EURURO Vol. 72 No. 5

725 860

the potential to save costs for payers when accounting for
care over a longer postoperative period and when optimal
cancer and quality-of-life outcomes with RARP are achieved.
3.1.3. RARP versus RRP from the societal perspective
Four studies addressed cost for RARP from the societal
perspective ). These were published more recently
than the other studies (2007–2016). Two studies were from
the USA, and one each from Australia and Denmark. The
Australian study found a cost per QALY gained of
approximately A$24 000 with RARP. However, the authors
used very optimistic outcome data from only one study and
did not perform sensitivity analyses . The study from
Denmark calculated the ICER for one successful surgery
comparing RARP with RRP, rather than calculating the ICER
per QALY gained—a somewhat unusual approach. Success-
ful surgery was defined as achieving cancer control without
incontinence or erectile dysfunction. They estimated an
ICER per extra successful surgery of s78 000 . This
study was limited by the significant amount of uncertainty
surrounding the assumptions on postoperative outcomes
. One study found a shortened sick leave after RARP
compared with RRP, but did not estimate the economic
impact of that difference . Finally, a comprehensive
study found that RARP saved approximately $1200 per case
due to faster recovery and fewer lost wages . This study
used a systematic review of the literature to estimate model
inputs and constructed detailed care pathways and cost
models, but there is a high risk of conflict of interest because
it was conducted and funded by Intuitive Surgical, the
manufacturer of the surgical robot . Nevertheless, it is
reasonable to conclude that RARP at least has the potential
for a moderate cost advantage from the societal perspective
if optimal outcomes are achieved, allowing patients to
return to work more quickly.
3.1.4. Quality of the evidence
The studies comparing the cost of RARP and RRP were all
observational in nature. There were some inconsistencies
across studies, but those could largely be explained by
differences in study designs. The risk of bias was mostly
deemed moderate or high. Based on this, we feel that the
overall quality of the evidence is low, indicating that the true
cost difference between RARP and RRP may be substantially
different from that reported in the reviewed studies
3.2. IMRT versus 3D-CRT
3.2.1. IMRT versus 3D-CRT from the payer’s perspective
We identified nine studies comparing the cost of IMRT and
3D-CRT from the payer’s perspective, while there were no
studies from the hospital or societal perspective ).
The lack of studies from the hospital perspective may be due
to the fact that radiotherapy is usually provided on an
outpatient basis, and capturing the related equipment,
overhead, and personnel costs may be more complex than
in the setting of inpatient surgery with RARP. The studies
spanned the years 2005–2016 and originated in five
different countries, including the USA (n = 4), Canada
(n = 2), Australia (n = 1), the UK (n = 1), and Hungary
(n = 1). Seven of the nine studies found an increased cost
for IMRT, ranging from $381 to $26 066 more per case. The
wide range of cost differential was affected by the time
horizon of the cost analyses (eg, incorporating costs in the
12 mo since diagnosis vs the lifetime of the patient after
treatment ), health system of the country in which the
study took place, whether or not the start-up phase of IMRT
was incorporated, and the specific costs being measured (eg,
Medicare reimbursements only vs reimbursement costs
along with the costs of equipment, supplies, personnel, and
overhead ). The study with the lowest cost difference
($381) specifically assessed a mature IMRT program
. When it examined a scenario in which IMRT was in
a start-up phase, the incremental cost of IMRT increased by a
factor of 11 to $4268 . Two studies found that IMRT costs
less than 3D-CRT. Their findings were likely driven by a long
time horizon (10 and 20 yr ), lack of inclusion of
capital costs and decreased estimated need for salvage
hormone therapy and chemotherapy These studies
originated in Australia and Hungary, and may not be
generalizable to the patient populations in other countries.
Six of the nine studies performed cost-utility analyses in
which they reported an ICER. The ICERs ranged from$16 182/
QALY to $41 572/QALY when comparing IMRT with 3D-CRT.
This implies that even though IMRT, in many cases, was
costlier than 3D-CRT, it was also more effective and—to this
end—was generally thought to be a more cost-effective
treatment. The caveat is that an ICER is dependent on the
assumptions about the toxicity and cancer control differ-
ences between the two treatments, which varied widely
between studies. However, these studies all performed a
variety of sensitivity analyses to test the robustness of their
findings across a range of parameters. In summary, the
majority of studies found IMRT to be more expensive from a
payer’s perspective, but also noted that it was still the
preferred strategy from a cost-effectiveness perspective,
when cost effectiveness is defined by an ICER of <$50 000/
QALY. Awide range of costswere included in the analyses (eg,
Medicare reimbursements only vs inclusion of capital and
other costs), and assumptions on differences in toxicity and
cancer control between the two treatments varied widely.
3.2.2. Quality of the evidence
All studies comparing IMRT with 3D-CRT were observa-
tional in nature with substantial inconsistencies across
studies, which could be largely explained by differences in
the populations examined and in study design. Thus, we
rate the overall quality of evidence as low, indicating that
the true cost difference between IMRT and 3D-CRT may be
substantially different from that reported in the reviewed
studies .
3.3. Proton beam therapy versus IMRT
3.3.1. Proton beam therapy versus IMRT from the payer’s perspective
Three studies compared the costs of proton beam therapy
and IMRT (two from the payer’s, one from the societal, and
none from the hospital’s perspective; ). The two
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 7 1 2 – 7 3 5 725