Proteomic Predictors Of AASK Renal Disease Progression

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Clinical treatment studies have demonstrated that there are several potential therapies including converting enzyme inhibitors, tight diabetes control and blood pressure control that can slow the progression of renal disease in several disease states. However in trials such as AASK the rate of progression of renal disease and incidence of ESRD remains substantial even in those subjects receiving optimal therapy. Current data provide little in the way of obvious therapeutic targets to ameliorate this inexorable progression. Recent studies predominately in ovarian and prostate cancer have demonstrated that serum proteomic patterns generated by Surface-enhanced Laser Desorption/Ionization time of flight (SELDI-tof) mass spectrometry can identify cancer patients versus normal subjects with high specificity and sensitivity. The Logical Analysis of Data (LAD) technique, developed by members of our group, has substantially improved the sensitivity and specificity for identifying ovarian cancer when applied to the same data sets. Further, this technique is able to identify predictive/prognostic patterns in clinical data sets. This is an exploratory study on a subset of AASK subjects to determine whether we will be able to develop a model that uses proteomic data to predict progression of renal disease and whether we will be able to identify predictive protein peaks from our model. We hypothesize that there are serum proteomic patterns that can be generated using the SELDI technique that will predict either progression or non-progression of renal disease in AASK subjects and that LAD can identify these patterns. We will use stored specimens from the AASK trial to test this hypothesis. At the minimum, this will allow identification of the subpopulation of patients with renal failure at risk for progression. Eventual isolation and identification of the proteins comprising the pattern may provide new targets for the therapy of progressive renal disease. We also hypothesize that LAD will detect patterns of baseline values in the AASK data set that will predict AASK outcomes; we will re-examine the AASK data set with LAD for this aim.