Search

Claudins, members of a large family of adherent junction proteins, regulate the integrity and function of tight junctions and influence tumorigenesis. Recent studies have shown that altered levels of the different claudins may be related to invasion and progression of carcinoma cells in several primary neoplasms. However, there is no reported study documenting the pattern of claudin expression in prostatic adenocarcinomas (PACs). Formalin-fixed paraffin-embedded sections from 141 PACs were immunostained by manual and automated methods on the Xmatrx (BioGenex, San Ramon, CA) using antihuman claudin-1, -3, -4, -5, and -7 antibodies (Zymed, San Francisco, CA). Membranous immunoreactivity for each protein was semiquantitatively scored in both the tumor and adjacent benign epithelium in each case. Results were correlated with clinicopathologic variables. Variable membranous positivity was noted in the adjacent benign glands for all 5 proteins in all cases. PACs showed variable membranous positivity ranging from decreased, similar to, and increased in relation to the adjacent benign epithelium for all claudins. Decreased expression of claudin-1 correlated with high tumor grade (P = .001) and biochemical disease recurrence (P = .01), whereas decreased claudin-7 correlated with high tumor grade (P < .0001). In contrast, expression of claudin-3 correlated with advanced stage tumors (P = .03) and recurrence (P = .02), and expression of claudin-4 correlated with advanced stage (P = .02). On multivariate analysis, advanced stage (P = .026) and decreased claudin-1 protein expression (P = .005) independently predicted disease recurrence. Immunohistochemical expression and prognostic significance of claudins are variable in PACs, with decreased expression of claudin-1 emerging as an independent prognostic variable warranting further study.

The tumor suppressor gene Smad4 (DPC4) has been localized to chromosome 18q21.1 and is a member of the Smad family that mediates the transforming growth factor beta signaling pathway suppressing epithelial cell growth. However, variable expression of this protein has been reported, with a loss in some cancers and increased expression in others. Given both the variability and lack of consensus reported regarding Smad4 expression in prostate cancer, we assessed Smad4 immunoreactivity in prostatic adenocarcinomas (PACs). Formalin-fixed, paraffin-embedded tissue sections from 133 PACs were immunostained by a manual method using indirect biotin streptavidin horseradish peroxidase and diaminobenzidine detection using a monoclonal mouse antihuman Smad4 antibody (sc-7966; Santa Cruz Biotechnology Inc, Santa Cruz, Calif). Nuclear immunoreactivity and cytoplasmic immunoreactivity were each semiquantitatively scored based on intensity and percentage of positive cells. Deoxyribonucelic acid ploidy was determined on Feulgen-stained tissue sections by static image analysis. Results were correlated with morphological and prognostic variables. Variable nuclear and cytoplasmic Smad4 positivity was noted in the adjacent benign glands in all cases. Of 133 PACs, 64 (48%) featured increased nuclear and 68 (51%) featured increased cytoplasmic protein expression. Nuclear Smad4 overexpression correlated with tumor grade (P = .02), stage (P = .04), and DNA ploidy (P = .04). Cytoplasmic overexpression correlated with tumor grade (P = .04) and DNA ploidy (P = .04) while showing a trend for correlation with tumor stage (P = .08). Neither nuclear nor cytoplasmic Smad4 overexpression correlated with postsurgical biochemical disease recurrence. Smad4 protein expression persists in PACs compared with benign glands, with both nuclear and cytoplasmic overexpression correlating with prognostic variables indicative of aggressive tumor behavior. Given the significant reported variability of Smad4 in several different cancers, further studies in prostate and other tumors are warranted to elucidate its role in tumorigenesis.

HER-2/neu expression has been established as a prognostic factor in breast and other cancers. In prostate cancer (PC), a similar predictive role has been hindered by variable immunohistochemical (IHC) results. The authors studied DNA amplification of the HER-2/neu gene on 4-microm sections obtained from 62 formalin-fixed, paraffin-embedded PCs by fluorescence in situ hybridization (FISH). The results were compared with HER-2/neu protein expression as determined by IHC and correlated by logistic regression analysis with Gleason tumor grade, DNA ploidy, serum prostate specific antigen (PSA), and pathological stage. The HER-2/neu gene was localized using the Oncor (Gaithersburg, MD) digoxigenin-labeled unique sequence probe. Amplified PCs had at least 20 malignant cells, with 5 or more copies of the sequence. Amplification of HER-2/neu correlated with Gleason score (P = .0001). The mean Gleason score of unamplified tumors was 5.7 and that of amplified tumors was 7.5. Nondiploid tumors had a significantly greater rate of HER-2/neu amplification compared with diploid tumors (P = .0003). Of the 62 cases evaluated by IHC and FISH, 18 cases (29%) were overexpressed by IHC, and 27 cases (44%) were amplified by FISH. A trend for similar HER-2/neu status in each PC by the two methods did not reach statistical significance (P = .23). HER-2/neu amplification by FISH was associated with advanced pathological stage; however, this relationship reached only near-statistical significance (P = .06). There was no correlation of HER-2/neu amplification by FISH with patient age or preoperative serum PSA levels. The authors conclude that HER-2/neu gene amplification status can be determined by FISH on archival prostate cancer specimens, significantly correlates with high tumor grade and nondiploid DNA content, and is more frequently encountered in tumors with advanced pathological stage. Also, FISH is more sensitive than IHC for detection of abnormalities in the HER-2/neu gene, and further studies should be undertaken to determine whether a FISH-based HER-2/neu detection method may prove of importance in the prediction of prognosis and planning of therapy in prostate cancer patients.

E-cadherin (E-CD), a cell adhesion molecule that plays a diverse role in cell-cell and cell-matrix interaction, has recently been associated with tumor invasion and metastasis. We evaluated the E-CD expression in 55 cases of urinary bladder papillary transitional cell carcinoma using a double-linked immunoalkaline phosphatase procedure on formalin-fixed, paraffin-embedded specimens quantified as percentage positive staining area with the Roche RPW image analyzer (Roche Image Analysis Systems, Elon College, NC). The 10 grade I tumors stained intensely for E-CD (66%) and showed enhanced staining compared with normal transitional epithelium. The 20 grade II lesions stained moderately at 45%, and the 25 grade III lesions showed weak expression of E-CD with a 33% positive stain. This loss of expression for high-grade versus low-grade tumors was statistically significant (P < .02). There was a similar decrease in E-CD expression for high-stage versus low-stage specimens with 30 stage 0 and stage A cases having an average of 51% positive stain and 25 stage B, C, and D cases having an average stain of 33%. This difference was also statistically significant (P < .001). We conclude that loss of expression of E-CD in high-grade, high-stage urinary bladder transitional cell carcinoma is associated with significant bladder wall invasion, indicates potential for metastasis, and may be of clinical value in the assessment of prognosis and planning of therapy for patients with bladder tumors.

To determine whether p53 immunoreactivity correlates with the Gleason tumor grade in primary adenocarcinoma of the prostate we analyzed 107 consecutive surgical specimens (78 radical prostatectomies and 29 transurethral resections). A hematoxylin-eosin-stained slide from a representative block of each tumor was examined, and primary and secondary Gleason scores were assigned in each case. Additional paraffin sections from the same block were stained immunohistochemically for p53 expression using the monoclonal antibody clone DO-1, a mouse IgG2a directed against a denaturation-resistant epitope of p53. Four of 54 (7.4%) low-grade tumors (combined Gleason score of 6 and below) and 11 of 53 (20.8%) high-grade tumors (combined Gleason score of 7 and above) revealed strong nuclear positivity for p53. When evaluated using only the primary Gleason score, none of 23 (0%) Gleason grade 2 tumors and 15 of 84 (17.9%) Gleason grade 3 or higher tumors were positive. These data demonstrate a positive association between p53 immunoreactivity and higher Gleason grade tumors (P = .04 for the combined score, P = .02 for primary score only). In addition, we noted occasional p53-positive nuclei in basal cells of benign glandular acini in regions flanking tumor. Focally positive nuclear staining also was demonstrated in basal cells from nine of 25 prostate glands exhibiting benign prostatic hyperplasia with no tumor. These results suggest that p53 overexpression might be associated with the known proliferative capacity of basal cells in benign hyperplastic prostate glands, and that mutations of p53 might play a role in the pathogenesis of a subset of high-grade prostate adenocarcinomas.