The role of the epithelial cell adhesion molecule EpCAM in cancer progression remains largely unclear. High expression of EpCAM in primary tumors is often associated with more aggressive phenotypes and EpCAM is the prime epithelial antigen in use to isolate circulating tumor cells (CTCs) and characterize disseminated tumor cells (DTCs). However, reduced expression of EpCAM was associated with epithelial-to-mesenchymal transition (EMT) and reports on a lack of EpCAM on CTCs emerged. These contradictory observations might reflect a context-dependent adaption of EpCAM expression during metastatic progression. To test this, EpCAM expression was monitored in esophageal cancer at different sites of early systemic disease. Although most of the primary esophageal tumors expressed high levels of EpCAM, the majority of DTCs in bone marrow lacked EpCAM. In vitro, downregulation of EpCAM expression at the plasma membrane was observed in migrating and invading cells, and was associated with a partial loss of the epithelial phenotype and with significantly decreased proliferation. Accordingly, induction of EMT through the action of TGFβ resulted in substantial loss of EpCAM cell surface expression on esophageal cancer cells. Knock-down or natural loss of EpCAM recapitulated these effects as it reduced proliferation while enhancing migration and invasion of cancer cells. Importantly, expression of EpCAM on DTCs was significantly associated with the occurrence of lymph node metastases and with significantly decreased overall survival of esophageal cancer patients. We validated this observation by showing that high expression of EpCAM promoted tumor outgrowth after xenotransplantation of esophageal carcinoma cells. The present data disclose a dynamic expression of EpCAM throughout tumor progression, where EpCAMhigh phenotypes correlate with proliferative stages, whereas EpCAMlow/negative phenotypes associated with migration, invasion and dissemination. Thus, differing expression levels of EpCAM must be taken into consideration for therapeutic approaches and during clinical retrieval of disseminated tumor cells.

Disseminierte Tumorzellen solider Tumoren in Knochenmark und Lymphknoten wie auch der Nachweis zirkulierender Tumorzellen im Blut der Patienten dienen zunehmend als Surrogatmarker für systemische Therapieansätze. Zur Detektion epithelialer Zellen in den genannten Gewebekompartimenten wird insbesondere das Zelloberflächenprotein EpCAM (CD326) verwendet (u.a. beim FDA-anerkannten CellSearch-System) und mit Edrecolomab und Catumaxomab stehen zudem zwei zugelassene therapeutische Antikörper zur Verfügung. Da bislang keine systematischen Daten zur Prävalenz von disseminierten EpCAM-positiven Tumorzellen vorliegen, haben wir EpCAM-positive Zellen aus dem Knochenmark von Patienten mit unterschiedlichen Karzinomen des Magen-Darm-Traktes isoliert.

Fragestellung: Weichteilsarkome (WTS) sind relativ seltene maligne Tumore, die etwa 1% aller Malignome ausmachen. Eines der häufigsten WTS sind die undifferenzierten pleomorphen Sarkome/NOS, die früher als maligne fibröse Histiozytome (MFH) klassifiziert wurden. Bei bis zu 23% der Patienten tritt ein Lokalrezidiv auf, das sich in 80% der Fälle bereits 2 Jahre nach initialer chirurgischer Therapie manifestiert. Bislang ist wenig über die genetische Progression dieser Tumorentität bekannt. Ziel unserer Studie war es daher, die zytogenetischen Veränderungen im Rahmen der Tumorprogression zu untersuchen.

Introduction: We previously identified the HER2 Oncogene as a potentially promising molecular target for adjuvant therapies in esophageal cancer. Here we describe the development and application of a quantitative genomic qPCR assay to determine HER2 gene copy numbers in single disseminated tumor cells (DTC). Material and Methods: The requirement for the application of our quantitative single cell qPCR assay is the whole genome amplification of single cells following the SCOMP-protocol. The HER2 assay consists of two primers for the HER2 gene and four primers for reference genes located in different genomic regions. To determine the HER2 copy numbers we first correlated the qPCR data of several breast cancer cell lines with results obtained by a standard fluorescence in situ hybridization (FISH) assay to determine HER2 copy numbers. We applied our assay to single cell amplification products from DTCs that were detected in a collective of 142 esophageal cancer patients with clinical follow-up data. Results: The qPCR data of single cell amplification products correlated significantly with HER2 FISH data in different breast cancer cell lines with different HER2 amplification levels (correlation coefficient of 0,96). When we applied our assay to single cell amplification products from DTCs that were detected in a collective of 142 esophageal cancer patients, we found that the calculated copy number of several DTCs exceeded that of 20 tested single cell amplification products from diploid mononuclear cells. Interestingly, HER2 copy numbers of some esophageal DTC were significantly higher than those observed in the tested model breast cancer cell lines (e. g. SKBR-3 or BT474). HER2 amplifications in single DTC conferred a strong prognostic impact on survival (p = 0.008). Furthermore, we observed a significant association between the shortness of survival and the extent of increased HER2 copy numbers (p = 0.021). Conclusion: The HER2 amplification in DTC is strongly linked to systemic esophageal cancer progression and should represent a suitable therapeutic target. Our single cell qPCR assay provides an excellent tool to adjust HER2 directed therapies directly on the actual targets for adjuvant systemic therapies, the disseminated tumor cells.

Prior in vivo studies supported the concept that Mallory bodies (MBs) are aggresomes of cytokeratins 8 and 18. However, to test this hypothesis an in vitro model is needed to study the dynamics of MB formation. Such a study is difficult because MBs have never been induced in tissue culture. Therefore, MBs were first induced in vivo in drug-primed mice and then primary cultures of hepatocytes from these mice were studied. Two approaches were utilized: 1. Primary cultures were transfected with plasmids containing the sequence for cytokeratin 18 (CK 18) tagged with green fluorescent protein (GFP). 2. Immunofluorescent staining was used to localize the ubiquitin-proteasome pathway components involved in MB-aggresome complex formation in primary hepatocyte cultures. The cells were double stained with a ubiquitin antibody and one of the following antibodies: CK 8, CK 18, tubulin, mutant ubiquitin (UBB+1), transglutaminase, phosphothreonine, and the 20S and 26S proteasome subunits P25 and Tbp7, respectively. In the first approach, fluorescence was observed in keratin filaments and MBs 48 h after the cells were transfected with the CK 18 GFP plasmid. Nascent cytokeratin 18 was preferentially concentrated in MBs. Less fluorescence was observed in the normal keratin filaments. This indicated that MBs continued to form in vitro. The immunofluorescent staining of the hepatocytes showed that CK 8 and 18, ubiquitin, mutant ubiquitin (UBB+1), P25, Tbp7, phosphothreonine, tubulin, and transglutaminase were all located at the border or the interior of the MB. These results support the concept that MBs are aggresomes of CK 8 and CK18 and are a result of inhibition of the ubiquitin-proteasome pathway of protein degradation possibly caused by UBB+1.

Early disseminated tumor cells (DTC) are potential precursor cells of later metastases and therefore the ultimate target for adjuvant therapies. To assess if EpCAM is a useful target for elimination of DTC in lymph nodes (LN) and bone marrow (BM), we determined the frequency of EpCAM+ and CK+DTC in these entities. Methods: After establishment of a protocol for double immunofluorescence labeling for simultaneous visualization of EpCAM expression on CK+ cells, we applied this protocol to a series of BM and LN preparations from operable esophageal cancer patients. After isolation of positively stained cells, their genomic DNA was globally amplified using the Mse-adapter PCR method. Then we applied comparative genomic hybridization (CGH) for the genome-wide screening of DNA-gains and-losses.