2012 EDUCATIONAL BOOK - American Society of Clinical Oncology

Overdiagnosis and Overtreatment of
Prostate Cancer
Overview: Prostate cancer is a ubiquitous disease, affecting
as many as two-thirds of men in their 60s. Through widespread
prostate-specific antigen (PSA) testing, increasing rates of
prostate biopsy, and increased sampling of the prostate, a
larger fraction of low-grade, low-volume tumors have been
detected, consistent with tumors often found at autopsy.
These tumors have historically been treated in a manner
similar to that used for higher-grade tumors but, more recently,
it has become evident that with a plan of active
surveillance that reserves treatment for only those patients
whose tumors show evidence of progression, very high
disease-specific survival can be achieved. Unfortunately, the
frequency of recommendation of an active surveillance strat-
WITH THE diffusion of PSA screening that first began
in the mid-1980s and with subsequent changes in
the way we diagnose prostate cancer, the tumors of today
are remarkably different from those of the 1980s or even
1990s. At its most basic level, the opportunity for overdiagnosis
of prostate cancer (overdiagnosis can be defined as
detection of cancers that will ultimately not harm the host)
must be understood to be related to the background rate of
the disease. The best insight into this potential opportunity
for diagnosis of prostate cancer comes from autopsy studies
of men who died as a result of other causes. Careful sectioning
of prostates in these men gives us an idea as to the lower
bound estimate for the risk of the disease. (It is the lower
bound estimate because men who have been treated with
surgery or radiation are effectively censored from these
analyses.) Nonetheless, these rates are substantial, as summarized
in Table 1. 1 Given that a man’s life expectancy
approaches 80 years in the United States, it would seem the
infrequent aging man who does not harbor at least a small
tumor in his prostate that, if struck with a prostate biopsy
needle, would result in a prostate cancer diagnosis. Indeed,
to make a diagnosis of prostate cancer, which is generally
asymptomatic until metastases develop, there are three
prerequisites: 1) the physician must suspect cancer, 2) the
patient must accept a prostate biopsy, and 3) the biopsy
needle must strike the tumor. As will be seen, these rates
have changed variably over the years.
How Is the Prostate Cancer of 2012 Different from
the Prostate Cancer of Prior Decades?
Physician Suspicion of Prostate Cancer
Before PSA, the only method for a physician to suspect the
presence of prostate cancer (at least, early prostate cancer)
was through an abnormal digital rectal examination (DRE).
This was generally uncommon. In one series we conducted
prospectively in the early 1980s, in 2,005 men undergoing
biopsy, only 65 abnormal examinations were identified. 2
Of these, only 17 proved to be cancer. With the advent of
PSA testing, it was not necessarily a substantially greater
percentage of men in whom prostate cancer was suspected
(approximately 8% of the population has a PSA greater than
4.0 ng/mL) but a far greater fraction of the population was
interested in PSA testing, likely as a result of the poor
By Ian M. Thompson Jr., MD
egy in the United States is low. An alternative strategy to
improve prostate cancer detection is through selected biopsy
of those men who are at greater risk of harboring high-grade,
potentially lethal cancer. This strategy is currently possible
through the use of risk assessment tools such as the Prostate
Cancer Prevention Trial Risk Calculator (www.prostate.cancer.
risk.calculator.com) as well as others. These tools can predict
with considerable accuracy a man’s risk of low-grade and
high-grade cancer, allowing informed decision making for the
patient with a goal of detection of high-risk disease. Ultimately,
other biomarkers including PCA3, TMPRSS2:ERG, and
[-2]proPSA will likely aid in discriminating these two types of
cancer before biopsy.
acceptance of DRE by U.S. males. It was then with this
testing that we witnessed a spike in prostate cancer diagnosis
in the early 1990s, which subsequently fell back (after a
harvest of prevalent tumors) to a rate approximately double
that of the pre-PSA 1980s. Presently, approximately 75
percent of at-risk men have undergone a PSA test and
approximately 50% are tested regularly.
Further changes in suspicion of prostate cancer then
developed. These included 1) the recognition that prostate
cancer was quite prevalent at lower levels of PSA, 2) the
recognition that other factors such as family history, race,
and age could be incorporated into a risk assessment to lead
to a biopsy in a man with a PSA less than 4.0 ng/mL, and
3) the use of changes in PSA (PSA velocity) to prompt biopsy
at even lower levels of PSA. 3-5 All of these biopsy prompts
have led to dramatic increases in the use of prostate biopsy.
The sum total of these events has dramatically increased the
likelihood that a man, if a PSA is obtained, will receive a
recommendation to consider prostate biopsy.
Patient Acceptance of Prostate Biopsy
This variable can dampen the rate of prostate cancer
diagnosis because many men who receive PSA results that
are greater than 4.0 ng/mL will opt to not have a prostate
biopsy. In the Prostate, Lung, Colorectal and Ovarian
(PLCO) Cancer Screening Trial in the United States, for
example, at the baseline PSA test, if a man’s PSA exceeded
4.0 ng/mL, within 1 and 3 years, the likelihood that he
underwent prostate biopsy was 41% and 64%, respectively. 6
Almost certainly, the way the PSA data are presented by the
physician will play a part in whether the patient undergoes
prostate biopsy.
From the Cancer Therapy and Research Center, University of Texas Health Science
Center at San Antonio, San Antonio, TX.
Author’s disclosure of potential conflicts of interest are found at the end of this article.
Address reprint requests to Ian M. Thompson Jr., MD, Cancer Therapy and Research
Center, University of Texas Health Science Center at San Antonio, 7979 Wurzbach Road,
San Antonio, TX; email: thompsoni@uthscsa.edu.
© 2012 by American Society of Clinical Oncology.
1092-9118/10/1-10
e35