Use of gene expression profiling to determine prognosis and therapeutic targets for patients with appendiceal carcinoma.

418 Background: Appendiceal carcinomas are rare neoplasms that often present as peritoneal surface malignancies. The tumors are treated similarly to primary colorectal metastases, despite having different biological environments. The aim of this study is to utilize genomic analyses to understand the underlying biology of appendiceal carcinoma in order to better guide prognosis and therapy. METHODS: Forty-one metastatic human tumors to the peritoneum (26 appendiceal carcinoma, 15 colorectal carcinoma) were subjected to global gene expression analysis. Unsupervised hierarchical clustering was used to differentiate phenotypes based on differential gene expression. Phenotypes were further analyzed to find differences in oncogenic pathway deregulation and findings were validated using gene specific inhibitors in vitro. RESULTS: Unsupervised hierarchical clustering revealed three distinct phenotypes, two consisting of predominantly low grade appendiceal samples (Cluster 1 and Cluster 2) and one consisting of predominantly colorectal samples (Cluster 3). Cluster 1 consisted of patients with good prognosis and Cluster 2 consisted of patients with poor prognosis (p = 0.046). Further characterization of Cluster 1 and Cluster 2 with oncogenic pathway signatures identified increased deregulation of the EGFR pathway in the poor prognosis cluster. Similarly, colon cancer cell lines with the poor prognosis phenotype were found to be more sensitive to gefitinib (EGFR inhibitor) (p = 0.038). CONCLUSIONS: The use of geneexpression profiling can be used to identify prognosis and novel therapeutic targets for patients with appendiceal carcinoma. These methods can serve as models to understand the biology of appendiceal carcinoma and to assist in the design of future trials for patients with appendiceal cancer. No significant financial relationships to disclose.

Duke Scholars

Author Dan German Blazer III Surgical Oncology