Evaluating Prognostic Accuracy of Biomarkers for Prostate Cancer Active Surveillance Study

Evaluating Prognostic Accuracy of Biomarkers for Prostate Cancer Active Surveillance Study

May 16, 2018

Address

727 E. Tyler St.
Tempe, AZ 85287

Location

Biodesign Institute, Auditorium

Date and Time

May 25, 2018, 2:00 pm (Length: 1 hour 0 minutes)

iCal Download

Yunro “Roy” Chung, Ph. D., Public Health Sciences Division, Fred Hutchinson Cancer Research Center

Active Surveillance for patients diagnosed with low-grade cancer has become a population management strategy for reducing overtreatment. Novel biomarkers have been sought for decision making during Active Surveillance. The accuracy of the decision tools for such a purpose needs to be quantified. The evaluation can be challenging due to the lack of definite measurement of outcomes on all subjects in the study. Yunro Chung proposes an estimation procedure for the time-dependent receiver operating characteristic curve by additionally using a failure-time surrogate outcome. The resulting estimator is asymptotically normal and efficient than the complete-case estimator, with a less strict assumption about the surrogate outcome. The results are demonstrated by conducting simulation studies. The method is illustrated with the data from the Canary Prostate Active Surveillance Study.