3852

KRT5

keratin 5, type II;KRT5;keratin 5, type II

58 kda cytokeratin|CK-5|cytokeratin-5|epidermolysis bullosa simplex 2 Dowling-Meara/Kobner/Weber-Cockayne types|keratin 5 (epidermolysis bullosa simplex, Dowling-Meara/Kobner/Weber-Cockayne types)|keratin, type II cytoskeletal 5|type-II keratin Kb5

12q13.13

protein-coding

OBJECTIVE: To investigate the correlation of CK5/6 and CK17 expression with clinical outcome in patients with triple-negative [ER(-), PR(-), Her-2(-)] breast cancer. METHODS: 112 patients with breast cancer treated by surgery between 2000 and 2002 were included in this study. All cases were immunohistochemically proven to be triple-negative. Samples of formalin-fixed and paraffin-embedded surgical specimens were obtained for immunohistological examination for CK5/6 and CK17 expression. The correlation of the gene expression with clinicopathological features and outcome of the patients was analyzed. RESULTS: Of the 112 triple-negative patients, five-year disease-free survival rate was 73.2% (82/112). The positive rate of both CK5/6 and CK17 was 21.4% (24/112), either CK5/6 or CK17 positive was 46.4% (52/112). It was shown by Kaplan-Meier curve that positive CK5/6, CK17 or CKs (CK5/6 or CK17 positive) was correlated with poor five-year disease-free survival (P = 0.020, P = 0.032, P = 0.003); and positive staining of CK5/6 or CKs was correlated with poor five-year overall survival (P = 0.027, P = 0.015). Of the 91 patients with invasive ductal carcinoma, a correlation of CK5/6 or CK17 positive staining with high grade differentiation was observed (P = 0.030), and with axillary lymph node metastasis was also noticed (P = 0.044). Multivariate analysis by Cox regression showed that differentiation grade, pathological stage and expression of CK5/6 were factors affecting both the disease-free-survival and overall-survival, while menopausal status was an independent factor affecting the disease-free-survival. CONCLUSION: Positive expression of CK5/6 or CK17 in patients with triple-negative breast cancer is correlated with poor prognosis, high grade differentiation and axillary lymph node metastasis.

BACKGROUND: Development of brain metastasis in patients with breast carcinoma correlates with poor outcome. Identification of tumor characteristics associated with breast cancer brain metastases (BCBM) could help identify patients at risk. PATIENTS AND METHODS: We studied 209 patients with BCBM. We evaluated a panel of proteins relevant to the biology of breast carcinoma on tissue microarrays of 133 primary tumors and 56 BCBM, including paired samples from 43 patients, and correlated the findings with the clinical outcome. RESULTS: The median survival after BCBM diagnosis was 19 months (95% confidence interval, 13-23 months). Patients presenting with solitary metastasis had a significantly longer median survival than those with multiple lesions (25 versus 11 months, P expression of tested markers, but identified a possible association between the expression of basal cytokeratin CK5/6 in the primary carcinoma and the development of multiple rather than solitary brain metastases. CONCLUSIONS: expression of antigens commonly associated with breast carcinoma does not differ significantly between the primary tumor and the corresponding brain metastases. Although no specific immunoprofile identifies breast carcinomas that develop brain metastases, we observed a possible association between CK5/6 expression in the primary tumor and multiple versus solitary BCBM.

In subgroups of breast cancer, the shortest disease-free and overall survival was observed in basaloid and human epidermal growth factor receptor-2 groups. CK5/6 expression is a marker used in diagnosing breast cancers in basaloid group and is associated with a poor prognosis. Similarly, loss of tumor suppressor gene PTEN and a high expression of c-Met has been associated with poor prognosis in breast cancer and many other cancers. In this study, we aimed to determine the effect of CK5/6 and c-Met expressions, and PTEN loss on the disease prognosis in triple-negative breast cancer patients. Ninety-seven patients pathologically diagnosed with triple-negative breast cancer were enrolled. The clinical and pathological characteristics of the patients were recorded. c-Met, PTEN, and CK5/6 expressions were evaluated with immunohistochemical methods from paraffin blocks. The median age of patients was 47 years. CK5/6 positivity was 50.5 %, PTEN loss was 44.3 %, and high c-Met expression was detected in 53.6 %. In multivariate analysis, predictors of the recurrence were loss of PTEN (HR = 2.99; P = 0.004), high c-Met expression (HR = 2.05; P = 0.06), CK5/6 expression (HR = 2.99; P = 0.02), increase in the number of metastatic lymph nodes (HR = 1.11; P = 0.001), and an increase in tumor size (HR = 1.226; P = 0.01). Also, PTEN loss (HR = 2.43; P = 0.05), CK5/6 expression (HR = 3.74; P = 0.01), and N2-3 tumors compared to negatives (HR = 3.63; P = 0.01) were associated with death. PTEN loss correlated with those of lymphovascular invasion. There was a correlation between CK5/6 expression and the number of metastatic lymph nodes. Also, a correlation was found among cancers with highly expressed levels of c-Met, T1-2 tumors, and high-grade tumors. The classical markers, lymph node involvement and tumor size, were found to be of prognostic value; however, high c-Met and CK5/6 expressions, and PTEN loss were found to increase risk of recurrence and death in patients with triple-negative breast cancer.

The differentiation of primary cutaneous adnexal neoplasms (CANs) from dermal-based metastatic lesions can be difficult. Cytokeratin 5/6 (CK 5/6) has a relatively limited expression profile, being relatively specific for mesothelium and other "pavement" type epithelium such as squamous epithelium. To date, the degree and distribution of CK 5/6 expression in cutaneous neoplasms has not been extensively studied. We speculate that since most of CANs arise from similar epithelium, they should express CK 5/6 and, therefore, CK 5/6 could potentially be helpful in distinguishing these lesions from most of metastatic neoplasms, which usually do not express this marker. Formalin-fixed, paraffin-embedded tissue sections from 228 previously classified CANs and 27 metastatic adenocarcinomas (17 breast, 4 colon, 2 prostate, 2 ovary, 1 lung, and 1 esophagus) were immunostained with anti-CK 5/6. Anti-CK 5/6 labeled 2 of 2 proliferating trichilemmal tumors, 6 of 6 poromas, 4 of 5 hydrocystomas, 10 of 10 cylindromas, 10 of 10 eccrine acrospiromas, 8 of 10 pilomatricoma, 10 of 10 nevus sebaceus, 9 of 9 desmoplastic trichoepitheliomas, 7 of 7 nevus sebaceus with basal cell carcinomas, 10 of 10 pilar cysts, 14 of 14 trichilemmomas, 10 of 10 syringomas, 6 of 7 chondroid syringomas, 10 of 10 hidradenoma papilliferum, 9 of 9 sebaceus adenomas, 3 of 3 microcystic adnexal carcinomas, 10 of 10 eccrine spiradenomas, 4 of 4 syringocystadenoma papilliferum, 3 of 5 ocular sebaceous carcinomas, 28 of 28 basal cell carcinomas, 16 of 16 trichoepitheliomas, and 33 of 33 trichoepitheliomas with basal cell features. By contrast, 9 of 27 metastatic adenocarcinomas stained positively, although only two of these stained strongly. Cumulatively, CK 5/6 was expressed by most (97%) of CANs, while only 33% of metastatic adenocarcinomas showed positive expression. The sensitivity of this marker in the malignant lesions (other than basal cell carcinoma) is 78%, while the specificity is 67%. If all lesions are considered, the sensitivity increases to 97%. Therefore, CK 5/6 may prove to be a useful adjunct marker in distinguishing CANs from metastatic lesions.

Adenosquamous carcinoma, a rare tumor of the skin with few reported cases in the literature, has not been studied with p63, cytokeratin 5/6 and cytokeratin 7. We stained nine such tumors with these markers. Histologically, the tumors showed superficial, atypical islands of keratinocytes in close association with islands displaying glandular differentiation. Clinically, lesions favored the head and trunk, and a subset of cases showed aggressive behavior. All tumors marked with p63 and cytokeratin 5/6, substantiating that diffuse positivity with these stains is supportive of a primary cutaneous origin. Six tumors stained focally in luminal areas with cytokeratin 7. Recognition of adenosquamous carcinoma is important for appropriate therapy, and stains for p63 and cytokeratin 5/6 may be helpful in ruling out metastatic adenocarcinoma.