Leonora de Boo1, Melania Pintilie2, Paul Yip3, Jack Baniel4, Neil Fleshner5, David Margel6. 1. Division of Urology, Department of Surgical Oncology,University Health Network, Toronto, ON and Erasmus MC-beurs, Erasmus University Rotterdam, Netherlands; 2. Clinical Study Coordination and Biostatistics, University Health Network, Toronto, ON; 3. Laboratory Medicine Program, University Health Network, Toronto, ON; 4. Division of Urology, Rabin Medical Center and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel; 5. Division of Urology, Department of Surgical Oncology, University Health Network, Toronto, ON; 6. Division of Urology and Davidoff Cancer Center, Rabin Medical Center, Beilinson Hospital, Israel.
Abstract
INTRODUCTION: In this study, we estimated the time from first detectable prostate-specific antigen (PSA) following radical prostatectomy (RP) to commonly used definitions of biochemical recurrence (BCR). We also identified the predictors of time to BCR. METHODS: We identified subjects who underwent a RP and had an undetectable PSA after surgery followed by at least 1 detectable PSA between 2000 and 2011. The primary outcome was time to BCR (PSA ≥0.2 and successive PSA ≥0.2) and prediction of the rate of PSA rise. Outcomes were calculated using a competing risk analysis, with univariable and multivariable Fine and Grey models. We employed a mixed effect model to test clinical predictors that are associated with the rate of PSA rise. RESULTS: The cohort included 376 patients. The median follow-up from surgery was 60.5 months (interquartile range [IQR] 40.8-91.5) and from detectable PSA was 18 months (IQR 11-32). Only 45.74% (n = 172) had PSA values ≥0.2 ng/mL, while 15.16% (n = 57) reached the PSA level of ≥0.4 ng/mL and rising. On multivariable analysis, the values of the first detectable PSA and pathologic Gleason grade 8 or higher were consistently independent predictors of time to BCR. In the mixed effect model rate, the PSA rise was associated with time from surgery to first detectable PSA, Gleason score, and prostate volume. The main limitation of this study is the large proportion of patients that received treatment without reaching BCR. It is plausible that shorter estimated median times would occur at a centre that does not use salvage therapy at such an early state. CONCLUSION: The time from first detectable PSA to BCR may be lengthy. Our analyses of the predictors of the rate of PSA rise can help determine a personalized approach for patients with a detectable PSA after surgery.
INTRODUCTION: In this study, we estimated the time from first detectable prostate-specific antigen (PSA) following radical prostatectomy (RP) to commonly used definitions of biochemical recurrence (BCR). We also identified the predictors of time to BCR. METHODS: We identified subjects who underwent a RP and had an undetectable PSA after surgery followed by at least 1 detectable PSA between 2000 and 2011. The primary outcome was time to BCR (PSA ≥0.2 and successive PSA ≥0.2) and prediction of the rate of PSA rise. Outcomes were calculated using a competing risk analysis, with univariable and multivariable Fine and Grey models. We employed a mixed effect model to test clinical predictors that are associated with the rate of PSA rise. RESULTS: The cohort included 376 patients. The median follow-up from surgery was 60.5 months (interquartile range [IQR] 40.8-91.5) and from detectable PSA was 18 months (IQR 11-32). Only 45.74% (n = 172) had PSA values ≥0.2 ng/mL, while 15.16% (n = 57) reached the PSA level of ≥0.4 ng/mL and rising. On multivariable analysis, the values of the first detectable PSA and pathologic Gleason grade 8 or higher were consistently independent predictors of time to BCR. In the mixed effect model rate, the PSA rise was associated with time from surgery to first detectable PSA, Gleason score, and prostate volume. The main limitation of this study is the large proportion of patients that received treatment without reaching BCR. It is plausible that shorter estimated median times would occur at a centre that does not use salvage therapy at such an early state. CONCLUSION: The time from first detectable PSA to BCR may be lengthy. Our analyses of the predictors of the rate of PSA rise can help determine a personalized approach for patients with a detectable PSA after surgery.
Authors: Stephen J Freedland; Elizabeth B Humphreys; Leslie A Mangold; Mario Eisenberger; Frederick J Dorey; Patrick C Walsh; Alan W Partin Journal: JAMA Date: 2005-07-27 Impact factor: 56.272
Authors: Andrew J Stephenson; Michael W Kattan; James A Eastham; Zohar A Dotan; Fernando J Bianco; Hans Lilja; Peter T Scardino Journal: J Clin Oncol Date: 2006-08-20 Impact factor: 44.544
Authors: Stephen J Freedland; William B Isaacs; Elizabeth A Platz; Martha K Terris; William J Aronson; Christopher L Amling; Joseph C Presti; Christopher J Kane Journal: J Clin Oncol Date: 2005-10-20 Impact factor: 44.544
Authors: Thomas Wiegel; Dirk Bottke; Ursula Steiner; Alessandra Siegmann; Reinhard Golz; Stephan Störkel; Norman Willich; Axel Semjonow; Rainer Souchon; Michael Stöckle; Christian Rübe; Lothar Weissbach; Peter Althaus; Udo Rebmann; Tilman Kälble; Horst Jürgen Feldmann; Manfred Wirth; Axel Hinke; Wolfgang Hinkelbein; Kurt Miller Journal: J Clin Oncol Date: 2009-05-11 Impact factor: 44.544
Authors: S García-Barreras; F Rozet; I Nunes-Silva; V Srougi; R Sanchez-Salas; E Barret; M Galiano; X Cathelineau Journal: Clin Transl Oncol Date: 2017-12-14 Impact factor: 3.405