Page 135 - An Evidence Review of Active Surveillance in Men With Localized Prostate Cancer

the physical functioning and bodily pain dimensions of the SF-36 instrument, the spiritual and
total wellbeing scores of the Quality of Life-Cancer Survivors (QoL-CS) instrument, and the
bowel function and bowel bother dimensions of Expanded Prostate Cancer Index Composite
(EPIC), compared to observational management. No other significant difference between
treatments was observed for general QoL, cancer-specific QoL, or disease-specific-QoL.

Comparison Between Observational Management Strategies and Combination Therapy or
Active Treatments Considered in Aggregate
Detailed results from the two studies that compared AS/WW and other active treatment
groups (a combined group of patients managed with RP or RT and a group receiving both RT
and EBRT) are shown in Appendix Table C4.4. No study reported incidence of metastatic
disease or quality of life outcomes.
One study analyzed data from the SEER-Medicare database up to year 2002, and compared
patients on observation with patients who received any active treatment, including RP, EBRT,
and BT. 208 Compared with patients on observation, patients on active treatment had significantly
lower risk of prostate cancer-specific mortality (adjusted HR 0.67; 95 percent CI 0.58, 0.77) and
all-cause mortality (adjusted HR 0.69; 95 percent CI 0.66, 0.72). 208
One study analyzed the data from the CaPSURE registry and found that a group of men
treated with EBRT and BT (combined treatment) had a higher rate of receiving treatments for
urethral stricture than men on WW over a median followup of 2.7 years (adjusted HR 4.56; 95
percent CI 1.23, 16.88). 218 No significant difference was found between patients on WW and
patients treated with combined RP and EBRT. 218

Costs

Findings from Studies of Actual Patient Costs
We identified four primary studies (three using U.S. data 159,235,236 and one from Sweden 237 )
reporting on comparisons of costs of active treatments and observational management strategies
for localized prostate cancer. All studies included groups of patients treated with WW and used
various active treatments as comparators; no comparative study reporting on costs included a
group managed with AS. Details from each study including the populations, treatments
compared, and cost estimates are presented in Appendix Table C4.5.
One study used the SEER-Medicare database (13,769 patients matched 1:1 with noncancer
controls; 2805 of the patients had been managed with WW) to estimate incremental treatment
costs during the first 5 years of treatment. Using inverse probability of treatment weights derived
from a propensity score to account for factors that affect treatment selection, this study found
that, WW has lower incremental costs ($8535) compared with RP ($19,481) or RT ($16,653)
over 5 years. 159
A second study used data from the CaPSURE database (235 patients; 37 managed with WW)
to estimate mean first year costs. 235 The unadjusted mean cost of WW ($484) was lower
compared with RP (without hormonal therapy, $7320) or RT (without hormonal therapy, $7430).
After adjusting for patient and disease characteristics, the difference in costs among treatments
was statistically significant (analysis of covariance P < 0.001).
A third study, again using CaPSURE data, estimated that WW had a total cumulative cost
over 5.5 years of follow up of $31,871 and $31,789, for low and intermediate risk prostate cancer
patients, respectively. The corresponding values were $28,366 and $41,419 for BT; $48,840 and

81