AUG 21, 2013 03:00 PM PDT

Discovery, Development, and Clinical Validation of Gene Expression Scores for Transplant Rejection and Coronary Artery Disease

  • Chief Scientific Officer, CardioDx, Inc.
      Steven Rosenberg joined CardioDx in 2006 and is responsible for the scientific staff and for research and development programs in the areas of genomics, genetics, and informatics. In addition, Steven is currently on the Board of Directors of KMT Hepatech. Prior to joining CardioDx, Steven served for three years as the Chief Scientific Officer of XDx, Inc., a molecular diagnostics company focused on diseases of the immune system. He led the group responsible for developing AlloMap, a molecular diagnostic for heart transplant patients, the first genomics-based test in transplant immunology. Prior to XDx, Steven spent a year as a consultant and visiting scholar at the University of California, Berkeley. From 1981 to 2001 he was at Chiron Corporation, where he was named one of only three Chiron Research Fellows in the company's history. He received his Ph.D. in biochemistry from the University of California, Berkeley, and is the author of more than 40 scientific publications, and an inventor of more than 20 U.S. patents.


    Introduction Although the focus of the genomics community has largely been on DNA polymorphisms which affect disease risk, gene expression, especially of blood cells, has the potential to reflect genetic and environmental factors and to inform on both current and future disease status. The immune system dysregulation underlying many human diseases, including transplant rejection, and coronary artery disease (CAD), coupled with development of global blood gene expression profiling and quantitative RT-PCR, has enabled the clinical use of blood-based gene expression tests. Methods Blood samples were obtained from heart transplant or coronary disease patients and paired with results from gold standards of pathologist core-lab read endomyocardial biopsies (EMB) or quantitative coronary angiographic analysis on invasive angiograms, respectively. Discovery, development and validation steps were used to identify genes, develop algorithms, and validate algorithm performance with pre-specified statistical and clinical analysis plans. Development and validation cohorts were obtained in prospective multi-center trials. Finally, the performance and impact of these tests on usual care was assessed in observational or interventional studies. Results For heart transplant rejection largely independent cohorts were used to discover and develop an 11 gene algorithm to discriminate high-grade rejection from quiescence. In a set of independent patient blood samples the ROC curve AUC was 0.72; subsequent comparison of this algorithm with EMB in a prospective interventional trial met the primary endpoint of non-inferiority with a significant reduction in EMB. For CAD diagnosis, after gene discovery by microarrays, independent cohorts from a multi-center angiographic study were used to develop and validate a 23-gene whole blood RT-PCR based algorithm to discriminate obstructive CAD, with a second validation study assessing performance in patients referred for myocardial perfusion imaging. ROC AUC in these two validation studies were 0.70 and 0.79, respectively. Conclusions Blood-based gene expression tests have been developed and are being using clinically.

    Show Resources
    Loading Comments...