Lung Cancer Diagnostic Test Validation in FNA Specimens

University Of Toledo Health Sci Campus, Toledo OH

Investigators

Linked publications, trials & patents

Abstract

DESCRIPTION (provided by applicant) Lung cancer is the most common cancer cause of death in this country. At the time of diagnosis, it usually is too advanced to be treated effectively with surgery. In these advanced cases, diagnosis is made by the least invasive, non-surgical methods available. This typically involves morphologic analysis of cells obtained by transthoracic fine needle aspirate (FNA) or bronchoscopic biopsy. Although it is the standard of care, cytomorphologic analysis of these small specimens is only approximately 80% sensitive for lung cancer, does not always allow morphologic sub-classification, and provides no information regarding expected survival outcome. Gene expression analysis is likely to improve diagnostic accuracy and provide information that will enable improved management. However the small cytology specimens obtained by FNA or bronchoscopy are not sufficient for most gene expression assay methods. In recent studies, we have developed methods for measuring expression of hundreds of genes in RNA from small cytology specimens, including improved techniques for preserving and extracting RNA combined with a quantitative RT-PCR method optimized for clinical diagnostic testing, standardized RT (StaRT)-PCR. Using these methods in preliminary tests, the E2F1 x c-myc/p21 interactive gene expression index distinguished 28 primary lung cancer samples from 27 primary normal lung samples with 100% sensitivity and specificity. Recently, microarray gene expression analysis was used to identify sets of genes associated with a) histologic class or b) time of survival in adenocarcinoma. This is a proposal to conduct a validation study of these molecular diagnostic tests for lung cancer, histologic class or survival prediction in lung cancer samples. In the R21 phase we will ensure, through the attainment of carefully detailed milestones, that the proposed diagnostic tests have the characteristics necessary to test the hypotheses proposed in the R33 phase. In the R33 phase we will test a hypothesis in each of three aims. AIM 1. Test the hypothesis that in FNA samples the c-myc x E2F-1/p21 index diagnoses lung cancer with higher accuracy than cyto-morphologic analysis and thereby has the potential to reduce the number of diagnostic tests done in lung cancer patients. AIM 2. Test the hypothesis that a set of genes measured by StaRT-PCR surgical or cytologic samples will discriminate SCLC from NSCLC with greater accuracy than cyto-morphology. AIM 3. Test the hypothesis that adenocarcinoma survival is correlated with expression of specific sets of genes measured by StaRT-PCR in surgical or cytologic samples, thereby validating use of FNA samples. We expect that the proposed project will validate our methods for obtaining detailed gene expression information from each individual tumor prior to treatment, whether from cytologic or histologic specimens, so that gene expression data may be correlated with overall outcome and response to treatment in every case. We expect that diagnostic tests resulting from these studies will reduce cost and suffering related to lung cancer diagnosis and treatment and will lead to improved design of clinical trials.