Recent research regarding the “watchful waiting” or as they
call it “active surveillance” has indicated that often what has been suspected
to be a rational and efficacious effort has feet of sand. In a recent piece in
Health Imaging the writer summarizes a recent study[1]:
Results revealed that men treated with RP (radical
prostatectomy) were significantly younger, had smaller prostates, lower
prostate specific antigen (PSA), fewer comorbidities, more sampled biopsy
cores, and more positive cores. The median total biopsy length was 140 mm and
the extent of cancer was six mm. Eighty-six percent of the men had a ratio of
cancer extent that was less than 15 percent. Half of the study group (2,235
patients) had adverse pathology at RP. Predictors of adverse pathology were
older age, higher PSA, PSA density greater than 0.15 ng/ml/cm 3 and palpable
disease and extent of cancer greater than four mm on biopsy. Adverse pathology
was evident in 33 to 45 percent of the men who met the study inclusion criteria
of six different AS protocols.
Namely the adverse results on a RP were often quite higher
than they had suspected. That means specifically two things. First, prostate
biopsies are notoriously weak in determining the Gleason grade. We know that
all too often they dramatically underestimate the extent and grade. This is not
surprising since they are often sextant biopsies and at best can garner some
adequate cellular samples to determine the presence of PCa. In many cases where
PCa is present, even multiple biopsies may miss it if it is somewhat confined.
Now the paper states[2]:
Many studies have demonstrated the frequent disparity
between Gleason scores reported on prostate biopsy and at radical
prostatectomy. In a recent review Epstein et al reported that about a third of
cases with a biopsy Gleason score of 5-6 were upgraded at RP.
Thus as they note, and as we have commented, even with a
good set of core samples, PCa may be identified but its grade is often
underestimated. They continue:
This issue is particularly germane to men with presumed
low risk prostate cancer considering active surveillance, for whom accurate
pretreatment risk stratification is paramount. As reviewed by Dall'Era et al,
in most AS programs candidates are chosen based on GS (Gleason Score), clinical
stage and PSA based parameters. Depending on the particular inclusion criteria
4% to 82% of men were eligible for AS, and conversion to active treatment was
reported in 11% to 33% of men on AS, with changes in tumor histology as the most
common reason for discontinuing AS.
Namely the methods to select for watchful waiting are
subject to substantial error and underestimation. Note that there has been no consideration
in this study of the many available genomic tests currently being developed. It
would have been useful if they had also pursued a parallel study using a
selection of the genomic studies we have discussed here. They continue:
A recent consensus conference concluded that AS is
underused. However, the limitations of current clinical staging and disparities
in selection criteria among current protocols are important in evaluating
obstacles for expanding the use of AS.
The claim of underuse is essentially a claim for excess
costs. Many Government controlled Health Care plans would like to shift the
costs of surgery to the patient’s mortality and morbidity. In the US we see
this coming with the development of CCE, Comparative Clinical Effectiveness[3]. They
continue:
These results have important implications for the
optimization of AS in the future. Most AS programs consider PSA and clinical
stage for patient selection, and we confirmed that there is a greater risk of
misclassification in men with a higher PSA and/or palpable disease. Despite
recent controversy over the usefulness of digital rectal examination in PCa
screening, these results suggest that it continues to have a role in staging.
Meanwhile we identified 2 other predictors of adverse
pathology that are not currently used by many AS programs and should be
considered in the future.
One predictor was PSA density, which was independently
associated with adverse pathology in all subsets evaluated, yet is currently
only used by a minority of AS programs. Our results concur with those of
Sfoungaristos and Perimenis, who reported that PSAD was a stronger predictor of
adverse pathology than PSA or biopsy Gleason score. Although Sfoungaristos and
Perimenis suggested a PSAD cutoff of 0.20 ng/ml/cm3, the threshold currently
used by the PRIAS protocol, we found that men with PSAD greater than 0.15
ng/ml/cm3 had a significantly greater risk of adverse pathology. The Johns
Hopkins AS program has shown that PSAD is an important predictor of progression
and uses a threshold of 0.15 ng/ml/cm3 for inclusion. In our population 21% of
men had a PSAD between 0.15 and 0.20 ng/ml/cm3, demonstrating that this
distinction affects a considerable proportion of men with otherwise low risk
disease. Overall PSAD is useful as it takes prostate volume into consideration
when evaluating PSA, as we also showed that lower prostate volume is associated
with a greater risk of adverse pathology.
We have argued for PSAD extensively in our studies. It is a
logical measure and normalizes for prostate volume. However we have also argued
for temporal analyses of PSAD, % PSA, PSA, and PSA Velocity. We have
demonstrated that use of these temporal studies can be exceedingly helpful in
even ascertaining between PCa and BPH for example. Moreover, no matter what we
look at, genomics studies most likely will be the sine qua non for ascertaining
results from biopsy studies.
The authors continue:
Older age was also positively associated with adverse
pathology as previously reported. Although older men have traditionally been
targeted for conservative management, they are more likely to have disease
under staged, but they also have higher mortality from competing causes.
Overall there is clearly still room for improvement in
PCa staging, including the biopsy protocol itself. Although 10 to 14 cores are
often sampled on initial biopsy, sampling error remains problematic. Ruijter et
al found an exact correlation between biopsy and prostatectomy grading in only
43% of cases, and less than 1 point difference in 77%. For men undertaking AS
these issues become more critical, suggesting a role for saturation biopsies or
early repeat biopsy to reduce misclassification. PSA density and cancer extent
on biopsy should be considered as inclusion criteria for active surveillance.
The core issue is always a debatable issue. In larger volume
prostates it seems logical that more cores are required. A baseline of 20 or
more cores will yield substantially better results. Yet, no matter how many
reasonable number of cores, we always face the risk of not detecting specific
lesions.
Thus this study highlights the weaknesses of watchful
waiting. The concern is that entities like PCORI which was established under
the ACA will focus their CCE efforts on the perceived patient morbidity and failed
to understand mortality. PCORI us dramatically unlike the FDA, where the FDA is
Government controlled and reportable to Congress, and follows well established
clinical trial protocols of substantial scientific validity. PCORI was set up
to be extra any Congressional oversight as a non-governmental entity and thus
can dictate whatever its management so desires with no recourse to the voters
and the patients.
References
Vellekoop, A., et al, Population Based Study of Predictors
of Adverse Pathology among Candidates for Active Surveillance with Gleason 6
Prostate Cancer, The Journal of Urology, Volume 191, Issue 2 , Pages 350-357,
February 2014
[1] http://www.healthimaging.com/topics/oncology-imaging/prostate-cancer-active-surveillance-criteria-needs-expansion
[3] See
Saver, R., Health Care Reform's Wild Card: The Uncertain Effectiveness of Comparative
Effectiveness Research, 159 U. PA. L. REV. 2147 (2011).
Posts
