The analysis population is defined as all the set of de-identified patient results received from the practices who met the inclusion / exclusion criteria. A target recruitment of 2000 subjects with an expected loss of 20% (400 subjects) and an overall event rate of 30% (480 subjects of 1600) will result in 480 cases and 1120 controls, where cases are defined as those patients with NCCN unfavorable intermediate risk or worse disease. With a significance level alpha of 0.05, the number of cases above will result in a statistical power (1-ß=0.2) of at least 80% to show a significant increase in performance of the full assay over clinical variables alone, assuming that performance characteristics are similar to those observed in a retrospective, independent training cohort and that the patients in the SAFELY cohort resemble this same training population.
