Chip Based Diagnosis of Problematic Liver Tumors

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Precise tumor diagnosis and classification is the first step in cancer management. While many tumor biopsies are diagnostic and form the cornerstone of cancer therapy, classification of tumor type and site of origin, is an ever-present clinical challenge. It is estimated that up to 10% of all tumors have no defined primary site of origin and that thousands of dollars are spent per case to identify the site of origin with limited overall success. The current standard of pathologic practice, using morphologic criteria and a panel of semi-quantitative immunohistochemical (IHC) analyses, is often limited in its capacity to define tumor type or site of origin. Moreover, the diagnosis of metastatic lesions can be quite difficult when no primary site of origin has been identified (unknown primary cancers). Since therapy is often based on site of origin, there is a clear need for the identification and validation of a classifier that will cleanly distinguish these histologically similar tumor types and augment standard pathological techniques in making the diagnostic call. We have recently demonstrated the feasibility of using gene expression profiling to discriminate 21 different tumor types with an accuracy of 88%. The performance of this classifier, however, was principally limited by the use of multiple platforms for analysis. This proposal seeks to build a new gene expression classifier on a single, commercially available, genome-wide oligonucleotide platform, with a focus on the most problematic primary and metastatic tumors of the liver. To demonstrate the clinical utility of this approach, we plan to validate the classifier with independent test sets that will also be profiled by standard IHC approaches; the results of molecular classification will then be compared head to head with standard pathological classification for accuracy of diagnosis. To further translate the technology, we will select core classifier genes and perform validation with real time quantitative PCR. Finally, the molecular classifiers will be tested using prospectively acquired biopsy samples on tissues of known and unknown sites of origin.