Statistical Methods Behind the CancerSEEK Blood Test and the DYNAMICS Study

• Activity Number: 596 - Statistical and Mathematical Methods in Cancer Etiology and Cancer Early Detection
• Type: Invited
• Date/Time: Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Kamel Lahouel*
• Companies: Johns Hopkins University
• Keywords: Cancer Early Detection; Classification; Blood Test for Cancer Detection; Cell-free DNA; Cancer Prediction; Bayesian Methods

Abstract

The classifying methods used in CancerSEEK, a novel blood test using circulating tumor DNA (ctDNA) for the early detection of eight different cancer types, will be presented (Cohen et al. Science 2018, 359(6378):926-930). The blood test uses a combination of features derived from genetic alterations and protein biomarkers: mutations in cell-free DNA and levels of circulating proteins. The performance of the classifier is assessed via two different tasks. The first task consists in classifying patients with cancer vs healthy controls (median sensitivity across the eight tissues 70%, with 99% specificity). The second task, for the positive cases only, consists in localizing the cancer site. The current developments of this algorithm, yielding improvements in its performance, will also be introduced. Finally, we will present the Bayesian statistical approach developed for DYNAMICS, a method for the early detection of minimal residual disease via ctDNA after stage II colon cancer resection.