EURURO Vol. 72 No. 5

697 860

RT in the salvage setting. In particular, the identification of
patients with distant metastases at the time of sRT would
improve our ability to deliver RT only to individuals who
would benefit from local disease control. Standard imaging
techniques such as computed tomography and bone scan
are limited by their poor sensitivity . Choline and
prostate-specific membrane antigen (PSMA) positron emis-
sion tomography/computed tomography scans play an
increasingly important role for assessing recurrence after
primary treatment. However, these imaging modalities
have poor performance characteristics at very low PSA
levels . Therefore, clinicians should rely on models
based on pathologic parameters, pre-RT PSA, and PSA
doubling time to identify patients who would benefit from
sRT . Tendulkar et al recently updated their
nomogram in a cohort of 2460 patients treated with sRT. At
a median follow-up of 5 yr, pre-RT PSA, Gleason score,
surgical margins, pathologic stage, use of ADT, and RT dose
were associated with the risk of recurrence after sRT and
were included in a predictive model that depicted a
concordance index of 0.69 at internal validation.
Besides pathologic variables, the major predictor of
response to sRT is PSA at the time of RT. Of note, pre-RT
PSA represents a parameter modifiable by the treating
physician who can decide the optimal timing for RT delivery
according to patient characteristics. When considering
exclusively patients who received early sRT (ie, at PSA
levels < 0.5 ng/ml), the 5-yr BCR- and metastases-free
survival rates increased to 63–80% and 85–90%, respectively
. A systematic review including more than
5500 patients treated with sRT observed a 2.6% loss of
BCR-free survival for each incremental of 0.1 ng/ml in the
PSA level at the time of sRT Moreover, a recent tumor
control probability model showed that the detrimental
effect of increased PSA levels at the time of RT can never be
compensated by increasing the dose Consequently,
international guidelines recommend initiating sRT at the
first sign of BCR . However, which is the best timing for
sRT administration is still debated. Indeed, while the vast
majority of studies report improved cancer control when sRT
is administered at low PSA levels in a small
proportion of patients a slowly growing postoperative PSA
might be the expression of residual benign prostatic tissue
. A multi-institutional study evaluating more than
700 node-negative patients reported that early sRT con-
ferred better cancer control when administered at the first
sign of PSA relapse especially in men with more aggressive
features . Conversely, selected individuals with favor-
able pathologic characteristics might be affected by an
indolent recurrence (rise in PSA levels that would not
increase the risk of metastases) and, therefore, would not
benefit from early sRT. A predictive nomogram to identify
patients who would benefit from sRT was developed in
472 patients with node-negative PCa treated with early sRT
. Pathologic stage, Gleason score, surgical margins, and
pre-RT PSA represented independent predictors of treatment
failure and were included in the model, which demonstrated
a discrimination of 0.74 at internal validation. Of note, PSA
doubling time cannot be accurately estimated in men
receiving sRT at low PSA levels and, therefore, has not been
included in predictive models developed on contemporary
patients
Genomic classifiers might play a role also in the salvage
setting Freedland et al showed that at a median
follow-up of 5.7 yr an increase in the genomic classifier score
corresponded to an augmented risk of metastases after sRT.
Moreover, the inclusion of the genomic classifier score in a
clinical model substantially improved its predictive accuracy
from 0.63 to 0.85. As discussed above, individuals with low
genomic classifier scores may or may not be able to postpone
RT safely. However, molecular markers could identify
patients at increased risk of metastases who should be
referred for systemic treatments (eg, ADT).
3.7. Adjuvant versus salvage RT
describes the characteristics of seven retrospective
investigations comparing aRT and sRT Six
studies reported superior oncologic outcomes with aRT
compared to sRT At 10-yr follow-up, a reduction in
rates of BCR and metastases of 32% and 17% was observed in
men undergoing aRT compared with sRT . Of note, these
investigations did not include men with aggressive disease
managed with observation who never recurred in the
salvage arm . Therefore, they compared the efficacy of
RT between patients theoretically at risk of recurrence (ie,
aRT group) versus individuals who already experienced
recurrence (ie, sRT group). Moreover, four studies included
men who received sRT at relatively high PSA levels
. This might impact on the efficacy of RT in the salvage
setting. A recent investigation reported the results of the
comparison between aRT and observation sRT adminis-
tered at PSA levels 0.5 ng/ml in a cohort of more than
500 patients with pT3N0 PCa At 8-yr follow-up the
authors did not observe differences in the metastases-free
survival and overall survival rates between patients treated
with aRT versus observation early sRT. These findings are
reassuring regarding the oncologic safety of observation sRT
at low PSA levels in men with aggressive features at RP at
intermediate-term follow-up.
Three prospective multi-center open-label trials are
currently randomizing patients with adverse pathologic
characteristics to aRT versus observation sRT (RADICALS,
RAVES, and GETUG-17) and their results are badly needed to
inform clinicians regarding the role of sRT as compared with
aRT in this setting ). The RADICALS trial is a phase
3 randomized-controlled study taking place in the UK, Canada,
Denmark, and Ireland that is divided in two parts: RT timing
comparison (RADICALS RT) and hormone duration comparison
(RADICALS HT). In the RADICALS RT, patients are randomized to
early versus deferred RT, which is administered at the time of
PSA failure after surgery. The RADICALS HT trial is randomizing
patients to RT alone (early or deferred) versus RT + 6 mo ADT
versus RT + 12 months ADT. The primary outcome of both
studies is represented by disease-specific mortality. At the
latest update, more than 1300 patients were randomized to
early versus deferred RT in the RADICALS RT trial . More-
over, more than 2500 patients are expected to be recruited in
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 6 8 9 – 7 0 9 697