0022-5347/00/1631-0013/0

THE JOURNAL OF UROLOGY

Copyright 2000 by AMERICAN UROLOGICAL ASSOCIATION, INC.

Vol. 163, 1320, January 2000

Printed in U.S.A.

Original Articles
SERUM PROSTATE SPECIFIC ANTIGEN IS A STRONG PREDICTOR OF
FUTURE PROSTATE GROWTH IN MEN WITH BENIGN PROSTATIC
HYPERPLASIA
CLAUS G. ROEHRBORN,*, JOHN MCCONNELL, JAIME BONILLA,* SIDNEY ROSENBLATT,
PERRY B. HUDSON,* GHOLEM H. MALEK, PAUL F. SCHELLHAMMER, REGINALD BRUSKEWITZ,
ALVIN M. MATSUMOTO,*, LLOYD H. HARRISON,* HAROLD A. FUSELIER, PATRICK WALSH,*
JOHNNY ROY,* GERALD ANDRIOLE,* MARTIN RESNICK AND JOANNE WALDSTREICHER* FOR
THE PROSCAR LONG-TERM EFFICACY AND SAFETY STUDY
From the University of Texas Southwestern Medical Center at Dallas, Dallas, Texas, Department of Veterans Affairs, VA Medical Center, Bay Pines,
Florida, Irvine Clinical Research Center, Irvine, California, Jackson Foundation, Madison, Wisconsin, Eastern Virginia Medical School, Norfolk,
Virginia, University of Wisconsin, Madison, Wisconsin, Veterans Affairs Puget Sound Health Care System, Seattle, Washington, Wake Forest
University School of Medicine, Winston-Salem, North Carolina, Ochsner Clinic, New Orleans, Louisiana, Johns Hopkins Hospital, Baltimore,
Maryland, University of Oklahoma, Oklahoma City, Oklahoma, Washington University, St. Louis, Missouri, University Hospital of Cleveland,
Cleveland, Ohio and Department of Clinical Research Endocrinology and Metabolism, Merck Research Laboratories, Rahway, New Jersey

ABSTRACT

Purpose: We analyze patterns of prostate growth in men diagnosed with benign prostatic
hyperplasia (BPH) and treated with placebo during 4 years, and determine which baseline
parameters were the strongest predictors of growth.
Materials and Methods: A total of 3,040 men were enrolled in the 4-year randomized, placebo
controlled Proscar Long-Term Efficacy and Safety study. Of these men a subgroup of 10%
underwent pelvic magnetic resonance imaging prostate volume measurement at baseline and
yearly thereafter. Absolute and percent volume changes during 4 years were calculated in the 164
placebo treated men in the subgroup. The ability of age, baseline prostate volume and prostate
specific antigen (PSA) to predict prostate growth in placebo treated patients was assessed by
multiple linear regression analyses, receiver operator characteristics curves, and evaluations of
growth stratified by tertiles of baseline serum PSA and decades of life.
Results: In placebo treated patients a steady increase in mean plus or minus standard deviation
prostate volume from year to year was noted (2.5 6 6.1, 4.9 6 6.8, 6.4 6 8.5 and 7.2 6 8.8 ml. at years 1,
2, 3 and 4, respectively). Mean volume changes at 4 years ranged from 29 to 130 ml. Mean percent
change from baseline ranged from 12.5% to 16.6% for men 50 to 59 years old to those 70 to 79 years old.
Baseline serum PSA was a strong predictor of growth with 7.4% to 22.0% change at 4 years from the
lowest to highest PSA tertiles. Annualized growth rates from baseline were 0.7 ml. per year for PSA 0.2
to 1.3, 2.1 for PSA 1.4 to 3.2 and 3.3 for PSA 3.3 to 9.9 ng./ml. Multiple linear regression analysis showed
that serum PSA was a stronger predictor of prostate growth than age or baseline prostate volume. All but
1 man with baseline serum PSA greater than 2.0 ng./ml. had prostate growth during 4 years, and 32.6%
of men with serum PSA less than 2.0 exhibited a decrease in volume.
Conclusions: Serum PSA is a stronger predictor of growth of the prostate in placebo treated patients
than age or baseline prostate volume. Since prostate volume is a risk factor for acute urinary retention
and the need for BPH related surgery, the ability of PSA to predict prostate growth may be an important
factor when considering individual treatment options for BPH. Such use of PSA represents a shift in
paradigm away from focusing solely on symptoms of BPH toward a more comprehensive approach with
consideration of predicting and preventing risk factors of BPH related outcomes.
KEY WORDS: prostatic hyperplasia, prostate-specific antigen, magnetic resonance imaging, prostate

Benign prostatic hyperplasia (BPH) is a common disorder

in aging men.1 Histological evidence of BPH is found in 60%

of men 60 to 69 years old, the majority of whom will eventually have lower urinary tract symptoms. Disease severity is
commonly measured with quantitative symptom frequency
and severity assessments,2, 3 which measure the impact of
disease and symptoms on activities of daily living,4 and dis-

Accepted for publication July 30, 1999.

Presented at annual meeting of American Urological Association,
San Diego, California, May 30 June 4, 1998.
* Financial interest and/or other relationship with Merck.
Financial interest and/or other relationship with Glaxo, Synthelabo, Abbott, U.S. Surgical, Vida Med, Dornier, Urologix and Pfizer.
Financial interest and/or other relationship with Serono,
Unimed, Biotek, Wyeth Ayerst, Organon and Smith Kline.
Merck, West Point, Pennsylvania.

Editors Note: This article is the first of 5 published in this

issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions
on pages 288 and 289.
13

14

SERUM PROSTATE SPECIFIC ANTIGEN PREDICTS PROSTATE GROWTH

ease specific quality of life assessment tools.5 An indirect

measure of disease severity is peak urinary flow rate which is
useful to assess the degree of bladder outlet obstruction,
although not specific enough to differentiate a weakening
detrusor muscle and outlet obstruction.6 10 Although the
term BPH suggests that the prostate is at the heart of the
pathophysiological condition, it has long been believed that
prostate volume alone is not useful to assess disease severity,
determine the need for treatment and/or help choose among
therapeutic interventions.11 The Agency for Health Care Policy and Research Guidelines for the diagnosis and treatment
of BPH and the International Consultation on BPH did not
recommend prostate volume assessment.5, 12 Nonetheless, it
has been shown that prostate volume increases with advancing age,13 and prostate size and shape often dictate treatment. For example, open prostatectomy is the surgical procedure of choice when the prostate gland is enlarged, while
transurethral resection of the prostate is the preferred surgical alternative in most other cases.
Recently it was recognized that baseline prostate volume
assessed during the initial evaluation of patients presenting
with lower urinary tract symptoms may have important ramifications. A meta-analysis of all available data from placebo
controlled, randomized finasteride trials of 1-year duration
demonstrated that therapeutic response to treatment with
the 5a-reductase inhibitor finasteride is volume dependent,
with larger prostates exhibiting better response.14 In addition, recent longitudinal community based studies have indicated that prostate volume is a strong predictor of episodes
of acute urinary retention. Men with a baseline prostate
volume of greater than 30 ml. are 3 times as likely to have
acute urinary retention compared to those with a volume of
less than 30 ml.15 In the 4-year placebo controlled Proscar
Long-Term Efficacy and Safety (PLESS) trial prostate volume and PSA were strong predictors of future episodes of
acute urinary retention as well as the need for BPH related
surgery in placebo treated patients with BPH.16
Because of the increasing risk of acute urinary retention
and need for surgery with increasing baseline prostate volume, knowing whether an individual presenting with symptoms of BPH has a lesser or greater tendency for further
prostate growth with time is of interest to clinicians and
patients. Such information would help decide whether to
initiate treatment or continue watchful waiting, and also
choose among therapeutic interventions. Baseline PSA, prostate volume and age were evaluated as risk factors for future
prostate growth in placebo treated patients from the PLESS
trial.
MATERIALS AND METHODS

Details of the overall study design, and the primary efficacy and safety results have been previously published.17, 18
A total of 3,040 men with clinical BPH diagnosed on the basis
of moderate to severe symptoms, a peak urinary flow rate of
less than 15 ml. per second with a voided volume of at least
150 ml. and an enlarged prostate gland on digital rectal
examination were enrolled in a 4-year study of finasteride
versus placebo. Men receiving a-blockers or antiandrogens
and those with a history of chronic prostatitis, recurrent
urinary tract infections, prostate or bladder cancer or surgery, or serum PSA greater than 10 ng./ml. were excluded
from study. Men with serum PSA between 4.0 and 9.9 ng./ml.

had to have a negative prostate biopsy before enrollment.

The present analysis is based only on placebo treated patients.
The study was approved by the Institutional Review Board
at all 95 participating centers and all men gave written
informed consent. After a 1-month, single-blind placebo
lead-in men were randomly assigned to receive placebo or 5
mg. finasteride daily. Symptoms, bothersomeness, adverse
events and flow rates were assessed every 4 months. Serum
PSA was measured every 4 months in year 1 and every 8
months during subsequent years at a central laboratory using the Hybritech* assay. Physical examinations, and routine hematological and serum chemistry studies were performed yearly. In a 10% subset of all patients at 13 centers
magnetic resonance imaging (MRI) and measurement of
prostate volume were performed at baseline and annually
thereafter. MRI was reviewed for quality assurance purposes
during the trial, and then read at the end of the study by the
central radiologist (J. B.) who was blinded to treatment allocation and time of imaging (that is baseline or followup).17
All statistical and receiver operating characteristics (ROC)
curve analyses were performed using computer software.
Simple linear and multiple linear regression analyses were
performed, with p ,0.05 considered significant.
RESULTS

A total of 164 patients were available at baseline with a

mean age of 63 6 6.1 years (range 50 to 77) and a mean
baseline prostate volume of 54.6 6 25.9 gm. (range 14 to 222)
on MRI (table 1). Mean baseline serum PSA was 2.7 6 2.1
ng./ml. (range 0.2 to 9.4). Of the patients 50 were 50 to 59
years old, 92 were 60 to 69 years old and 22 were older than
70 years (table 1). Patients dropped out of the study to
undergo BPH related surgery, to receive other active medical
therapy (for example finasteride or a-blockers), or because of
worsening symptoms, lack of improvement, adverse events,
logistical reasons or lost to followup.18 Baseline and followup
data for all parameters were available for 145 men at 1, 125
at 2, 107 at 3 and 91 at 4-year followup. A total of 83 men had
complete sets of prostate volume and serum PSA at baseline
and all followup times. A steady increase in prostate volume
was noted from a mean baseline of 54.6 6 25.9 gm. (fig. 1).
Mean increase in absolute volume was 2.5 6 6.1 ml. at 1,
4.9 6 6.8 at 2, 6.4 6 8.5 at 3 and 7.2 6 8.8 at 4 years, which
represent a percent increase of 5.2 6 11.8%, 9.0 6 11.1%,
11.8 6 14.5% and 14.1 6 17.5%, respectively.
Absolute and percent prostate volume changes were stratified by baseline PSA (tertiles) and age (decade of life) (table
1). Volume changes from baseline at the 4 followup times by
PSA (tertile) and decade of life are listed in table 2. Volume
changes ranged from 2.8% for men 50 to 59 years old to 6.4%
for those 70 to 79 years old at year 1 and from 12.0% for those
50 to 59 years old to 16.6% for those 70 to 79 years old at year
4. Absolute volume changes ranged from 1.1 ml. for men 50 to
59 years old to 3.1 for those 70 to 79 years old at year 1, and
from 5.8 to 9.6 at year 4 for these age groups (fig. 2, A). The
annualized growth rate was 1.45 ml. per year for men 50 to
59 years old and 2.4 for those 70 to 79 years old. Absolute
volume changes stratified by PSA ranged from 4.2% (1.1 ml.)
to 6.5% (3.9) for the lowest to the highest PSA tertile at year
* Hybritech, Inc., San Diego, California.

TABLE 1. Baseline data on all patients and stratified by decade of life

All Pts.
No. pts.
Mean baseline age 6 SD (range)
Mean baseline (ml.) prostate vol. 6 SD (range)
Mean baseline (ng./ml.) PSA 6 SD (range)

164
63 6 6.1 (50 to 77)
54.6 6 25.9 (14 to 222)
2.7 6 2.1 (0.2 to 9.4)

Age 5050 Yrs.

Age 6069 Yrs.

Age Older Than 70 Yrs

50
55.7 6 2.8
50.0 6 22.2 (23 to 146)
2.2 6 1.8 (0.2 to 7.6)

92
64.9 6 2.8
56.2 6 28.1 (14 to 222)
2.9 6 2.1 (0.3 to 9.4)

22
72.1 6 2.1
58.9 6 24.6 (30 to 117)
3.1 6 2.3 (1.0 to 9.4)

SERUM PROSTATE SPECIFIC ANTIGEN PREDICTS PROSTATE GROWTH

15

FIG. 1. Absolute volume changes during 4 years of followup with mean (thick line), median (thin line), 25th to 75th percentile (box), 10th
to 90th percentile (whiskers) and individual outliers.

TABLE 2. Absolute and percentage changes from baseline in

prostate volume stratified by PSA tertiles and decade of life at 12
to 48 months of followup
Mean 6 95% CI

Variable
Absolute changes by PSA:
0.21.3
1.43.2
3.39.4
% Changes by PSA:
0.21.3
1.43.2
3.39.4
Absolute changes by age:
5059 Yrs.
6069 Yrs.
7077 Yrs.
% Changes by age:
5059 Yrs.
6069 Yrs.
7077 Yrs.

12 Mos.

24 Mos.

36 Mos.

48 Mos.

1.1 6 1.6
2.5 6 1.8
3.9 6 2.1

2.5 6 1.2
5.5 6 2.3
7.0 6 2.7

1.6 6 1.6
8.7 6 3.0
10.6 6 3.7

2.8 6 1.8
8.3 6 3.5
13.3 6 4.2

4.2 6 3.7
5.1 6 3.3
6.5 6 3.7

6.8 6 3.3
9.4 6 3.4
11.3 6 4.0

4.8 6 4.3
15.7 6 4.6
16.7 6 5.7

7.4 6 4.8
16.2 6 7.4
22.0 6 6.9

1.1 6 2.1
3.0 6 1.4
3.1 6 2.8

4.5 6 2.7
5.5 6 1.6
3.7 6 2.9

4.4 6 3.6
6.8 6 2.1
9.8 6 4.1

5.8 6 2.6
7.5 6 3.1
9.6 6 4.9

2.8 6 3.9
6.2 6 2.7
6.4 6 5.1

7.9 6 3.4
10.1 6 3.0
7.8 6 4.4

8.1 6 5.3
12.8 6 4.0
16.3 6 7.2

12.0 6 5.0
14.8 6 6.4
16.6 6 9.1

1 but from 7.4% (2.8 ml. per year to 22.0% (13.3), respectively, at year 4 (fig. 2, B). Mean percent growth at 4 years
varied little among age groups compared to differences in
growth rates among patients when stratified by PSA tertiles.
The relationships among baseline age, prostate volume
and baseline serum PSA are shown in figure 3. The estimated
correlation between age and baseline prostate volume was
modest (correlation coefficient r 5 0.17, r2 5 0.027, p 5 0.04,
fig. 3, A), and the typical modest increase of serum PSA with
advancing age, which led to the development of age specific
reference ranges for PSA, was apparent (r 5 0.19, r2 5 0.037,
p 5 0.01, fig. 3, B). By far the strongest relationship was
between baseline prostate volume and PSA (r 5 0.53, r2 5
0.28, p ,0.001, fig. 3, C). This relationship is best expressed
as a double logarithmic presentation (fig. 3, D). Similarly, we
evaluated the relationship between baseline parameters and

percent volume change with time (fig. 4). There was little
relationship between age and percent volume change (r 5
0.09, r2 5 0.007, p 5 0.42, fig. 4, A). The relationship between
baseline prostate volume and percent volume change was
poor (r 5 0.014, r2 5 0.0002, p 5 0.9, fig. 4, B). The relationship between baseline prostate volume and absolute volume
change at year 3 was stronger (r 5 0.32, r2 5 0.1, p 5 0.003),
which is a reflection of the impact of baseline volume on the
potential absolute changes in volume. However, the strongest relationship existed between baseline serum PSA, and
absolute (r 5 0.35, p 5 0.001) and percent prostate volume
change (r 5 0.31, p 5 0.004, fig. 4, C).
Of 46 patients 15 (32.6%) who had baseline PSA less than
2.0 ng./ml. had a net decrease in prostate volume during 4
years. In contrast, only 1 patient who had baseline PSA
greater than 2.0 ng./ml. had a decrease while all others had
an increase in prostate volume. Using a cutoff of 5 ml. or
more to represent significant growth during 4 years, a ROC
curve analysis was performed to determine which of the
baseline parameters of age, prostate volume and serum PSA
best identified men destined to experience such growth. The
area under the curve representing the probability of correctly
distinguishing men with and without significant prostate
volume growth was 0.584 for age, 0.719 for baseline prostate volume and 0.787 for baseline serum PSA (fig. 5).
Multiple linear regression was performed using absolute or
percent volume change at 4 years as the dependent variable,
and age, baseline volume and baseline PSA as independent
variables. PSA was the strongest predictor of absolute changes
in prostate volume (age p 5 0.88, volume p 5 0.54, PSA p
,0.001) and percent changes in prostate volume (age p 5 0.58,
volume p 5 0.1, PSA p 5 0.001) with time. In contrast, age and
baseline volume contributed only marginally to the model. Data
were also analyzed by serum PSA tertiles at baseline (fig. 6).
While patients in the second (PSA 1.4 to 3.2 ng./ml.) and third
(3.3 to 12) tertiles had substantial volume increases during the
4-year followup, those with baseline PSA from 0 to 1.3 (lowest
tertile) had small volume increases.

16

SERUM PROSTATE SPECIFIC ANTIGEN PREDICTS PROSTATE GROWTH

FIG. 2. Mean percent prostate growth from baseline by year and decade of life (A), and PSA tertile (B)

DISCUSSION

To our knowledge this study is the longest randomized

clinical trial in men with BPH to date which documents
longitudinal changes in prostate volume growth with time in
placebo treated patients. The trial was designed using modern MRI and quality controls with a central reviewer blinded
to time and treatment assignment to provide for a high

degree of reproducibility and confidence in the data.17 The

164 placebo treated patients followed for up to 4 years demonstrated an overall annualized prostate growth rate of 1.8
ml. per year, ranging from 1.45, 1.88 and 2.40 for men 50 to
59, 60 to 69 and 70 to 79 years old, respectively. These growth
rates are higher than those described in the Olmsted County
study in randomly selected community dwelling men (0.4 ml.

SERUM PROSTATE SPECIFIC ANTIGEN PREDICTS PROSTATE GROWTH

17

FIG. 3. Relationships at baseline between age and baseline prostate volume (A, r 5 0.17, p 5 0.04), age and baseline serum PSA (B, r 5
0.193, p 5 0.01), baseline prostate volume and baseline PSA (C, r 5 0.534, p ,0.001), and baseline prostate volume versus baseline serum
PSA expressed as double logarithmic linear relationship (D). r2 Values derived from regression analyses.

per year in men 40 to 60 years old and 1.2 in men 60 to 80

years old).19 This difference is most likely due to the fact that
all men in the present study had clinically diagnosed BPH,
while the Olmsted County study is based on a community
population sample, suggesting that growth in those already
diagnosed with BPH is greater than in unselected community dwelling men. Interestingly, prostate growth in the
present cohort of men with BPH occurred steadily, with an
increase of 5.2% at 1, 9.0% at 2, 11.8% at 3 and 14.1% at 4
years.
Although they were almost eliminated from the routine
evaluation of BPH,12 prostate size assessment and consideration of future prostate growth have recently generated renewed interest for several reasons. Prostate size as well as
urinary flow rates, symptoms and age have been found to
predict health care seeking behavior for urinary symptoms in
men.20 In addition, a stronger correlation between the volume of the transition zone of the prostate, and symptom
severity, peak urinary flow rate and invasive pressure flow
studies has been described.21 A most important contribution
to our understanding of the importance of prostate size was
the Olmsted County study by Jacobsen et al which demonstrated a 3-fold increase in the risk of acute urinary retention
in men with prostate volume greater than 30 ml. at baseline.15 While data from the Olmsted County study are based
on community dwelling men randomly selected without regard to symptoms of BPH, results from the PLESS study also
indicated a strong association between prostate volume or
PSA, and the risk of acute urinary retention and a need for
BPH related surgery.16 The risk of either of these 2 complications occurring during 4 years in placebo treated patients
ranged from 8.9% to 22.0% when stratified by increasing

prostate volume (by tertiles in our MRI subset) and from

7.8% to 19.9% when stratified by increasing serum PSA (by
tertiles of the entire study population). Compared to all other
urological measures and assessment tools, such as symptom
scores, flow rate, residual urine volume and so forth, serum
PSA and prostate volume were the most powerful predictors
of spontaneous acute urinary retention in placebo treated
patients with an area under the curve of 0.70 and 0.81,
respectively.16 These observations refocused attention on
prostate size as an important predictor of BPH related outcomes and stressed the need for understanding which patients with BPH are at high risk for future increasing prostate volume growth and, by inference, future BPH related
outcomes.
Previous placebo controlled trials have provided limited
data on prostate volume growth due to shorter followup,
more variable ultrasound measurements and less rigorous
standardization than used in our study. The 2-year transrectal ultrasound prostate volume data are available on 197 of
707 patients with moderate symptoms of BPH in the 2-year
randomized, controlled trial at 59 centers in 5 Scandinavian
countries.22 Prostate volume increased from a baseline of
41.7 by 11.5 ml. for an annualized growth rate of 5.75 ml. in
these patients. Similar growth patterns have also been reported from the 2-year placebo controlled, randomized finasteride trials conducted in Canada.23 However, except for
these trials few data are available on prostate growth with
time in patients treated with watchful waiting or placebo. In
placebo controlled a-blocker trials prostate volume assessment was not typically done as no effect on prostate size is
expected from this therapy.24 When assessing all of these
other data sets the sample sizes and variability associated

18

SERUM PROSTATE SPECIFIC ANTIGEN PREDICTS PROSTATE GROWTH

FIG. 4. Relationship between age and percent volume change (A, r 5 0.09, p 5 0.42), baseline prostate volume and percent volume change
(B, r 5 0.014, p 5 0.9) and baseline serum PSA versus percent volume changes (C, r 5 0.31, p 5 0.004) at 4 years. Additional lines in C
indicate zero growth line (horizontal) and 2.0 ng./ml. PSA baseline (vertical). All but 1 patient with PSA greater than 2.0 ng./ml. had growth
during 4 years. r2 Values derived from regression analyses.

with transrectal ultrasonography at multiple centers without

a central reviewer must be considered.25
Our objective was to identify the key predictors of prostate
growth with time since they would likely be key predictors of
BPH related outcomes as well. Thus, we performed regression analyses between various baseline parameters and the
percent changes at 4 years. The relationships between baseline prostate volume or age and percent changes at 4 years
were weak, while those of absolute change and baseline prostate volume were stronger. In contrast, annualized percent
increases in prostate volume in randomly selected community men followed for 5 years were highly dependent on
baseline prostate volume (r 5 0.75, unpublished data). In the
current 4-year controlled clinical trial prostatic enlargement
on digital rectal examination and symptoms of BPH were
entry criteria, whereas distribution of prostate volumes at
baseline in the Olmsted County community based study was
broader.
The strongest relationship was identified between baseline
PSA and percent volume changes with time. Observed average percent growth from baseline ranged from 7.4% to 22%
when stratified by PSA tertile but only from 12.5% to 16.6%
by decade of life. Similar relationships held true at 36 months
but differences at 12 and 24 months were less evident. Average annualized growth rate was 0.7 ml. per year in men
with the lowest PSA tertile (0.2 to 1.3 ng./ml.), which is
similar to that seen in randomly selected community dwelling men, compared to 2.1 ml. in those with the second PSA
tertile (1.4 to 3.2) and 3.3 ml. in those with the highest
PSA tertile (3.3 to 9.4). Figure 6 suggests that lower urinary
tract symptoms have been similar between men with the
lowest and highest PSA tertiles but the underlying disease

FIG. 5. ROC curve for baseline PSA and prostate volume to predict growth defined as net increase of 5 ml. from baseline during 4
years. Areas under curve for prostate volume (0.719) and PSA (0.787)
are not significantly different (p 5 0.270).

SERUM PROSTATE SPECIFIC ANTIGEN PREDICTS PROSTATE GROWTH

19

FIG. 6. Mean percent changes in prostate volume during 4 years stratified by serum PSA tertile. Numbers below panels represent number
of patients available for followup at each time. Vertical bars represent standard error.

process in the prostate may have been considerably different.

Differences regarding histological composition (stromal versus PSA producing glandular epithelial tissue) may be suspected of being at the core of such differences.
There are multiple important implications of these findings for the clinical practice of urology. The relationship
between serum PSA, and baseline symptom severity and
peak urinary flow rate in men with BPH has been shown to
be weak. Prostate volume has been shown to be a predictor of
BPH related outcomes, such as risk of retention and surgery,15, 16 and rates of prostatic growth (unpublished data).
Due largely to its relationship with prostate volume, serum
PSA in the PLESS trial was strongly predictive of acute
urinary retention and the need for BPH related surgery16 as
well as prostatic growth. With its greater accuracy it may be
an even stronger predictor. It is recommended that men older
than 50 years have annual physical examinations, which
often include serum PSA for prostate cancer screening. Thus,
consideration should be given to the use of PSA, already
readily available and routinely obtained, to predict future
prostatic growth and risk for long-term negative BPH related
outcomes.
Despite the strong correlation between serum PSA and
prostate size, the estimate of prostate size for any given PSA
value still retains some variability. For example, if a patient
presents at age 60 years with a PSA of 1.0, a prostate volume
of 33 ml. and a symptom severity score of 10 points, watchful
waiting may be considered a reasonable alternative. Based
on PSA and prostate volume, the risk of disease progression
with the possibility of retention or need for surgery is likely
to be extremely small. However, if the same patient has a
high symptom score, one might consider treatment with
a-blockers to be a reasonable option with a high probability of
reducing symptom severity quickly. On the other hand, if a
patient presents at age 60 years with a PSA of 3.3, a prostate
volume of 45 or 50 ml. and a moderate to severe American
Urological Association symptom severity score, one might
consider treatment with finasteride for several reasons. The
5a-reductase inhibition has been shown to be most effective
to improve symptoms compared to placebo in men with larger
prostate glands.14 Additionally, prostate shrinkage is induced with finasteride, thereby reducing the risk of retention
and need for surgery.
Clinical decision making becomes more interesting when
an otherwise healthy man presents at age 65 years with a
PSA of 6.0 and a prostate of about 60 ml. on transrectal
ultrasonography but a sextant biopsy is negative for cancer.
The patient has a depressed urinary flow rate but only modest symptoms. Assuming an annualized prostatic growth rate
in patients with a PSA in the range of 5.5 ng./ml., he might
have a 55% increase in volume during the next 10 years,
resulting in a volume of approximately 90 ml. By recognizing
which patients are at risk for significant prostate growth
with time (for example those with PSA greater than 1.3

ng./ml.), one may be able to assess better which symptomatic

patients may be best treated with finasteride to arrest prostate growth, and reduce the risk of surgery and retention.16, 18 In these patients it is unknown whether a-blockers
will affect the future risk of surgery or retention. On the
other hand, in symptomatic men with low PSA who are likely
to have little prostate growth with time and are at low risk
for retention or needing surgery, finasteride may not be beneficial as the risk of complications is low, and an a-blocker
might be considered more appropriate for symptomatic improvement.24, 26
The limitations of this study are noteworthy. Of the 3,040
patients originally enrolled in the PLESS trial only 10% had
prostate volume assessed by MRI and only 164 in the placebo
arm had prostate volume assessed. Of those patients only 83
had complete data sets at each time in the study. Fortunately, however, these patients were reasonably well stratified from a low value to a PSA of 10 ng./ml., and equally well
stratified by decade of life, thus allowing conclusions to be
drawn. In addition, the average growth rate when calculated
based on an intention to treat last value carried forward
approach, including data on patients who dropped out of the
study, was similar to the analysis of those who completed
the study (data not shown). Another limitation is that patients were selected based on moderate to severe symptoms
of BPH as well as an enlarged prostate on digital rectal
examination. Therefore, these data may not be applicable to
men without symptoms or enlarged glands.
The translation of clinical research results into daily clinical
practice is often difficult to achieve. Even more difficult is translating what amounts to a paradigm shift in the assessment and
treatment of men with lower urinary tract symptoms and BPH.
While in the past prostate volume was considered unimportant
and the only goal of therapy was to improve BPH symptoms,
recent data have demonstrated that prostate volume and serum
PSA are strong predictors of adverse outcomes, such as urinary
retention and the need for surgery.15, 16 We demonstrated that
serum PSA is an even better key to understanding future prostate growth potential. By reviewing serum PSA, which is usually obtained routinely, physicians can evaluate future potential for development of detrimental BPH related outcomes.
Using PSA beyond the current use as a screening tool for prostate cancer, physicians now have the opportunity to counsel
patients regarding future risk of BPH related outcomes and to
incorporate preventive medicine into the therapeutic paradigm.
A simple PSA can provide the key information on future risk of
prostate volume growth, and risk factors of retention and surgery in individual patients, and can help therapeutic management of BPH.
CONCLUSIONS

The 164 placebo treated patients of the 4-year randomized

PLESS trial who underwent baseline and yearly MRI pros-

20

SERUM PROSTATE SPECIFIC ANTIGEN PREDICTS PROSTATE GROWTH

tate volume measurement had continual increases in volume

with time. Prostate volume is an important predictor of future development of acute urinary retention and the need for
BPH related surgery. Although prostatic growth was related
to baseline prostate volume, the strongest predictor of future
prostate growth was baseline serum PSA. Thus, serum PSA
is a useful predictor to assess future prostate growth and the
risk of future BPH related outcomes, and to help make therapeutic decisions for individuals with BPH. Such thinking
reflects a shift from focusing solely on symptoms and flow
rate changes toward a more comprehensive approach with
preventative considerations of care of patients with BPH.

13.
14.

15.
16.

REFERENCES

1. Guess, H. A., Arrighi, H. M., Metter, E. J. et al: Cumulative

prevalence of prostatism matches the autopsy prevalence of
benign prostatic hyperplasia. Prostate, 17: 241, 1990.
2. Barry, M. J., Fowler, F. J., OLeary, M. P. et al: The American
Urological Association symptom index for benign prostatic
hyperplasia. J Urol 148: 1549, 1992.
3. Barry, M. J. and OLeary, M. P.: Advances in benign prostatic
hyperplasia. The developmental and clinical utility of symptom scores. Urol Clin North Am, 22: 299, 1995.
4. Barry, M. J., Fowler, F. J., Jr., OLeary, M. P. et al: Measuring
disease-specific health status in men with benign prostatic
hyperplasia. Med Care, 33: AS145, 1995.
5. Cockett, A. T. K., Khoury, S., Aso, Y. et al: The 3rd International
Consultation on Benign Prostatic Hyperplasia (BPH). Channel
Islands: Scientific Communication International Ltd., 1996.
6. Abrams, P. H.: Prostatism and prostatectomy: the value of urine
flow rate measurement in the preoperative assessment for
operation. J Urol 117: 70, 1977.
7. Jensen, K. M., Jorgensen, J. B. and Mogensen, P.: Urodynamics
in prostatism. I. Prognostic value of uroflowmetry. Scand
J Urol Nephrol, 22: 109, 1988.
8. Jensen, K. M., Jorgensen, J. B. and Mogensen, P.: Urodynamics
in prostatism. IV. Search for prognostic patterns as evaluated
by linear discriminant analysis. Scand J Urol Nephrol, suppl,
114: 84, 1988.
9. Jensen, K. M., Jorgensen, J. B. and Mogensen, P.: Urodynamics
in prostatism. II. Prognostic value of pressure-flow study combined with stop-flow test. Scand J Urol Nephrol, suppl, 114:
72, 1988.
10. Chancellor, M. B., Blaivas, J. G., Kaplan, S. A. et al: Bladder
outlet obstruction versus impaired detrusor contractility: the
role of uroflow. J Urol, 145: 810, 1991.
11. Barry, M. J., Cockett, A. T., Holtgrewe, H. L. et al: Relationship
of symptoms of prostatism to commonly used physiological and
anatomical measures of the severity of benign prostatic hyperplasia. J Urol, 150: 351, 1993.
12. McConnell, J. D., Barry, M. J., Bruskewitz, R. C. et al: Benign
prostatic hyperplasia: diagnosis and treatment. Clinical Practice Guideline, In: No. 8. Rockville, MD: Agency for Health

17.
18.

19.

20.
21.

22.

23.

24.

25.

26.

Care Policy and Research, Public Health Service, U.S. Department of Health and Human Services, 1994.
Oesterling, J. E., Jacobsen, S. J., Chute, C. G., et al: The establishment of age specific reference ranges for prostate specific
antigen. J Urol, 149: 510A, abstract 1188, 1993.
Boyle, P., Gould, A. L. and Roehrborn, C. G.: Prostate volume
predicts outcome of treatment of benign prostatic hyperplasia
with finasteride: meta-analysis of randomized clinical trials.
Urology, 48: 398, 1996.
Jacobsen, S. J., Jacobson, D. J., Girman, C. J. et al: Natural
history of prostatism: risk factors for acute urinary retention.
J Urol, 158: 481, 1997.
Roehrborn, C. G., McConnell, J. D., Lieber, M. et al: Serum
prostate specific antigen concentration is a powerful predictor
of acute urinary retention and need for surgery in men with
clinical benign prostatic hyperplasia. Urology, 53: 473, 1999.
Bonilla, J., Stoner, E., Grino, P. et al: Intra- and interobserver
variability of MRI prostate volume measurements. Prostate,
31: 98, 1997.
McConnell, J. D., Bruskewitz, R., Walsh, P. et al: The effect of
finasteride on the risk of acute urinary retention and the need
for surgical treatment among men with benign prostatic hyperplasia. N Engl J Med, 338: 557, 1998.
Rhodes, T., Girman, C. J., Jacobsen, S. J., et al: Longitudinal
measures of prostate volume in a community-based sample:
3.5 year followup in the Olmsted County study of health status
and Urinary symptoms among men. J Urol, 153: 301A, abstract 291, 1995.
Jacobsen, S. J., Girman, C. J., Guess, H. A. et al: Do prostate size
and urinary flow rates predict health care-seeking behavior for
urinary symptoms in men? Urology, 45: 64, 1995.
Kaplan, S. A., Te, A. E., Pressler, L. B. et al: Transition zone
index (TZI) as a method of assessing benign prostatic hyperplasia: correlation with symptoms, uroflow and detrusor pressure. J Urol, 154: 1764, 1995.
Andersen, J. T., Ekman, P., Wolf, H. et al: Can finasteride
reverse the progress of benign prostatic hyperplasia? A twoyear placebo-controlled study. The Scandinavian BPH Study
Group. Urology, 46: 631, 1995.
Nickel, J. C., Fradet, Y., Boake, R. C. et al: Efficacy and safety of
finasteride therapy for benign prostatic hyperplasia: results of
a 2-year randomized controlled trial (the PROSPECT Study).
Can Med Assoc J, 155: 1251, 1996.
Roehrborn, C. G., Oesterling, J. E., Auerbach, S. et al: Effectiveness and safety of terazosin versus placebo in the treatment of
men with symptomatic benign prostatic hyperplasia in the
HYCAT study. Urology, 47: 159, 1996.
Collins, G. N., Raab, G. M., Hehir, M. et al: Observer variability
and reproducibility of transrectal ultrasound in the measurement of prostate volume. Ultrasound Med Biol 21: 1101,
1995.
Lepor, H., Williford, W. O., Barry, M. J. et al: A randomized
placebo controlled clinical trial of the safety and efficacy of
therapies in men with clinical benign prostatic hyperplasia.
N Engl J Med, 335: 533, 1996.