EURURO Vol. 72 No. 3

445 476

advantage of the present work. Finally, the presence of
detailed core biopsy information enabled us to examine the
potential predictive value of these elements. Biopsy
information was unable to predict GS upgrading accurately,
neither using a continuous nor a dichotomized model.
In a comparable study, Ploussard et al analyzed
2323 patients with localized GS 3 + 4 PCa who underwent
RP and applied various AS criteria to test their ability to
stratify the risk of unfavorable disease in this group. The
overall rate of unfavorable disease was 46%, and for patients
without any additional risk factor (PSA level <10 ng/ml,
PSAD 0.15 ng/ml per gram, T1c, two or fewer positive cores),
the same rate was 19%. Not surprisingly, the study showed
large heterogeneity in the results, strictly dependent on the
AS criteria used (Prostate Cancer Research International
Active Surveillance [PRIAS], Toronto Criteria, Royal Marsden
Hospital). However, the authors were not able to use
detailed biopsy information, such as percentage of core/
specimen surface involvement, which certainly can be
relevant when counseling a man about AS eligibility.
In the present work, we used a FIR definition and
consequently found a risk of unfavorable disease (24.8%)
slightly lower than the 30.5% reported by the authors using
the most restrictive criteria. In our study, clinical stage
higher than T1 and information on positive cores were
predictors of unfavorable disease, confirming the findings
of Ploussard et al. We were able to assess the role of age
and perineural invasion that also played a significant role in
the prediction of unfavorable disease.
Prediction of GS upgrade plays a major role when
considering AS, given the known risk of underestimation
in biopsy samples. In our series, only age, PSAD, and
percentage of surface involvement showed association with
Gleason upgrade, whereas in the Ploussard et al clinical
stage, PSA, PSAD, and total number of cores more than two
reached significance. These discrepancies are probably
reflective of a more heterogeneous population in their
study, with an overall 30% of GS upgrade (vs 13.1% in our
cohort, which was restricted by PSA and cT at the inclusion).
Interestingly, percentage of positive biopsy cores cate-
gorized as <50% or 50%, and 33% did not show a significant
value in predicting GS upgrade either in Ploussard et al or in
the present study, suggesting a limited role for the
parameter in this setting.
Motamedinia and colleagues also analyzed a low-
risk group and reported a risk classification rate of 33% at
rebiopsy, withmicrofoci of GS 4 and the presence of prostatic
intraepithelial neoplasia as important risk factors for
progression, whereas PSA, PSAD, and the total number of
positive cores did not play a significant role. Jain and
colleagues found a 30% upgrade at 1-yr follow-up biopsy
in a cohort of 592 men with low-risk or FIR disease
undergoing AS. T2, higher PSA at biopsy, and higher
percentage of involved cores on initial biopsy predicted
upgrading. However, only 9.8% of patients were GS 3 + 4, and
most of them were all nonsurgical candidates >75 yr of age,
reducing the applicability to contemporary AS candidates.
These studies were not able to differentiate GS underesti-
mation at initial biopsy and GS progression over time.
The role of age as an independent predictor of upgrading
confirms previous findings in the setting of low- and lower
IR PCa . The precise reason for this phenomenon is
incompletely understood, but it might be related to the
influence of aging on tumor biology, leading to more
aggressive differentiation.
As far as downgrading is concerned, PSAD, clinical stage,
and the presence of perineural invasion were inversely
associated with the outcome in our cohort, and biopsy core
information, although statistically significant, did not show
a clinically meaningful association. These results partially
compare with Ploussard et al , where the overall
prevalence was 8.4%, rising to 14.6% when they applied the
more restrictive PRIAS criteria. Age, prostate-specific
antigen doubling time (PSA DT), and biopsy core informa-
tion were significant predictors in their series.
Our results fit well in the current debate about extension
of AS eligibility. IR PCa has been historically considered an
indication for immediate treatment, and the Prostate
Cancer Intervention Versus Observation Trial (PIVOT)
and the SPCG-4 trials make quite a strong case for
definitive treatment of IR men, with a mortality risk
reduction of 31% in the PIVOT and 15.5% in the SPCG-4.
These results are based on older risk-defining criteria and
lack the additional information used today. It is therefore
debatable whether they apply to contemporary newly
diagnosed GS 3 + 4 patients.
Some major prospective AS experiences have enrolled a
subset of GS 3 + 4 patients. The University of Toronto
protocol, initially including patients with GS 3 + 4 (17%),
was eventually restricted to GS 6 after 4 yr, or better IR
patients (PSA 10–20 ng/ml and/or GS 3 + 4) were admitted
only if they had significant comorbidities and a life
expectancy <10 yr. GS >6 and Gleason pattern 4 were
independent predictors of deferred treatment in that
cohort, but the receipt of ISUP 2005 criteria is not certain
in all patients. A limitation of this approach is the inclusion
of patients who would not be good candidates for curative
treatment, indeed reducing the applicability of the findings
to the modern AS concept.
Yamamoto et al analyzed the risk of metastatic
progression in 980 patients in the Toronto cohort, 133 of
whom (13.6%) were GS 7 (3 + 4 or 4 + 3). Men with Gleason
pattern 4 on biopsy have an increased risk for metastatic
progression when treated with AS (hazard ratio: 2.9). PSA
DT <3 yr and more than three positive cores are additional
risk factors. However, we do not know how many of these
patients would fit the GS 3 + 4 FIR definition and what
outcome would have been observed in this specific subset.
Cooperberg et al reported on 90 patients with IR PCa
in their AS cohort, only 27 of whom had GS 3 + 4. The rates of
active treatment-free survival did not vary between low-
risk and IR patients; in spite of that, 30% of IR men were
upgraded at repeated biopsy. Among patients who eventu-
ally underwent RP, the upstage rate was 50%. The Prostate
Testing for Cancer and Treatment (ProtecT) randomized
trial , whose results are awaited in 2016, included 20%
of GS 7 men randomized to AS, RP, or radiation therapy.
These data will shed some more light on long-term
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 4 4 2 – 4 4 7 445