VADuke 4x8 Poster Template

1
WHAT ARE THE FACTORS ASSOCIATED WITH A SHORT PSA
DOUBLING TIME? A REPORT FROM THE SEARCH
DATABASE GROUP
Anna E. Teeter, Lionel L. Bañez, William J.
Aronson, Joseph C. Presti, Jr, Martha K. Terris,
Christopher L. Amling, Christopher J. Kane, and
Stephen J. Freedland
Veterans Affairs Medical Centers (Durham, West
Los Angeles, Palo Alto and Augusta) The Duke
Prostate Center, Duke University Medical Center
University of California, Los Angeles University
of California, San Diego Stanford University,
Palo Alto University of Alabama, Birmingham
Medical College of Georgia, Augusta
Abstract
Background
Results
Summary of Findings
Introduction A short prostate specific antigen
doubling time (PSADT) following biochemical
recurrence after radical prostatectomy (RP) has
been shown to portend a poor prognosis for men
with prostate cancer. We sought to determine
which demographic and clinicopathological
variables were predictive of a short PSADT in a
cohort of men with clinically localized prostate
cancer treated with RP. Methods Data on 856
men from the Shared Equal Access Regional Cancer
Hospital (SEARCH) database who underwent RP for
node-negative prostate cancer between 1988 and
2003 were used in the analysis. We used logistic
regression analysis to determine the independent
factors associated with a short PSADT (lt9 months)
versus a longer PSADT (gt9 months) or no
recurrence. The variables considered for
analysis included age, race, logarithmically
transformed pre-operative PSA, body mass index,
year of surgery, pathologic Gleason sum,
extraprostatic extension, surgical margin status
and seminal vesicle invasion. Results On
multivariate analysis, higher pre-operative PSA
(odds ratio (OR) 2.20 95 confidence interval
(CI) 1.52-3.19 plt0.001), pathological Gleason
sum of 8-10 (OR 4.70 95 CI 2.11-10.43 plt0.001)
and 7 (OR 2.11 95 CI 1.09-4.08 p0.026) tumors
with extraprostatic extension and/or positive
surgical margins (OR 2.08 95 CI 1.48-3.91
p0.023) and seminal vesicle invasion (OR 3.26
95 CI 1.48-7.21 p0.003) were all independent
predictors of a short PSADT. Based upon these
risk factors, we developed a table to estimate
the risk of recurring with a PSADT lt9
months. Conclusions The factors which are
invariably used to predict overall biochemical
recurrence (high PSA, high grade and adverse
pathology) following RP also predict aggressive
recurrence.
Multivariate analysis of factors predicting
aggressive recurrence (PSADT lt9 month) PSA
OR 95 CI p-value
lt10 Reference 10-19.9 2.33 1.30-4.19
lt0.001 gt20 5.67 2.91-11.07 lt0.001 Pat
hological Gleason sum 2-6 Reference 7
2.10 1.08-4.08 0.028 8-10 4.80 2.14
-10.80 lt0.001 Pathologic Stage
Grouping Organ-confined, margin
negative Reference EPE and/or positive
margins 2.07 1.10-3.90 0.024 Seminal
vesicle invasion 3.17 1.43-7.01 0.004

• Short PSADT at the time of recurrence after
  radical prostatectomy predicts poor prognosis
• PSADT not known until recurred and have at least
  2 PSA values after recurrence
• Thus, knowing PSADT is a delayed event

• PSA, Gleason sum, and pathological stage
  reliably predict aggressive recurrence
• AUC of 0.79 for model predicting aggressive
  recurrence

Objective

• Identify clinical, demographic, and pathologic
  factors predictive of an aggressive recurrence
  (i.e. recurrence with a PSADT lt9 months)

Conclusions

• Aggressive recurrence can be reliably predicted
  based upon standard clinical and pathological
  variables
• PSA
• Pathological Gleason sum
• Pathological findings

Risk of aggressive recurrence (PSADT lt9
months) Pathologic
Gleason sum   2-6 7
8-10 PSA lt10 ng/ml Organ-confined, margin
negative 2 (1-3) 4 (2-7) 8 (4-16) EPE
and/or positive margins 4 (2-7) 7
(4-12) 15 (8-26) Seminal vesicle invasion 5
(2-12) 11 (5-20) 21 (11-38) PSA gt10 ng/ml and
lt20 ng/ml Organ-confined, margin negative 4
(2-8) 8 (4-15) 17 (8-32) EPE and/or positive
margins 8 (4-15) 15 (10-23) 29
(18-45) Seminal vesicle invasion 12
(5-24) 22 (12-35) 39 (23-57) PSA gt20
ng/ml Organ-confined, margin negative 9
(4-19) 18 (9-31) 33 (16-56) EPE and/or
positive margins 17 (8-32) 31 (19-45) 50
(31-70) Seminal vesicle invasion 24
(11-45) 40 (24-59) 61 (39-79)
Materials and Methods

• 856 men from the SEARCH Database, treated with
  radical prostatectomy for node-negative prostate
  cancer between 88 and 03
• Stepwise logistic regression analysis to
  identify independent factors associated with an
  aggressive recurrence (PSADT lt9 months) vs.
  non-aggressive (PSADT gt9)/no recurrence.
• Variables entered into the model were age,
  race, logPSA, BMI, surgery year, pathologic
  Gleason sum, extraprostatic extension, surgical
  margin status, and seminal vesicle invasion

Acknowledgements
This study was supported by the Department of
Veterans Affairs, Department of Defense, National
Institutes of Health, the Georgia Cancer
Coalition, and the American Urological
Association Foundation/Astellas Rising Star in
Urology Award.