0022-5347/01/1653-0757/0

THE JOURNAL OF UROLOGY® Vol. 165, 757–760, March 2001

Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.® Printed in U.S.A.

Original Articles

THE INCIDENCE OF PROSTATE CANCER IN A SCREENING

POPULATION WITH A SERUM PROSTATE SPECIFIC ANTIGEN
BETWEEN 2.5 AND 4.0 NG./ML.: RELATION TO BIOPSY STRATEGY
RICHARD J. BABAIAN,* DENNIS A. JOHNSTON, WILLIAM NACCARATO, ALBERTO AYALA,
VIJAYA A. BHADKAMKAR AND HERBERT A. FRITSCHE, JR.†
From the Departments of Urology, Biomathematics, Pathology and Research Laboratory Medicine, The University of Texas, M. D.
Anderson Cancer Center, Houston, Texas, and Dade Behring, Inc., Newark, Delaware

ABSTRACT

Purpose: It has recently been suggested that the diagnostic threshold for the prostate specific
antigen (PSA) assay be lowered to enhance prostate cancer detection. A 22% incidence of prostate
cancer has been reported in men with PSA between 2.5 and 4.0 ng./ml. We designed a study to
confirm this observation.
Materials and Methods: Men who participated in our free early detection program and who had
serum PSA between 2.5 and 4.0 ng./ml. were asked to undergo prostate biopsy. Of 268 eligible
men 151 (56%) agreed to participate in this free trial. All men underwent biopsy using an 11-core
multisite directed biopsy scheme. All biopsy cores were color coded for location specificity and
examined by 1 pathologist.
Results: Cancer was identified in 24.5% (37 of 151) of the men biopsied. The median age of men
with cancer was 62 years (range 43 to 74). Conventional systematic sextant biopsies, which
accounted for 6 of the 11 cores, detected 73.0% (27 of 37) of the cancers and the alternate site
biopsies identified the remaining 10. Gleason score was 6 in 25 men, 3 ⫹ 4 in 5, 4 ⫹ 3 in 4 and
8 or greater in 3 (median Gleason score 6). There were 14 men who had 1 core positive for cancer,
9 had 2 and 14 had more than 2 (median number of positive cores 2). Of the 14 men with 1
positive core 11 had a less than 3 mm. focus of cancer and 8 had only a positive alternate site
biopsy. There were 11 cases of abnormal results on digital rectal examination, 5 of which were
cancer, and 31 cases of abnormal results on ultrasonography, 13 of which were cancer. Median
biological variability in PSA was ⫾15% (range 0.4% to 440.0%).
Conclusions: We found a significant incidence of cancer (24.5%, 37 of 51) in men with serum
PSA between 2.5 and 4.0 ng./ml. In our study 67.6% of the detected cancers were significant
based on the biopsy data. If the PSA threshold is lowered the conventional systematic sextant
technique may be preferable to an extended strategy.
KEY WORDS: prostate-specific antigen, mass screening, incidence, biopsy

The prostate specific antigen (PSA) assay is widely used
throughout the world to detect prostate cancer. The estab-
lished absolute cutoff value commonly used to differentiate
the risk between cancer and no cancer is 4.0 ng./ml.1 The
limitations of this marker, specifically its false-negative and
false-positive rates, are well known. Numerous investigators
have attempted to enhance the sensitivity and improve the
specificity of the PSA assay through various enhancements,
including volume and age based indexes.2– 4 These modifica-
tions have added little, if any, to our diagnostic armamen-
tarium.4 Some investigators have suggested lowering the
diagnostic threshold to increase the sensitivity of PSA in
Accepted for publication September 15, 2000.
* Financial interest and/or other relationship with Astra Zeneca
and Bayer.
† Financial interest and/or other relationship with IMI, Inc.,
Tosoh, dia Dexus and Bayer.
Editor’s Note: This article is the first of 5 published in this
issue for which category 1 CME credits can be earned. In-
structions for obtaining credits are given with the questions
on pages 942 and 943.
757
detecting cancer.5, 6 However, a lower threshold would signif-
icantly increase the number of men who would become can-
didates for biopsy.
Approximately 12% of all men have PSA in the range of 2.5
to 4.0 ng./ml., and presumably the vast majority do not have
this disease and even fewer would have clinically significant
prostate cancer.7 To our knowledge the relationship between
curable prostate cancer and serum PSA has not yet been
clearly defined. About 30% of men with PSA 4.1 to 10 ng./ml.
have extraprostatic extension of cancer at prostatectomy,
and this finding is associated with a less favorable outcome
compared with those with organ confined disease.8 Will low-
ering the PSA threshold significantly improve outcome with-
out overtreatment of some patients compared to the outcome
achieved using the conventional PSA 4.0 ng./ml. threshold?
The task of investigators today is to enhance detection,
thereby improving outcome, minimizing unnecessary biop-
sies and identifying those specific patients who would benefit
from therapeutic intervention.
As a result of the challenge to lower the 4.0 ng./ml. PSA
758 SERUM PROSTATE SPECIFIC ANTIGEN VALUE AND ITS RELATION TO BIOPSY STRATEGY

threshold, we designed a study to determine the incidence of

prostate cancer in men with PSA 2.5 to 4.0 ng./ml. who partic-
ipate in early detection programs. Furthermore, because of the
emergence of new biopsy strategies to enhance cancer detection,
we compared the systematic sextant biopsy and an 11-core
multisite directed biopsy in this patient population.

METHODS

Between September 1998 and September 1999, 3,172 men

participated in a free early detection program at The Univer-
sity of Texas, M. D. Anderson Cancer Center. Total serum
PSA was obtained before the digital rectal examination using
the TOSOH‡ immunometric assay. Serum PSA in the range
Schema of prostate depicts 11 sites of biopsy strategy. R Base,
of 2.5 to 4.0 ng./ml. was detected in 322 (10.2%) men. Of these right base (sextant). R Mid, right mid (sextant). R Apex, right apex
men 268 were eligible to participate in the study based on the (sextant). L Base, left base (sextant). L Mid, left mid (sextant). L
eligibility criteria of age between 40 and 75 years, no prior Apex, left apex (sextant). RAH, right anterior horn. LAH, left ante-
prostate biopsy, screening PSA 2.5 to 4.0 ng./ml. regardless of rior horn. RTZ, right transition zone. LTZ, left transition zone. ML,
midline.
bladder outlet obstructive symptoms, life expectancy 5 years
or greater, no history of bleeding dyscrasia, no anticoagulant
therapy, no medication or food supplement that could de-
zone. A sextant biopsy site only was positive in 27% (10 of 37)
crease or increase serum PSA, except saw palmetto, and no
of men with cancer. Positive biopsy results in sextant and
history of transurethral resection of the prostate or open
alternate sites were observed in 46% (17 of 37) of patients. A
prostatectomy. These 268 men were asked to undergo a pros-
repeat biopsy was recommended in 10 men for either intra-
tate biopsy at no charge.
prostatic epithelial neoplasia (5) or atypical suspicious
A total of 151 (56%) men agreed to participate in this
glands (5). The repeat biopsy was positive for cancer in 2 of 4
institutional review board approved study and signed an
men.
informed consent. All 151 men underwent an 11-core multi-
The distribution of a positive biopsy site in all 37 patients
site directed biopsy (see figure).9 Any hypoechoic area or
and in those 12 with clinically insignificant cancer based on
abnormal digital rectal examination finding that was not
the 2 biopsy strategies is shown in the table. Only 1 core
biopsied with the 11-core technique was also biopsied. Me-
positive for cancer was seen in 14 men, 8 of whom had a
dian time from initial screening PSA to biopsy was 60 days
positive alternate site. Gleason scores were 6 in 12 patients,
(range 2 to 213). No interval PSA values were obtained be-
7 (4 ⫹ 3) in 1 and 8 in 1. The length of cancer in these biopsies
tween the initial PSA and time of biopsy. All biopsy cores
was 3 mm. or less in all 12 patients with a Gleason score of 6.
were color coded with dye and submitted in 3 separate con-
Of 9 cases with 2 cores positive for cancer Gleason score was
tainers, including 1 marked right, 1 left and 1 alternate sites,
6 in 6, 7 (4 ⫹ 3) in 2 and 8 in 1. Of 14 cases with 3 or more
to reduce cost and enable accurate identification of the origin
cores positive for cancer, including 1 with 7 positive cores,
location of each core. All biopsies were reviewed by 1 pathol-
Gleason score was 7 in 5 (3 ⫹ 4 in 4 and 4 ⫹ 3 in 1) and 8 in
ogist.
1. The designation of clinically significant disease was based
All patients underwent a digital rectal examination and
on the definition proposed by Terris et al, that is more than
transrectal ultrasonography of the prostate immediately be-
1 positive core, Gleason score 7 or greater or a cancer focus in
fore biopsy. The prostate gland volume was determined by
1 core greater than 3 mm.11 All 37 cases could have been
the elliptical method. No patient was excluded from study
evaluated by this definition and, using these criteria, 67.6%
because of an abnormal digital rectal examination or trans-
(25 of 37) could have been classified as having clinically
rectal ultrasonography. Demographic information, including
significant disease. Clinically significant cancer was detected
patient age, ethnicity, PSA history, family history of prostate
in 1 patient by only a sextant biopsy compared to 2 using only
cancer and use of any medication or food supplement that
an alternate site biopsy strategy. A positive biopsy at sextant
could influence the PSA value, such as finasteride, saw pal-
and alternate sites was observed in 22 patients.
metto and any androgen or estrogen, dehydroepiandros-
To date, 19 men have undergone radical prostatectomy,
terone, herbs and vitamins, was recorded. Tumor volume in
including 14 at our institution. The tumor volume was 0.5 cc
the radical prostatectomy specimens was determined by a
or less in 8 patients (mean 0.18, range 0.03 to 0.36) and
previously described computer assisted volumetric pro-
greater than 0.5 cc in 6 (mean 1.27, range 0.61 to 3.29). The
gram.10 We statistically analyzed the relationship between
pathological stage was pT2 in 12 patients and pT3 in 2. A
variables using the chi-square statistic.
positive margin of resection was questionably focal in the
patient who had the largest tumor, which was present in
RESULTS
extraprostatic tissue. Of the men with tumor volume 0.5 cc or
Overall, cancer was detected in 24.5% (37 of 151) of the less Gleason score was 6 in 4 and 7 in 4. Of the 6 patients
men biopsied. The median age of all 151 men was 62 years with tumor volume greater than 0.5 cc Gleason score was 6 in
(range 43 to 74). There was no difference in the median age 2, 7 in 3 and 9 in 1. Radiation therapy was selected by 6
of the men with positive or negative biopsy (64 versus 62 patients, watchful waiting by 3, 6 patients were undecided as
years). The ethnic distribution of the men in our study was to which therapeutic option to choose and 1 was lost to
75% (114 of 151) white, 17% (26 of 151) black and 7% (11 of followup.
151) Hispanic. The cancer incidence was 26% (30 of 114) Mean total prostate gland volume as determined by ultra-
among white men, 23% (6 of 26) among black men and 9% (1 sound showed a significant difference (p ⫽ 0.016) between
of 11) among Hispanics. Conventional systematic sextant patients with cancer (37.6 cc) and those who had a negative
biopsies, which accounted for 6 of the 11 cores in our biopsy biopsy (45.7 cc). Mean gland volume in patients with only a
scheme, detected 73.0% (27 of 37) of the cancers. Cancer was positive alternate site was 39.2 cc compared to 31.1 cc and
detected only in an alternate site in 27% of cases, including 8 40.5 cc in those who had a positive sextant site biopsy or
in the anterior horn right and/or left and 2 in the transition positive alternate and sextant sites. These differences were
not significantly different.
‡ TOSOH Medics, San Francisco, California. Results of the digital rectal examination were abnormal
 SERUM PROSTATE SPECIFIC ANTIGEN VALUE AND ITS RELATION TO BIOPSY STRATEGY
Location of positive biopsy cores in patients with cancer and those with clinically insignificant cancer
Biopsy Strategy
Anterior Horn
No. pos. biopsy cores Ca:
Alternate only 9 2
Sextant only – –
Both 19
No. pos. biopsy cores clinically
insignificant Ca:
Alternate only 5 2
Sextant only – –
and suspicious for cancer in 11 (7.3%) men, including 6 (5.5%)
who had negative and 5 (13.5%) who had positive biopsy
results. Results of transrectal ultrasonography were abnor-
mal in 31 (20.5%) men, including 18 (15.8%) with negative
and 13 (35.1%) with positive biopsy results. Of the 136 men
evaluable for a family history of prostate cancer 23 (22%)
with negative biopsy results had a family history of prostate
cancer compared with 12 (34.3%) who had positive biopsy
results. This difference was not significant (p ⫽ 0.1863). A
total of 20 patients reported taking saw palmetto, including 4
(10.8%) with cancer and 16 (14%) with a negative biopsy
result.
A history of PSA testing was available for only 46 men
because of inability to recall prior PSA values. Of the 39 men
who could recall PSA history and who had negative biopsy
results 36 consistently had values greater than or equal to
2.0 ng./ml. but less than 4.0 ng./ml., and 3 had levels consis-
tently less than or equal to 2.0 ng./ml. Of the 7 men with
positive biopsy results 1 had PSA 4.8 ng./ml. and 6 had PSA
2.0 to 4.0 ng./ml.
DISCUSSION
There are relatively few studies that document the inci-
dence of prostate cancer when the PSA assay value is 2.5 to
4.0 ng./ml. In 1996 Shroder et al reported a 14% incidence of
PSA in the reflex range of 1 to 3.9 ng./ml., and 28% of these
men had cancer.12 In a larger study from Washington Uni-
versity biopsy in 36% of men in a screening population with
PSA 2.6 to 4.0 ng./ml. revealed cancer in 22%. As is readily
apparent, these findings are similar to those found in men
with PSA between 4.1 and 10 ng./ml.13 Our finding of a 24.5%
positive biopsy rate is in agreement with these previous
reports.
The significance of our study is that approximately 56% of
all eligible patients underwent biopsy, reducing the effect of
bias on the result. There were no significant differences be-
tween men with negative and those with positive biopsy
results in regard to abnormal findings on digital rectal ex-
amination, or in the proportions of those taking saw pal-
metto. Furthermore, no man in our study had ever taken
finasteride. As expected, a higher proportion of men with
positive than those with negative biopsy results had a family
history of prostate cancer, although the difference was not
statistically significant. The only preliminary conclusion that
can be drawn from the PSA history data is that men have a
poor recollection of prior PSA results and that most evaluable
men did not have any prior values that were significantly
different from those at entry in this study.
The only rationale to detect cancer early is to improve the
survival rate and decrease disease related morbidity. Cata-
lona et al reported that 83% of patients with PSA between 2.6
and 4.0 ng./ml. who underwent prostatectomy had significant
disease defined as greater than 0.5 cm.3 tumor volume.5 On
the basis of an earlier study we showed that 32% of cancers
with a tumor volume between 0.5 and 1.5 cm.3 had extra-
prostatic extension, which has been reported to increase bio-
chemical failure rates.10 Whether lowering the PSA cutoff
759
Pos. Biopsy Site
Transition Zone Apex Mid Base
– – –
7 5 5
5 11 8 6
– – –
2 1 2
value to 2.5 ng./ml. will increase the disease specific survival
rate remains to be determined. However, it has been amply
demonstrated that treatment outcome for surgery and radi-
ation therapy is associated with preoperative PSA, with pa-
tients with PSA less than 4 ng./ml. fairing better than those
with PSA 4.1 to 10 ng./ml.14, 15
Comparison of the conventional systematic sextant biopsy
strategy and an 11-core multisite directed scheme revealed
cancer detection rates of 17.9% and 24.5%, respectively. An
alternate biopsy site only was positive in 10 patients. Clini-
cally insignificant disease based on the previously stated
criteria was detected in 12 patients, 7 (58.3%) of whom had
only a positive alternate site. In addition, only 8% (2 of 25) of
the significant cancers would have gone undetected if only
the conventional systematic sextant biopsies were per-
formed. These data suggest that extended biopsy strategies
could be avoided in men with PSA between 2.5 and 4.0
ng./ml. in an attempt to minimize the diagnosis of insignifi-
cant cancers without jeopardizing the detection of significant
cancer.
CONCLUSIONS
A significant number (24.5%) of men with serum PSA in
the range of 2.5 to 4.0 ng./ml. have prostate cancer and based
on current clinical criteria the majority (70%) of these can-
cers are clinically significant. Furthermore, if a decision is
made to perform a biopsy when serum PSA is between 2.5
and 4.0 ng./ml., the conventional systematic sextant tech-
nique may be preferable to an extended schema to decrease
the detection of clinically insignificant cancers. A more de-
finitive answer regarding the optimum biopsy strategy in
these patients awaits larger multi-institutional studies.
REFERENCES
1. Benson, M. C., Whang, I. S., Pantuck, A. et al: Prostate specific
antigen density: a means of distinguishing benign prostatic
hypertrophy and prostate cancer. J Urol, 147: 815, 1992
2. Babaian, R. J. and Camps, J. L.: The role of prostate-specific
antigen as part of the diagnostic triad and as a guide when to
perform a biopsy. Cancer, 68: 2060, 1991
3. Oesterling, J. E., Jacobsen, S. J., Chute, C. G. et al: Serum
prostate-specific antigen in a community-based population of
healthy men. Establishment of age-specific reference ranges.
JAMA, 270: 860, 1993
4. Babaian, R. J., Kojima, M., Ramirez, E. I. et al: Comparative
analysis of prostate specific antigen and its indexes in the
detection of prostate cancer. J Urol, 156: 432, 1996
5. Catalona, W. J., Smith, D. S. and Ornstein, D. K.: Prostate cancer
detection in men with serum PSA concentrations of 2.6 to 4.0
ng/mL and benign prostate examination. Enhancement of spec-
ificity with free PSA measurements. JAMA, 277: 1452, 1997
6. Schroder, F. H., van der Cruijsen-Koeter, I., de Koning, H. J. et
al: Prostate cancer detection at low prostate specific antigen.
J Urol, 163: 806, 2000
7. Kane, R. A., Littrup, P. J., Babaian, R. et al: Prostate-specific
antigen levels in 1695 men without evidence of prostate can-
cer. Findings of the American Cancer Society National Pros-
tate Cancer Detection Project. Cancer, 69: 1201, 1992
760 SERUM PROSTATE SPECIFIC ANTIGEN VALUE AND ITS RELATION TO BIOPSY STRATEGY
8. Walsh, P. C., Partin, A. W. and Epstein, J. I.: Cancer control and
quality of life following anatomical radical retropubic prosta-
tectomy: results at 10 years. J Urol, 152: 1831, 1994
9. Babaian, R. J., Toi, A., Kamoi, K. et al: A comparative analysis of
sextant and extended 11-core multisite directed biopsy strat-
egy. J Urol, 163: 152, 2000
10. Babaian, R. J., Troncoso, P., Steelhammer, L. C. et al: Tumor
volume and prostate specific antigen: implications for early
detection and defining a window of curability. J Urol, 154:
1808, 1995
11. Terris, M. K., McNeal, J. E. and Stamey, T. A.: Detection of clini-
cally significant prostate cancer by transrectal ultrasound-
guided systematic biopsies. J Urol, 148: 829, 1992
12. Shroder, F. H., Damhuis, R. A., Kirkels, W. J. et al: European
randomized study of screening for prostate cancer—the Rot-
terdam pilot studies. Int J Cancer, 65: 145, 1996
13. Catalona, W. J., Smith, D. S., Ratliff, T. L. et al: Measurement of
prostate-specific antigen in serum as a screening test for pros-
tate cancer. N Engl J Med, 324: 1156, 1991
14. Zagars, G. K. and Pollack, A.: Radiation therapy for T1 and T2
prostate cancer: prostate-specific antigen and disease out-
come. Urology, 45: 476, 1995
15. Partin, A. W., Pound, C. R., Clemens, J. Q. et al: Serum PSA
after anatomic radical prostatectomy. The Johns Hopkins
experience after 10 years. Urol Clin North Am, 20: 713,
1993