|
"Can Aggressive Prostatic Carcinomas be Identified and Can Their Natural
History be Altered by
Treatment" (March 2005)
This question is the title of the article by
Quinlan, Partin and Walsh, UROLOGY 46 (Suppl 3A), 1995, for which a
partial answer is gradually emerging. The extent of relevance of this
issue today is documented by the prevalence of high-grade prostate cancer
at diagnosis reported in “Prevalence of Prostate Cancer Among Men with a
Prostate-Specific Antigen Level < 4.0 ng per Milliliter” (Ian
Thompson et al, NEJM May 27, 2004). Fifteen percent of the total of 2950
biopsied men [the control arm of the finasteride study] showed cancer, and
15% of these had high-grade cancer (Gleason 4+3, 8, and 9). Based
on an estimated 232,000 new cases for 2005, and considering that many men
will be diagnosed with a PSA value of greater than 4 ng/mL, in 2005
high-grade disease will be found in more than 35,000 men!
Organ Confinement of Prostate Cancer at
Diagnosis: A Critical Factor for Optimal Treatment Outcome for High-Grade
Disease
It is well recognized that the magnitude of
“aggressiveness” of high-grade disease and the outcome of its treatment
are significantly influenced by the interplay of the PSA level at
diagnosis, Gleason score, and the clinical and pathologic tumor stage. The
Partin tables and the nomogram built on CaPSURE data, which substitutes
percentage of biopsy cores positive for clinical tumor stage, are relied
upon to predict the underlying prostate pathology that is likely
associated with the various permutations of these three predictive
factors. The Kattan, Wheeler, and Scardino article, “Postoperative
nomogram for disease recurrence after radical prostatectomy for prostate
cancer” - the title of their JCO May 1999 analysis, extends predictions
based on these variables to an estimation of a man’s freedom from PSA
recurrence at 2, 5, and 7 years. (This is available on the Web at
www.mskcc.org/mskcc/html/5794.cmf). Underlying all these calculations is
the primacy of organ confinement of prostate cancer as the major predictor
of favorable treatment outcome, and the most heavily weighted factor that
determines organ confinement is the absence of extra-capsular extension of
cancer. The pathologic description of capsular invasion may be further
refined by measurement of the extent of capsular penetration and a
determination of “specimen confinement” in the surgical specimen. The
critical importance of capsular invasion was highlighted by Wheeler,
Scardino et al in their August 1998 Human Pathology article: “Prostate
cancer does not appear to metastasize in the absence of invasion into the
capsule regardless of the volume or grade of the intracapsular tumor”.
Outcomes of Surgical Treatment of
High-Grade Prostate Cancer
Is there a documented benefit to treatment of
high-grade cancer when it is organ confined? Two representative surgical
series document improved outcome when high-grade disease is treated at
this stage. Mian et al (M.D.Anderson) analyzed this issue in “Outcome of
Patients with Gleason Score 8 or Higher Prostate Cancer Following Radical
Prostatectomy” (J UROL, April, 2002) and concluded: “Patients with
specimen confined disease had a significantly higher 5-year disease
free survival than those with non specimen confined disease (84% and 50%,
p <0.0001)” and “the pathological status of the surgical specimen was the
most significant independent predictor of disease recurrence”.
Hull, Wheeler, Scardino et al (J UROL
February 2002) - “Cancer Control with Radical Prostatectomy Alone in 1,000
Consecutive Patients” - found that “poorly differentiated cancer that is
confined to the prostate pathologically is highly likely to be controlled
with radical prostatectomy.” Additionally, “Further evidence of the
efficacy of surgery in poorly differentiated [organ confined] cancer was
the 57.6% + 21% 10 year metastases free survival rate for
high-grade cancer in this series...”
The outcome of diagnosis and surgery of organ
confined high-grade (in this case Gleason 8-10) was reported by Epstein,
Watch, Kattan et al in “Probability of Biochemical Recurrence by Analysis
of Pathologic Stage, Gleason Score, and Margin Status for Localized
Prostate Cancer, UROLOGY 62(5),2003. Men with Gleason scores 8-10 organ
confined cancer showed a bRFS at five years of 62% (range 51-70) and 41%
(range 29-55) at 10 years. The Kattan post-operative nomogram
individualizes estimates by entering the parameters as continuous
variables. For example, a man with a pretreatment PSA of 4, a specimen
Gleason score of 9, with organ confined cancer is predicted to have a 91%
probability of bRFS at 7 years.
“Early
Diagnosis” Doesn’t Diminish the Likelihood of Finding High-Grade Disease
Unfortunately, “early” diagnosis of prostate
cancer at a low PSA level and in low volume doesn’t have a commensurate
“pay off” in finding less aggressive histology. Kupelian and Klein in
their Cleveland Clinic series (CANCER. Dec. 2002) observed “that tumors of
Gleason score 8 and above are not associated with the dominant pathologic
stage migration observed with lower grade tumors during the PSA era...”.
In the data from the Prostate Cancer Prevention Trial referred to earlier,
the prevalence of cancer was 15.2% in men with normal DRE exams and PSA
levels of less than PSA 4 ng/mL. Within this group high-grade
disease was found in 16% of men in the PSA range <.5 to 3.0 ng/mL
and in 25% in the PSA range between 3.1 and 4.0 ng/mL. Although in the
past it was believed that a more aggressive histology gradually evolved
over time as a result of increasing mutations, currently there is an
emerging body of evidence that in most instances high-grade disease is
“born that way”. PSA screening and increased prostate cancer awareness has
resulted in the diagnosis of cancer at the cT1c stage in 60% to 70% of new
cases suggesting that the major gains of finding low volume prostate
cancer have been made and that clinicians will continue to confront a
significant percentage of high-grade disease at diagnosis.
The Dilemma Posed By The Decreasing
Specificity of PSA At Low
Levels For Identifying Low Volume, Organ Confined
High-Grade Prostate Cancer
The question that arises from these
considerations is what methods are available to clinicians to optimize the
usefulness of the PSA test so as to increase the likelihood of finding
organ confined disease in this relatively small, but important cohort of
men with high-grade cancer? The amount of prostate-specific antigen that
can be attributed to prostate cancer is considered to be
proportional to the volume of that cancer, although when in association
with significant amounts of BPH, initially the contribution of the
prostate cancer to the total PSA can be hidden. Nonetheless it will always
be advantageous to find prostate cancer - all grades of cancer - at the
lowest possible PSA, but the diagnosis of high-grade disease at a point of
low volume and organ confinement is especially crucial for optimal
treatment.
Drs. Stamey and McNeal in their article in the Journal of Urology, October
2004, highlighted the increasing difficulty of identifying high-grade
cancer in current PSA screening programs. Their analysis spanning 20 years
of PSA screening found a decreasing utility of the PSA
level itself in identifying low volume, high-grade
cancer. Clinicians may have to transition from focusing on a “normal” PSA
level of 4 ng/mL to other types of diagnostic information that can be
gained from PSA interpretation particularly in the range of less than 4
ng/mL. The principle candidates for facilitating a more sophisticated
interpretation of PSA values are: age-specific PSA ranges, percentage free
PSA in PSA levels below 4 ng/mL, and the rate of PSA increase over time (PSA
velocity).
Age-Specific PSA Ranges
The average PSA in men between 41-50 years is
1.2 ng/ml; for 51-60, 1.7; 61-70, 2.2; and for 71-80, 3.8. These values
are soberingly lower than the upper range of “normal” set generously by
Oesterling to include 99% of normal men (containing, of course, many men
with cancer): age 40-49, <2.5 ng/mL; 50-59, <3.5; 60-70, <4.5; and 70-80,
<6.5. By reducing the sensitivity level to only include 97.5% of normals
the values would change, for example, to <1.81 for age 40-49 and <3.36,
50-59. It is interesting that the ranges chosen in the European literature
are considerably lower. In the background for these ranges is the estimate
that “For a healthy 60-year old man with no evidence of prostate cancer,
the PSA concentration increases by approximately 3.2% per year, i.e. about
0.04 ng/mL yearly (Oesterling JE. JAMA. 1993 Aug).
In Judd Moul’s excellent article “Population
Screening for Prostate Cancer and Emerging Concepts for Young Men”
(Clinical Prostate Cancer, Sept. 2003) he reports that the median PSA for
army officers age 40 - 49 is between 0.7 and 0.8 ng/mL with only 8.2% over
1.5 ng/mL. Because of his heightened alertness to cancer Dr. Moul may be a
more aggressive diagnostician than average, but it is instructive to note
that and he would consider biopsying a man in this age group if his PSA
were consistently >1.7 ng/mL.
In men younger than 60 years the use of
age-specific PSA range to trigger a biopsy increases cancer detection by 8
- 18% and “ 81% had favorable pathological results [i.e., organ
confinement or capsular penetration with a Gleason score of < 7] (Partin &
Oesterling, J Urol. 1996 Apr.).
Percentage Free PSA In The PSA Range of 2
- 4 ng./mL.
The application of the percent free PSA in
this PSA range is controversial. However there is agreement that within
this PSA window 20% to 30% of men with negative DREs harbor prostate
cancer, and “up to 80% of those prostate cancers are clinically
significant”. Catalona found that in the PSA range 2.51 to 4 ng/mL 91%
were pathologically organ confined (Urology, 54:220,1999).
The crux of the issue is the
sensitivity/specificity trade off; but also relevant to the total issue is
the intensity of a clinician’s inclination to biopsy into this pool in
search of high-grade cancers, for which there is the greatest gain from
finding organ confined disease. If a clinician’s tendency is to biopsy
early and often, then restricting biopsies to those men with normal DRE
exams and free PSA values below 18% to 20% will avoid 73% of unnecessary
biopsies (Haese and Partin. J Urol. 2002). But if a clinician has a
conservative inclination, then he may be motivated by the information that
a prostate biopsy of a man with < 20% free PSA will yield a
positive result in 50% of cases (Haese). Haese supports his suggestion by
the comment “Assuming regular monitoring of individual patients
with PSA of 2 to 4 ng./mL., sensitivity may not need to be as high as in
the total PSA 4 to 10 ng./mL. range”; i.e. missed cancers will be found in
subsequent testing.
Roehl and
Catalona (J Urol 2002) were less sanguine about the application of % free
PSA to guide biopsy decisions in the PSA range 2.6 - 4.0 in men with
normal DRE exams. In their study of the outcome of biopsies in 965 men
they found: “A 25% free PSA cutoff detected 85% of cancers and avoided 19%
negative (cancer-free) biopsies...” and “Of those men who underwent
radical prostatectomy 132 (80%) had pathologically organ confined
disease.” However, it is not standard clinical practice to perform
biopsies in most men who have normal DRE exam and a PSA between 2.6 and
4.0 ng/mL. It could be concluded that by using Haese’s cutoff of 18% to
20%, biopsies will be done, and cancers will be discovered - ones that
would otherwise not have been found - and there will be a high probability
of organ confinement
PSA
Velocity
Probably the most informative dynamic of PSA
is its rate of change over time - the PSA velocity. It is generally
recognized that an increment of > .75 ng/mL per year signals a 72%
likelihood of underlying cancer, but what is less recognized is that this
figure is best applied in the PSA of 4 to 10 ng/mL. Because of the lower
volume of cancers in the PSA range of < 4 ng/Ml a considerably smaller
velocity can be informative.
This issue was addressed nicely by Fang,
Ballentine Carter, et al in “PSA Velocity for Assessing Prostate Cancer
Risk in Men with PSA Levels Between 2.0 and 4.0 ng/mL” (UROLOGY, June
2002). Access to sequential PSA levels in 21 men over a period of 18 or
more months preceding the diagnosis of cancer allowed them to make a
comparison with 68 controls and conclude that “The sensitivity and
specificity [for predicting cancer] of a PSAV of 0.1 ng/mL/year was 81%
and 50%, respectively”. Also, they found that “The relative risk of
prostate cancer was 6.53 (range 1.90 to 22.51) when the PSAV
was 0.1 ng/mL per year or more compared with a PSAV of less than 0.1 ng/mL
per year (P = 0.0029)”. “The median PSA velocity across three measurements
was 0.02 ng/mL for those with cancer and 0.01 ng/mL for those without
cancer...”. Their study suggests that in this low range of PSA a yearly
increase of > 20% may indicate cancer.
Riffenburgh’s
analysis, “Use of early PSA velocity to predict eventual abnormal PSA
values in men at risk for prostate cancer” (Prostate Cancer and Prostate
Diseases. 2003. June) comes to a similar conclusion: a yearly rise of 0.13
ng/mL was enough to “trigger the assignment of the patient to a high risk
group”.
D’Amico has emphasized the perspective that
can be provided by awareness of the PSA velocity over several
observations. The PSA velocity places any subsequent PSA test into its
dynamic context. This has led him to suggest that several years of PSA
testing should precede the now conventional starting point of age 50 for
the average man, so that the most informed interpretation can be made.
Bottom Line:
In the effort to find organ confined high-grade cancer - the stage
associated with optimal treatment outcome - an informed use of
age-specific PSA ranges, percent free PSA, and PSA velocity - singly or in
combination - offers the strongest leverage.
«
Back to Article List |
