Individualized Disease Diagnostics: Improving Diagnostic Tests with Genomic/Genetic/Clinical Information

• Activity Number: 457
• Type: Topic Contributed
• Date/Time: Wednesday, August 3, 2011 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Meijuan Li*+
• Companies: U.S. Food and Drug Administration/CDRH
• Keywords: Individualized Disease Diagnostics ; cutoff ; diagnostics medicine ; predictive value ; diagnostic accuracy ; overdiagnosis and overtreatment

Abstract

A patient's certain clinical or genetic/genomic characteristics may affect his diagnostic test level independent of his disease status. However, a uniform cutoff (i.e. a population based cutoff) regardless of patients' characteristics is often used in diagnostics medicine where a cutoff is defined as the threshold at which the test differentiates a positive from a negative outcome. Consequently, such test can be controversial because of its modest predictive value and diagnostic accuracy; the potential overdiagnosis and overtreatment of the disease. Therefore, using more individualized thresholds based on a test; molecular measures; and patient age, ethnicity, and family history of disease can improve the predictive ability and the accuracy of this test. In this talk, we will discuss issues related to population-based diagnostic approaches and discuss some statistical methods for individualized diagnostic approaches.