BACKGROUND: The objective of this study was to evaluate the adequacy in calculating prostate-specific antigen doubling time (PSADT) using an ultrasensitive assay of prostate-specific antigen (PSA) for biochemical failure after radical prostatectomy (RP). METHODS: In this research, we included 52 consecutive patients who experienced more than or equal to three consecutive increases in ultrasensitive PSA values after RP at two institutions. PSADT is to be calculated by fitting a linear model to post-nadir longitudinal logarithm ultrasensitive PSA (lnPSA) values. The goodness of fit of linear model for longitudinal lnPSA values was evaluated by applying a mixed effect model for all patients or two groups divided by median PSADT. RESULTS: Median PSADT was 4.7 months. A mixed effect model showed that the fit of linear model for longitudinal lnPSA values was not good in lower ultrasensitive PSA values for all patients (p < 0.0001) or the group with a value longer than the median PSADT (p < 0.0001), while it was not worse for the group with a value shorter than the median PSADT (p = 0.079). CONCLUSIONS: These findings suggest that calculating PSADT using an ultrasensitive assay for biochemical failure after RP may still be open for discussion. (c) 2007 S. Karger AG, Basel.
BACKGROUND: The objective of this study was to evaluate the adequacy in calculating prostate-specific antigen doubling time (PSADT) using an ultrasensitive assay of prostate-specific antigen (PSA) for biochemical failure after radical prostatectomy (RP). METHODS: In this research, we included 52 consecutive patients who experienced more than or equal to three consecutive increases in ultrasensitive PSA values after RP at two institutions. PSADT is to be calculated by fitting a linear model to post-nadir longitudinal logarithm ultrasensitive PSA (lnPSA) values. The goodness of fit of linear model for longitudinal lnPSA values was evaluated by applying a mixed effect model for all patients or two groups divided by median PSADT. RESULTS: Median PSADT was 4.7 months. A mixed effect model showed that the fit of linear model for longitudinal lnPSA values was not good in lower ultrasensitive PSA values for all patients (p < 0.0001) or the group with a value longer than the median PSADT (p < 0.0001), while it was not worse for the group with a value shorter than the median PSADT (p = 0.079). CONCLUSIONS: These findings suggest that calculating PSADT using an ultrasensitive assay for biochemical failure after RP may still be open for discussion. (c) 2007 S. Karger AG, Basel.
Authors: Judd W Moul; Hans Lilja; O John Semmes; Raymond S Lance; Robert L Vessella; Martin Fleisher; Clarisse Mazzola; Mark J Sarno; Barbara Stevens; Robert E Klem; Jonathan E McDermed; Melissa T Triebell; Thomas H Adams Journal: Urology Date: 2012-10-26 Impact factor: 2.649