Immunostaining Patterns of Posttransplant Liver Biopsies Using 2 Anti-C4d Antibodies.

Histopathologic diagnosis of antibody-mediated rejection in posttransplant liver biopsies is challenging. The recently proposed diagnostic criteria by the Banff Working Group on Liver Allograft Pathology require positive C4d immunohistochemical staining to establish the diagnosis. However, the reported C4d staining patterns vary widely in different studies. One potential explanation may be due to different antibody preparations used by different investigators. In this study, posttransplant liver biopsies from 69 patients histopathologically diagnosed with acute cellular rejection, chronic rejection, or recurrent hepatitis C were immunohistochemically stained using 2 polyclonal anti-C4d antibodies. On the basis of the distribution of C4d immunoreactivity, 5 different staining patterns were observed: portal vein and capillary, hepatic artery, portal stroma, central vein, and sinusoids. The frequency, extent, and intensity of positive C4d staining with the 2 antibody preparations differed significantly for portal veins/capillaries and central veins, but not for hepatic arteries and portal stroma. Positive sinusoidal staining was seen in only 1 case. There were no significant differences in the frequency, extent, and intensity of positive C4d staining among the acute cellular rejection, chronic rejection, and recurrent hepatitis C groups with the 2 anti-C4d antibodies. These data show that different anti-C4d antibodies can show different staining patterns, which may lead to different interpretation. Caution is thus needed when selecting C4d antibodies for clinical use to aid in the diagnosis of antibody-mediated rejection.

Diagnostic value of maspin in distinguishing adenocarcinoma from benign biliary epithelium on endoscopic bile duct biopsy.

Histopathologic distinction between benign and malignant epithelia on endoscopic bile duct biopsy can be extremely challenging due to small sample size, crush artifact, and a propensity for marked inflammatory and reactive changes after stent placement. Our previous studies have shown that the insulin-like growth factor II mRNA-binding protein 3, S100P, and the von Hippel-Lindau gene product (pVHL) can help the distinction. This study analyzed 134 endoscopic bile duct biopsy specimens (adenocarcinoma 45, atypical 31, and benign 58) by immunohistochemistry for the expression of maspin, a serine protease inhibitor. The results demonstrated that (1) maspin expression was more frequently detected in malignant than in benign biopsies; (2) malignant biopsies frequently showed diffuse, strong/intermediate, and combined nuclear/cytoplasmic staining patterns for maspin, which were much less commonly seen in benign biopsies; (3) the malignant staining patterns for maspin observed in atypical biopsies were consistent with follow-up data showing that 67% of these patients were subsequently diagnosed with adenocarcinoma; (4) a maspin+/S100P+/pVHL- staining profile was seen in 75% of malignant biopsies but in none of the benign cases. These observations demonstrate that maspin is a useful addition to the diagnostic immunohistochemical panel (S100P, pVHL, and insulin-like growth factor II mRNA-binding protein 3) to help distinguish malignant from benign epithelia on challenging bile duct biopsies.

Use of IMP3, S100P, and pVHL immunopanel to aid in the interpretation of bile duct biopsies with atypical histology or suspicious for malignancy.

Histologic evaluation of an endoscopic bile duct biopsy for malignancy is a known challenge. Our prior study has shown that the insulin-like growth factor-II mRNA binding protein-3 (IMP3), S100P, and the von Hippel-Lindau gene product (pVHL) are a useful immunopanel for the distinction between adenocarcinoma and benign biliary epithelium. To further evaluate the usefulness of the IMP3, S100P, and pVHL immunopanel to aid in the interpretation of bile duct biopsies, 16 histologically challenging bile duct biopsies that exhibited atypical histology or features suspicious for malignancy were immunohistochemically stained for IMP3, S100P, and pVHL. Clinical follow-up data were obtained for each case. The results showed that in the 11 cases that showed adenocarcinoma during follow-up, the following staining patterns in atypical/suspicious cells in the initial biopsies were observed: IMP3-positive/S100P-positive/pVHL-negative or reduced (n=6), IMP3-negative/S100P-positive/pVHL-negative or reduced (n=4), and IMP3-positive/S100P-negative/pVHL-negative (n=1). In the 5 follow-up-proven benign cases, 2 biopsies showed an IMP3-negative/S100P-positive/pVHL-positive pattern in atypical cells and 1 was negative for all 3 proteins. The remaining 2 biopsies exhibited an IMP3-positive/S100P-positive/pVHL-negative or reduced pattern in atypical cells that were histologically considered dysplastic on retrospective review. These observations reaffirm that bile duct adenocarcinoma frequently shows positive IMP3 and/or S100P staining with reciprocal negative or reduced pVHL expression. This staining pattern can also be seen in dysplastic epithelium in the absence of invasive carcinoma. On the contrary, benign biliary epithelium typically lacks IMP3 immunoreactivity and may retain normal pVHL expression. However, caution should be exercised when using this immunopanel in the interpretation of challenging bile duct biopsies, because S100P and pVHL stains may give rise to variable patterns that can be difficult to interpret.

Diagnostic utility of CD10 in benign and malignant extrahepatic bile duct lesions.

CD10, a cell surface enzyme with neutral metalloendopeptidase activity, is a marker for intestinal epithelial brush border. It is also present in normal bile ducts and gallbladder epithelia but is absent in cholangiocarcinomas. However, the expression profile of CD10 in benign and malignant extrahepatic biliary lesions has not been studied. In this study, 69 biopsies, 9 resections, and 9 cell blocks prepared from fine-needle aspirations of the extrahepatic bile ducts from 86 patients were studied immunohistochemically for CD10 expression. The majority of cases contained normal biliary epithelium (NL, n=64), along with foci of benign or malignant lesions in various combinations. Benign lesions included reactive atypia (n=35), low-grade dysplasia of unknown significance (n=21), and bile duct adenoma (BDA, n=1). Malignant lesions included high-grade dysplasia (HGD, n=45) and invasive adenocarcinoma (IC, n=30). As expected, the NL showed strong continuous staining at the apical surface in all cases. Benign lesions were also CD10 positive in all but 3 cases; however, the staining pattern was discontinuous, with positive cells varying from 20% to 80%. None of the malignant lesions showed CD10 immunoreactivity, except for 2 HGD cases and 1 IC case, which exhibited focal staining. The Pearson χ2 and Fisher exact tests showed significant statistical difference in CD10 expression among the study groups (P<0.001). Our findings suggest that absence of CD10 expression in strips of atypical biliary epithelial cells may be a phenotype associated with malignant transformation and may serve as a useful marker to aid in the evaluation of bile duct biopsies, in which distinction between benign and malignant lesions on biopsies or cytology specimens can be extremely challenging because of limited sampling, crush artifact, and frequent inflammatory/reactive changes.