Table 4.

Cox proportional hazards regression model for recurrence following immediate and delayed radical prostatectomy based on CaPSURE data and for prostate cancer death based on T-stage ≤ T2a JH-PCM data

EndpointCovariateCoefficientHR95% Confidence intervalP value
Recurrence under immediate radical prostatectomy (fit to low-risk CaPSURE cohort)Age0.011.01(0.98, 1.03)0.527
PSA0.151.16(1.07, 1.25)<0.001
Recurrence under delayed radical prostatectomy (fit to T-stage ≤ T2a CaPSURE cohort)Age0.011.01(1.00, 1.03)0.187
PSA0.031.03(1.02, 1.03)<0.001
Gleason score
≤6Ref.
70.471.59(1.25, 2.03)<0.001
≥81.434.16(2.92, 5.93)<0.001
Prostate cancer mortalityTime to recurrence−0.440.65(0.51, 0.82)<0.001
Gleason score
≤6Ref.
70.561.75(0.96, 3.17)0.066
≥80.892.43(1.21, 4.86)0.012

NOTE: The immediate radical prostatectomy model was fit to 2,150 cases with low-risk disease (T-stage ≤ T2a, Gleason score ≤ 6, and PSA level ≤ 10 ng/mL); the delayed radical prostatectomy model, used to simulate times to recurrence following active surveillance, was fit to 3,470 cases with T-stage ≤ T2a disease.