=0.06).MGB1在IDPCa组染色明显减少(P值分别为0.002和0.007),p53染色各组均呈阴性.结论 IDPN是一组组织学改变复杂的疾病,应注意其诊断标准的掌握.肌上皮、基底型角蛋白和腺腔上皮标志物联合应用在该组复杂病变中有很好的诊断和鉴别诊断价值.     

Morphological and immunophenotypic analysis of breast carcinomas with basal and myoepithelial differentiation

The aim of this study was to assess the morphological characteristics and immunohistochemical profile of breast carcinomas with basal and myoepithelial phenotypes to obtain a better understanding of their biological behaviour and nature. One thousand nine hundred and forty-four invasive breast carcinomas were examined, using tissue microarray (TMA) technology and immunohistochemistry, to identify those tumours that showed basal and myoepithelial phenotypes, and their immunophenotype profile was characterized using a variety of markers. In addition, haematoxylin and eosin-stained sections of these tumours were studied for several morphological parameters. The findings were correlated with patient and tumour characteristics and outcome data. Tumours were classified into two groups: (1) tumours with basal phenotype [expressing one or both basal markers (CK5/6 and/or CK14)] and (2) tumours with myoepithelial phenotype (expressing SMA and/or p63). Group 1 was further subdivided into two subgroups: (A) dominant basal pattern (more than 50% of cells positive) and (B) basal characteristics (10-50% of cells positive). Group 1 tumours constituted 18.6% (8.6% and 10% for groups 1A and 1B, respectively) and group 2 constituted 13.7% of the cases. In both groups, the most common histological types were ductal/no specific type, tubular mixed and medullary-like carcinomas; the majority of these tumours were grade 3. There were positive associations with adenoid cystic growth pattern, loss of tubule formation, marked cellular pleomorphism, poorer Nottingham prognostic index, and development of distant metastasis. In addition, associations were found with loss of expression of steroid hormone receptors and FHIT proteins and positive expression of p53 and EGFR. The most common characteristics in group 1 were larger size, high-grade comedo-type necrosis, development of tumour recurrence, and absence of lymph node disease. Group 2 tumours were more common in younger patients and were associated with central acellular zones, basaloid change, and positive E-cadherin protein expression. Group 1 characteristics were associated with both reduced overall survival (OS) [log rank (LR) = 22.5, p < 0.001] and reduced disease-free interval (DFI) (LR = 30.1, p < 0.001), while group 2 characteristics showed an association with OS (LR = 5, p = 0.02) but not with DFI. Multivariate analysis showed that basal, but not myoepithelial, phenotype has an independent value in predicting outcome. Breast cancers with basal and myoepithelial phenotypes are distinct groups of tumours that share some common morphological features and an association with poor prognosis. The basal rather than the myoepithelial phenotype has the strongest relationship with patient outcome.     

Breast carcinoma with basal differentiation: a proposal for pathology definition based on basal cytokeratin expression

AIM: To assess the expression and coexpression of a range of different biomarkers that have been used to define breast carcinomas with a basal phenotype (BP) and their relationship with prognosis in an attempt to refine the definition of BP and to evaluate the reliability of using a single biomarker to identify these tumours. METHODS AND RESULTS: The expression pattern of basal cytokeratins (CK5/6 and CK14), oestrogen, progesterone and androgen receptors, epidermal growth factor receptor, HER2, BRCA1, P-cadherin and myoepithelial markers (smooth muscle actin and p63) were studied in a well-characterized series of invasive breast carcinoma (1872 cases) with long-term follow-up using immunohistochemistry and tissue microarray. Although the additional markers were associated with basal CK expression, they did not serve to improve recognition of cases with differing outcome when compared with basal CKs alone and, if used to define cases, reduced considerably the proportion of cases allocated to this poor prognostic type of breast cancer. CONCLUSION: BP can be defined based on the expression of basal CKs regardless of the expression of other markers.     

The role of immunohistochemistry for smooth-muscle actin, p63, CD10 and cytokeratin 14 in the differential diagnosis of papillary lesions of the breast

BACKGROUND: Histological differentiation of mammary papillary lesions can be difficult. The evaluation of myoepithelial cells can be helpful, with benign papilloma showing a continuous myoepithelial cell layer, which becomes attenuated or absent in malignant papillary lesions. METHODS: A large series of 100 papillomas (28 papillomas with florid epithelial hyperplasia) and 68 papillary carcinomas (9 invasive, 44 in situ, and 15 ductal carcinomas in situ (DCIS) involving papillomas) of the breast were stained for myoepithelial cells by immunohistochemistry using antibodies to smooth-muscle actin (SMA), p63, CD10 and cytokeratin (CK) 14. RESULTS: In the papillomas, using these four antibodies, myoepithelial cells were positive in 88%, 99%, 91% and 95% of cases, respectively, with SMA showing marked stromal component cell staining and CD10 showing epithelial and stromal staining. CK14 also showed epithelial staining in 71% of papillomas and 96% of papillomas with florid epithelial hyperplasia. In the papillary carcinomas, 36 (53%) cases showed staining of myoepithelial cells that were scattered, discontinuous and diminished in number and the remaining 32 (47%) cases did not show myoepithelial cells. Invasive papillary carcinoma has the lowest proportion (33%) with myoepithelial cells, and DCIS involving papillomas had the highest proportion (87%). CONCLUSIONS: p63 had the highest sensitivity and did not cross-react with stromal cells and only rarely with epithelial cells. CK14 has the added ability to distinguish between florid epithelial hyperplasia involving papilloma and DCIS involving papillomas. CK14 and p63 may be used as an adjunct in assessing difficult papillary lesions of the breast.     

Anatomie der Brustdrüse

The human breast consists of lobes with a luminal glandular and a basal myoepithelial layer. Immunofluorescence studies have shown that the breast epithelium contains cytokeratin (CK)5/14-positive precursor cells which give rise to CK8/18-positive glandular or sm-actin-positive myoepithelial cells. Only some of the glandular cells contain estrogen receptors. The luminal epithelium of the lobules shows a much higher glandular differentiation than the ductal system. Diagnostically important cytokeratins of normal breast epithelium and its proliferative epithelial processes include luminal cytokeratins (CK7, CK8 and CK18) as markers of glandular differentiation and basal cytokeratins (CK5, CK14 and CK17) as markers of progenitor cells and early cells of the glandular and myoepithelial differentiation pathway. The most important myoepithelial markers are currently CD10, SMA, SMM-HC and Calponin.     

CD109 expression in basal-like breast carcinoma

Breast cancer can be classified into several subtypes based on gene expression profiling. Basal-like breast carcinoma (BLC) has a triple negative phenotype, that is, the subtype lacks the estrogen receptor (ER), progesterone receptor (PgR) and human epidermal growth factor receptor 2 (HER2). It has been recently reported that CD109, a glycosylphosphatidylinositol (GPI)-anchored cell surface protein, is a new breast myoepithelial marker. In the present study CD109 expression was investigated in invasive ductal carcinomas (IDC) of the breast on immunohistochemistry. Eighty-eight formalin-fixed, paraffin-embedded breast carcinoma sections were immunostained with anti-CD109, anti-cytokeratin 5/6 (CK5/6), anti-calponin, anti-vimentin and anti-p63 antibodies. CD109 expression was detected in 18 of 30 basal-like breast carcinomas (BLC) but not in other types of 53 IDC (non-BLC) that were positive for ER, PgR and/or HER2. The percentage of CD109-positive tissues (60%) in BLC was similar to that of CK5/6 (63%) and higher than that of other myoepithelial markers including p63 (23%), calponin (33%) and vimentin (33%). Statistical analysis indicated that the CD109-positive group in BLC, but not the CK5/6-positive group in BLC, was associated with reduced fat invasion ( P  < 0.05). These findings indicate that CD109 is a useful diagnostic marker for BLC and that CD109 expression may affect biological properties of cancer cells.     

p63 correlates with both BRCA1 and cytokeratin 5 in invasive breast carcinomas: further evidence for the pathogenesis of the basal phenotype of breast cancer

AIMS: To study the expression of p63, cytokeratin (CK) 5 and CK8/18 in invasive ductal carcinomas and their relationship with BRCA1 and other pathological and immunohistochemical features of clinical significance. METHODS AND RESULTS: Immunohistochemistry with the antibodies p63, CK5, CK8/18, BRCA1, oestrogen receptor, progesterone receptor, p53, c-erbB-2 and Ki67 was performed in 102 formalin-fixed paraffin-embedded samples of invasive ductal carcinomas. The CK5+ cases were submitted to a double-immunolabelling study with p63. There was a strong relationship between CK5 and p63 expression and both markers were associated with hormonal receptor-negative high-grade carcinomas with high proliferative rate. Furthermore, there was coexpression of CK5 and p63 in neoplastic cells, indicating that p63, like CK5, is a marker of the basal phenotype of breast cancer. There was a strong relationship between reduced expression of BRCA1 with both p63 and CK5 expression as well as an inverse correlation between p63 and CK8/18 expression, suggesting that loss of p63 expression is required for the transition between a basal to a luminal phenotype of breast carcinoma. CONCLUSIONS: Since p63 is thought to be a marker of stem cells and may act as an oncogene, our data support the idea that BRCA1 acts as stem cell regulator.     

Utility of high molecular weight cytokeratins, but not p63, in the differential diagnosis of neuroendocrine and basaloid carcinomas of the head and neck

High-grade neuroendocrine carcinomas of the head and neck overlap significantly in morphology with both basaloid squamous and solid-type adenoid cystic carcinomas. High-grade neuroendocrine carcinomas have sheets of small cells with scant cytoplasm, granular chromatin, and inconspicuous nucleoli. Basaloid squamous and adenoid cystic carcinomas are aggressive variants of their respective tumor types which both have nests of basaloid tumor cells with round nuclei, little cytoplasm, and inconspicuous nucleoli. As the management and prognosis of these tumors are very different, it is important to differentiate them. We performed high molecular weight cytokeratin (CK) and p63 immunohistochemistry on 19 neuroendocrine carcinomas, 18 basaloid squamous carcinomas, and 11 solid-type adenoid cystic carcinomas. All tumors were immunostained for p63, CK 34betaE12, CK 5/6, synaptophysin, chromogranin-A, S-100, and smooth muscle actin. All basaloid squamous and adenoid cystic carcinomas were positive for CK 5/6 and 34betaE12. Only 4 and 5 of the 19 neuroendocrine carcinomas, respectively, were positive for these markers. Staining was focal in the neuroendocrine cases when positive, whereas almost all basaloid squamous and adenoid cystic carcinomas showed strong staining. Almost all tumors of each type were positive for p63, including neuroendocrine carcinomas, but with different staining patterns. Basaloid squamous carcinomas were diffusely positive, neuroendocrine carcinomas were diffusely positive, but with weak staining, and adenoid cystic carcinomas showed a distinct pattern with staining at the periphery of the cell nests only. We conclude that high molecular weight cytokeratin immunostaining is helpful in distinguishing high-grade neuroendocrine carcinomas from similar tumor types.     

Expression of p63, keratin 5/6, keratin 7, and surfactant-A in non-small cell lung carcinomas

In this study, we sought to validate the importance of p63, CK5/CK6, CK7, and surfactant-A (SP-A) to classify 42 non-small cell lung cancers in autopsy and surgical resection specimens and to study the usefulness of these markers in distinguishing between squamous cell carcinomas and adenocarcinomas because of their different implications regarding treatment and prognosis. All adenocarcinoma cases were negative for p63; 9 (56.2%) of 16 were CK5/CK6 positive, 16 (94.1%) of 17 were CK7 positive, and 4 (26.6%) of 15 were SP-A positive. In squamous cell carcinoma, 1 case was CK7 and SP-A positive and 14 (77.8%) of 18 were p63 positive. The latter appears to be useful in differentiating squamous cell carcinoma from adenocarcinoma of the lung in small biopsies without keratinization or glandular differentiation; thus, for advanced-stage cases, where there is no possibility of surgical resection and the treatment of choice is radiotherapy plus chemotherapy, we would be able to differentiate between the two histological types, establishing then a different therapy.     

P-cadherin and cytokeratin 5: useful adjunct markers to distinguish basal-like ductal carcinomas in situ

Gene expression profiles of invasive breast carcinomas have identified a subgroup of tumours with worse prognosis, which have been called "basal-like". These are characterized by a specific pattern of expression, being estrogen receptor (ER) and HER2 negative, and frequently expressing at least one basal marker such as basal cytokeratins or epidermal growth factor receptor (EGFR). Previously, our group characterized basal-like tumours in a series of invasive breast carcinomas using P-cadherin (P-CD), p63 and cytokeratin 5 (CK5). Based on that study, we hypothesized that those high-grade basal-like invasive carcinomas might have a pre-invasive counterpart, which could be identified using the same approach. A series of 79 ductal carcinomas in situ (DCIS) were classified into distinct subgroups according to their ER, HER2 and basal markers expression. Luminal DCIS expressed ER and constituted 64.6% of the series; the HER2 overexpressing tumours did not express ER and represented 25.3% of the cases, whereas 10.1% lack the expression of ER and HER2 and expressed at least one basal marker (P-CD, CK5, CK14, p63, vimentin and/or EGFR). These basal-like DCIS were mostly high-grade, with comedo-type necrosis, and consistently showed expression of P-CD and CK5. In conclusion, DCIS with a basal-like phenotype represent a small percentage in our series, being P-CD and CK5, the most useful adjunct markers to distinguish this subset of carcinomas in situ of the breast.     

Mesothelial markers in high-grade breast carcinoma

Duhig E E, Kalpakos L, Yang I A & Clarke B E
(2011) Histopathology 59 , 957–964 Mesothelial markers in high‐grade breast carcinoma Aims: Advances in molecular profiling have subdivided breast carcinomas into distinct subtypes. Basal carcinomas are generally oestrogen receptor (ER)?progesterone receptor (PR)?/human epidermal growth factor receptor 2 (HER2)?, and cytokeratin (CK)5/6+. This profile overlaps with that of mesothelial cells. This study of high‐grade breast carcinomas was undertaken to determine the expression of mesothelial markers. Methods and results: Immunohistochemistry was performed on 23 basal‐like breast carcinomas and 30 high‐grade breast carcinomas with variable ER, PR and HER2 expression. The incidence of staining of CK5/6, CK14, calretinin, Wilms’ tumour 1 (WT1), thrombomodulin and epithelial membrane antigen was assessed statistically. CK14 staining was more specifically associated with triple‐negative tumours than CK5/6. Calretinin positivity was statistically associated with basal‐like carcinomas. WT1 and thrombomodulin expression was infrequent and limited to a small number of non‐basal carcinomas. Conclusions: There is an overlap between the immunophenotype of mesothelial cells and that of basal‐like carcinomas of breast. Positive calretinin and CK5/6 are not specific, and may be seen in both mesothelial cells and basal‐like breast carcinomas. Negative ER and PR of basal carcinomas may also bias the observer against a breast origin. However, other negative mesothelial markers, such as WT1 and thrombomodulin, may help point to the correct diagnosis.     

Immunohistochemical study of metaplastic carcinoma and central acellular carcinoma of the breast: central acellular carcinoma is related to metaplastic carcinoma

Metaplastic breast cancers (MBCs) [spindle cell carcinoma (SpCC), squamous cell carcinoma (SCC), and matrix-producing carcinoma (MPC)] and invasive carcinomas with central acellular zones (CACs) were analyzed with respect to biological potential by immunohistochemical analyses. Specimens from 40 patients [20 with MBCs (7 with SCC, 6 with SpCC, 5 with MPC, and 2 with mixed type)] and 20 with CACs were analyzed using antibodies to cytokeratin (CK) 8, 5/6, 14, AE1/AE3, 34αE12, involucrin, c-kit, vimentin (VIM), alpha-smooth muscle actin, p63, epidermal growth factor receptor, epithelial cell adhesion molecule, and estrogen receptor (ER)/progesterone receptor (PR)/HER2. Expression of CK5/6, 34βE12, VIM, nuclear p63, and cytoplasmic p63 was significantly higher with MBCs than CACs (38%/13%, 70%/43%, 85%/33%, 68%/40%, and 48%/18%, respectively). Other markers were expressed at various levels in these tumors, but the difference between them was not significant. Eighteen MBC and 8 CAC cases were triple (ER/PR/HER2) negative; 17 MBCs and 7 CACs were basal-like tumors. Several differences were seen in MBCs and CACs, but they were heterogeneous, differentiating multipotentially into mesenchymal, myoepithelial, basal-like phenotypes with "stem cell-like" features. Thus, CACs are related to MBCs by immunohistochemical analyses as well as according to morphological findings.     

Immunophenotypical profiles of salivary gland tumours: a new evidence for their histogenetic origin

The histogenetic origin of salivary gland tumours is not clear. In normal tissues smooth muscle actin (SMA) is expressed in myoepithelial cells, CK14 immunoreactivity is seen in myoepithelial and basal cells and CK10 in keratinized squamous epithelium. In this study, we examine the immunophenotypic properties of salivary gland tumours in order to obtain further insight into their histogenesis. 30 cases of salivary gland tumours (18 pleomorphic adenomas, 8 Warthin's tumours, 2 basal cell adenomas, 2 acinic cell carcinomas) were included in our study. Cytokeratin (CK) 10, CKI4, CKI7, CK18, CK 19, and smooth muscle actin (SMA) immunostains were applied to the sections. Immunoreactivities were detected and the statistical significance was evaluated by chi square test. SMA was not detected in Warthin's tumour (p < 0.0001). CK14 was found in all tumours except acinic cell carcinomas (p < 0.0001). CK10 immunoreactivity was observed in 5 Warthin's tumour. In conclusion, pleomorphic adenomas and basal cells adenomas originate from stem cells. Immunophenotypic profile of Warthin's tumour is suggestive of an embryological remnant origin.     

Immunohistochemical differentiation between type B3 thymomas and thymic squamous cell carcinomas

Type B3 thymomas and thymic squamous cell carcinomas have some overlapping histological features, so it is difficult to make the differential diagnosis between these two entities, especially when the biopsy specimen is small. Only a few markers such as CD5 and CD 117 were applied to the differential diagnosis, the purpose of this study is to identify other diagnostic markers to help making the differential diagnosis more accurate. GLUT-1, MUC-1, CD117, CD5, CEA, P63, CK19, CK5/6, CD1a and TdT were evaluated using 16 cases of type B3 thymoma and 20 cases of thymic squamous cell carcinoma. Staining scores were obtained by calculating the percentage of positive cells. The sensitivity of GLUT-1 or MUC-1 for thymic squamous cell carcinomas was highest (100%), followed by CK5/6 (95%), CD117 (90%), P63 (85%), CD5 (80%) and CEA (75%). The specificities of CD5, CD117 and CEA for thymic squamous cell carcinomas all were 100%, next was MUC-1 (56.3%), followed by GLUT-1 (50%), P63 (25%), CK5/6 (12.5%). The sensitivities of CK19, TdT, and CD1a for type B3 thymomas were 100%, 93.8% and 87.5%, respectively. The specificity of CD1a for type B3 thymomas was highest (100%), followed by TdT (95%), CK19 (10%). The reactivity of GLUT-1, MUC-1, CD117, CD5, CEA, CD1a and TdT in thymic squamous cell carcinomas and type B3 thymomas had significant difference. Usually a panel of markers is needed, if we combine GLUT-1 or MUC-1 which sensitivity for thymic squamous cell carcinomas is highest with CD5, CD117, CEA, CD1a or TdT which have high specificity, we can make the differential diagnosis effectively.     

The clinical use of a P63/cytokeratin7/18/cytokeratin5/14 antibody cocktail in diagnostic breast pathology

An antibody cocktail directed against p63, cytokeratin (CK)5/14, and CK7/18 is reported to be useful in distinguishing noninvasive from invasive breast lesions and for the characterization of intraductal epithelial proliferations. However, limited studies evaluate its use in clinical practice. A retrospective review of breast material at a university medical center identified cases that were immunostained with the above antibody cocktail. Additional p63 immunostaining alone was performed to further determine the utility of the antibody cocktail in the evaluation of invasion. Of 50 breast cases identified, the antibody cocktail was used to confirm or exclude invasion in 44 (88%). Twenty-two (50%) of these had easily identifiable p63/CK5/14-positive myoepithelial cells, whereas the remainder lacked such staining, confirming the diagnosis of invasive carcinoma. In 27 cases with available diagnostic material for additional p63 immunostaining, the cocktail better highlighted myoepithelial cells by staining nuclei and cytoplasm. Easier identification of invasion was also facilitated by CK7/18 expression in invasive foci, especially those composed of single cells. Ten cases were immunostained to help determine the nature of an intraductal proliferation. The cocktail demonstrated a mosaic staining pattern of both CK7/18- and CK5/14-positive epithelial cells in 3 (30%) cases consistent with usual hyperplasia; homogenous CK7/18 expression in the remaining cases supported the diagnosis of atypical ductal hyperplasia or carcinoma in situ. In summary, the p63/CK7/18/CK5/14 cocktail stain appears to be a useful tool in diagnostic breast pathology, in the evaluation of possible invasion, particularly in the setting of minute foci of invasion as well as in epithelial proliferations.