Genotype and phenotype predictors in therapy response in renal cell carcinoma

点击翻译按钮获取中文摘要

ABSTRACT Therapy of renal cell carcinoma (RCC) has been transformed in recent years by the efficacy of targeted agents that inhibit kinases involved in critical cellular signaling pathways and VEGF, a primary driver of tumor angiogenesis. However, patient response to these therapeutics is highly variable and currently cannot be predicted based on clinical or pathological data or available laboratory/genetic testing. The goal of this proposal is to develop predictors of therapeutic response for patients with the most common subtype of RCC, clear cell (ccRCC), based on novel tests. Our approach is based primarily on the hypothesis that the targeted agents used in ccRCC therapy - bevacizumab, sunitinib and sorafenib - inhibit tumor vessel activation and that the activation status and stability of tumor vessels are major determinants of response. We also hypothesize that the phenotype of ccRCC tumor vessels is linked to the molecular pathobiology of the tumor cells. In particular, as these drugs also can inhibit tumor cell signaling and modulate their behavior, ccRCC tumor cell signaling, HIF-1/HIF-2 (hypoxia-inducible factor) expression and VHL function are potential determinants of response. To examine parameters of vascular phenotype, tumor cell phenotype and VHL genotype as predictors of response to therapy, ccRCC specimens will undergo multiplex immunostaining for the appropriate biomarker antigens and be analyzed by a novel computer-assisted image analysis system that objectively quantifies analyte staining on a cellular (cytometric) basis as well as by traditional pixel-based analysis. These studies will be performed on tumors of ccRCC patients treated with single-agent bevacizumab, sunitinib or sorafenib in ongoing multi-institutional phase II (ECOG2804) and phase III (ECOG2805) clinical trials, using tumor blocks from a subset of 90, 170 and 170 appropriate ccRCC patients for therapy with the respective drugs. The specific aims of this proposal are (Aim 1) to analyze the vascular and endothelial cell activation phenotype and vessel pericyte coverage in ccRCC tumors; (Aim 2) to analyze tumor cell signaling, HIF phenotype and VHL genotype in ccRCC tumors; and (Aim 3) to correlate parameters quantified in the prior aims for relationships to each other, to therapeutic outcome and to develop parsimonious predictive models of therapeutic response using biostatistical and bioinformatics approaches. The results of these studies should allow identification of the most appropriate drugs for treating individual patients with ccRCC and assist in the rational development of second-line and combination drug therapies. Beyond ccRCC, the targeted agents under study are used in the therapy of an ever expanding number of cancers, and the response predictors developed for ccRCC, where these agents are best studied because they are used as single-agents, may be useful for predicting response of these other cancers to combination therapy incorporating the targeted agents.