EURURO Vol. 72 No. 5

732 860

studies that examined the cost effectiveness of proton beam
therapy from a payer’s perspective were both from the USA.
The first study found an ICER of $63 578/QALY for a 70 yr old
and $55 726/QALY for a 60 yr old . The second study
examined 553 Medicare beneficiaries who received proton
beam therapy and 27 094 patients who received IMRT in
2008 and 2009 After matching IMRT and proton beam
therapy patients based on known confounders, the study
found the median sum of Medicare reimbursements in the
3 mo after treatment to be $32 428 for proton beam therapy
and $18 575 for IMRT. Patients in this study who received
proton beam therapy were younger, healthier, and from
more affluent areas. There were no differences in the two
treatments in terms of gastrointestinal and genitourinary
complications at 12 mo, but proton beam therapy was
significantly costlier .
3.3.2. Proton beam therapy versus IMRT from the societal
perspective
A Swedish study examined the costs of proton beam
therapy and IMRT from a societal perspective They
used a Markov model to assess cost effectiveness and found
that the mean cost of proton beam therapy was s13 491
compared with s5477 for IMRT, resulting in an increased
cost per QALY of s26 776. They accounted for both direct
and indirect costs, including the purchasing costs of proton
beam and IMRT facilities, amortization of these facilities,
and transportation and hotel accommodation. The authors
concluded that proton beam therapy may be cost effective if
appropriate patients are selected, such as those with a
higher risk of death from cancer . However, their
analyses were hindered by limited clinical effectiveness
data, leading to uncertainty in their assumptions . In
summary, there is little doubt that proton beam therapy is
costlier than IMRT. However, it remains unclear whether
this increased cost is worth it because the comparative data
on important outcomes such as cancer control and quality
of life remain very limited in the literature.
3.3.3. Quality of the evidence
All studies were observational and found a higher cost for
proton beam therapy. There were inconsistencies in the
point estimates that could be explained by different
populations and different time horizons. Two studies relied
on cost modeling. One study included patient-specific data,
but only 553 proton beam therapy patients were included,
increasing the risk for imprecise estimates. Thus, the quality
of the evidence on cost of proton therapy versus IMRT is
very low, indicating that the true effect is likely substan-
tially different from the estimates of the included studies
3.4. Uncertainties surrounding the benefits of new technologies
Studies addressing the cost utility or cost benefit from the
payer’s or societal perspective depend on whether out-
comes are improved by the use of new technology and on
the magnitude of that improvement. The main outcomes of
interest include cancer control, urinary function, and sexual
function. There have been several systematic reviews of the
literature addressing these outcomes for RARP. Regarding
cancer control, a meta-analysis performed in 2012 showed
clinically and statistically equivalent positive margin rates
with RARP and RRP (95% confidence interval [CI] for
difference ranging from –1.9% to 2.4% after propensity
score adjustment) . Similar positive surgical margin rates
were also found in a recent randomized trial from Australia
(90% CI for difference ranging from –1% to 11%) Bio-
chemical recurrence rates have largely been equivalent
between these two surgical approaches .
In a meta-analysis of observational studies, RARP
appears to have a statistically significant albeit small
advantage over RRP when comparing urinary function,
with 12-mo incontinence rates of 7.5% compared with
11.3% (absolute risk reduction 3.8%, p = 0.03) However,
these cumulative data were driven by results from only two
studies with only one of them using a validated question-
naire to assess urinary continence . Regarding erectile
function, patients undergoing RARP appear to have an
increased likelihood to recover potency at 12 mo based on a
cumulative analysis of six observational studies (recovery
rate of 75.8% for RARP vs 52.2% for RRP, p = 0.002)
However, the quality of these six observational studies
was limited, as they defined potency as an erection
sufficient for intercourse, which is neither a very objective
nor a reproducible definition More recently, a
randomized trial from Australia compared functional out-
comes between RARP and RRP at 12 wk and found no
significant differences between the two techniques
One meta-analysis that included 23 studies compared
outcomes after IMRT and 3D-CRT. IMRT was associated with
decreased acute (risk ratio 0.59; 95% CI 0.44–0.78) and late
gastrointestinal toxicity (risk ratio 0.54; 95% CI 0.38–0.78),
improved biochemical control (risk ratio 1.17; 95% CI 1.08–
1.27), and no change in late genitourinary toxicity or overall
survival . IMRT was associated with a slight increase in
acute genitourinary toxicity (risk ratio 1.08; 95% CI 1.00–
1.17) . The authors concluded that IMRT may be a better
choice, but cautioned that their analysis included hetero-
geneous studies and that the potentially increased risk of
secondary malignancies with IMRT needed further evalua-
tion . The comparative effectiveness of IMRT and
proton beam therapy is unclear. There are no completed
randomized clinical trials between these two modalities,
although there is one ongoing randomized trial examining
gastrointestinal toxicity among patients with low- and
intermediate-risk disease (NCT01617161) . Several
retrospective studies have shown that proton beam therapy
likely has similar rates of urinary and bowel toxicity to IMRT
while another study showed more gastrointestinal
toxicity with proton beam therapy In short, more
clinical information is needed before the true effectiveness
of proton beam therapy is known.
In summary, given the overall low level of evidence with
new technologies, there is significant uncertainty whether
the increased treatment costs are associated with better
outcomes. If ideal outcomes can be achieved on a
population level, RARP and IMRT have the potential to be
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 5 732