Primary treatments for clinically localised prostate cancer: a comprehensive lifetime cost-utility analysis

Authors


Correspondence: Matthew R. Cooperberg, University of California, San Francisco, Box 1695, 1600 Divisadero St, A-624, San Francisco, CA 94143-1695, USA.

e-mail: mcooperberg@urology.ucsf.edu

Abstract

What's known on the subject? and What does the study add?

  • Multiple treatment alternatives exist for localised prostate cancer, with few high-quality studies directly comparing their comparative effectiveness and costs.
  • The present study is the most comprehensive cost-effectiveness analysis to date for localised prostate cancer, conducted with a lifetime horizon and accounting for survival, health-related quality-of-life, and cost impact of secondary treatments and other downstream events, as well as primary treatment choices. The analysis found minor differences, generally slightly favouring surgical methods, in quality-adjusted life years across treatment options. However, radiation therapy (RT) was consistently more expensive than surgery, and some alternatives, e.g. intensity-modulated RT for low-risk disease, were dominated – that is, both more expensive and less effective than competing alternatives.

Objective

  • To characterise the costs and outcomes associated with radical prostatectomy (open, laparoscopic, or robot-assisted) and radiation therapy (RT: dose-escalated three-dimensional conformal RT, intensity-modulated RT, brachytherapy, or combination), using a comprehensive, lifetime decision analytical model.

Patients and Methods

  • A Markov model was constructed to follow hypothetical men with low-, intermediate-, and high-risk prostate cancer over their lifetimes after primary treatment; probabilities of outcomes were based on an exhaustive literature search yielding 232 unique publications.
  • In each Markov cycle, patients could have remission, recurrence, salvage treatment, metastasis, death from prostate cancer, and death from other causes.
  • Utilities for each health state were determined, and disutilities were applied for complications and toxicities of treatment.
  • Costs were determined from the USA payer perspective, with incorporation of patient costs in a sensitivity analysis.

Results

  • Differences across treatments in quality-adjusted life years across methods were modest, ranging from 10.3 to 11.3 for low-risk patients, 9.6–10.5 for intermediate-risk patients and 7.8–9.3 for high-risk patients.
  • There were no statistically significant differences among surgical methods, which tended to be more effective than RT methods, with the exception of combined external beam + brachytherapy for high-risk disease.
  • RT methods were consistently more expensive than surgical methods; costs ranged from $19 901 (robot-assisted prostatectomy for low-risk disease) to $50 276 (combined RT for high-risk disease).
  • These findings were robust to an extensive set of sensitivity analyses.

Conclusions

  • Our analysis found small differences in outcomes and substantial differences in payer and patient costs across treatment alternatives.
  • These findings may inform future policy discussions about strategies to improve efficiency of treatment selection for localised prostate cancer.
Abbreviations
ADT

androgen-deprivation therapy

BT

brachytherapy

CaPSURE

Cancer of the Prostate Strategic Urologic Research Endeavor

CSM

cancer-specific mortality

ED

erectile dysfunction

HRQL

health-related quality of life

QALYs

quality-adjusted life years

(O)(L)(RA)RP

(open) (laparoscopic-assisted) (robot-assisted) radical prostatectomy

(3DC)(EB)(IM)RT

(three-dimensional conformal) (external-beam) (intensity-modulated) radiation therapy

Introduction

Clinical practice guidelines for localised prostate cancer endorse active surveillance, radical prostatectomy (RP), external-beam radiation therapy (EBRT), and brachytherapy (BT) as alternatives that should be offered to men with clinically localised disease [1, 2]. However, few high-quality comparative effectiveness studies exist to guide decisions among these alternatives. Recently, studies from large observational cohorts have identified differences in long-term oncological outcomes across treatment methods [3, 4], but randomised trials comparing treatments have not been completed. Absent consensus about the optimal treatment, prostate cancer treatment is both preference- and supply-sensitive, and tremendous variation exists in primary management strategies [5]. Differences across treatments in definitions of recurrence [6], health-related quality of life (HRQL) domains affected [7], and other considerations complicate efforts to compare surgical with radiation-based treatments. Substantial differences in cost also have been documented [8].

Decision and cost-effectiveness analyses have examined specific topics, e.g. the utility of active surveillance [9] and proton-beam therapy [10], but no such analysis has yet addressed the larger question of relative cost-effectiveness, at various strata of disease risk of the most commonly used treatments, surgery vs radiation therapy (RT). We aimed to determine costs and quality-adjusted outcomes between surgery and RT, including the various methods within these two broad categories.

Patients and Methods

A four-phase literature search was conducted. In Phase 1, the published literature on local prostate cancer treatments was searched via PubMed and yielded 7008 candidate articles. Limiting to English articles reporting on human subjects since 2002 reduced the pool to 3583, and further restricting to clinical trials (randomised or not), meta-analyses, and other explicit comparative studies yielded 988 articles. Titles and abstracts were then manually reviewed and studies were selected that reported a sample size of ≥20 men with clinically localised disease and did not combine results from different treatment methods (e.g. BT and BT+EBRT). Meta-analyses were excluded at this stage, as were papers that were superseded by subsequent reports from the same cohort. A final set of 374 articles was thus identified at the end of Phase 1 (eTable 1).

Phases 2 and 3 of the literature search were performed concurrently. Systematic application of inclusion/exclusion criteria specific to each clinical parameter was conducted for all articles from Phase 1 and 60 selected hand-picked manuscripts. For three-dimensional conformal RT (3DCRT), to reflect contemporary practice, only papers reporting results from dose-escalated series were included in the base-case analysis [11, 12]. In all, 22 cost and utility information sources were hand-selected. When duplicates were eliminated at the end of Phase 3, a total of 202 publications remained. In Phase 4, 30 additional articles were used in manuscript preparation, yielding a final set of 232 unique publications provided sources for all study data (eTable 2). The final list of references is presented in eTable 3. Probabilities for all outcomes were derived from the literature review and validated by the expert panel.

A decision-analytic Markov model was developed to evaluate the clinical outcomes, quality-adjusted life years (QALYs), and lifetime costs for a hypothetical cohort of men with clinically localised (clinical stage ≤ T3aN0M0) prostate cancer. After each treatment analysed (open RP [ORP], laparoscopic-assisted RP [LRP], robot-assisted RP [RARP], 3DCRT, intensity-modulated RT [IMRT], BT, and EBRT+BT), possible health states for each 1-month Markov cycle were remission, biochemical recurrence, metastasis, death from prostate cancer, and death from other causes. With each cycle, patients incurred costs, and those experiencing complications or adverse effects of treatment accrued disutilities. eFigure 1 presents the full decision tree. The analysis was stratified by clinical risk at diagnosis according to the three-level classification endorsed by the clinical practice guideline [1]; however, because this schema is frequently modified or adapted in various published studies, strict adherence to the risk criteria was not required for study inclusion.

Treatments

Men undergoing ORP, LRP, or RARP were assigned probabilities of erectile dysfunction (ED) and incontinence at each Markov cycle (Table 1). In all, 76% of surgery patients in biochemical recurrence were assumed to receive salvage treatment [4]. Low-, intermediate-, and high-risk patients were 75%, 50%, and 25% likely, respectively, to receive salvage RT and the remainder of patients received androgen-deprivation therapy (ADT) alone. Salvage RT was assumed to be IMRT given with 6 months of ADT [4, 13]. Salvage RT yielded a possibility of returning to the remission state, whereas salvage ADT alone did not. Both salvage methods entailed costs and potential adverse effects.

Table 1. Surgical complications and RT-related toxicities.
 RARP, %ORP, %LRP, % 
Proportion of patients with surgical complications:    
Hernia 0.61.82.1 
Urinary retention1.31.53.3 
Rectal injury0.51.11.4 
Lymphocele0.71.81.5 
Sepsis0.30.30.1 
Ileus0.61.31.3 
Bleeding episode1.26.80.8 
UTI1.61.82.2 
Deep vein thrombosis0.51.10.5 
Pulmonary embolism0.50.50.5 
Myocardial infarction0.20.20.2 
Anastomotic leakage3.55.25.7 
Urinary stricture/bladder neck contracture1.03.21.2 
Proportion of patients with acute RT-related toxicity:IMRT, %BT, %3DCRT, %EBRT+BT, %
GI grade 215.62.533.910.3
GI grade ≥ 30.102.21.1
GU grade 229.811.235.219.5
GU grade ≥ 32.33.33.73.5
Percentage of ED at baseline:All methods
Age 50–59 years26
Age 60–69 years40
Age ≥ 70 years61
Proportion of patients with new-onset ED:RARP, %ORP, %LRP, % 
3 months 66 (50; 83)66 (50; 83)75 (56; 94) 
6 months50 (38; 63)63 (47; 79)58 (44; 73) 
12 months42 (32; 53)58 (44; 73)53 (40; 66) 
24 months28 (21; 35)49 (37; 61)40 (30; 50) 
 IMRT, %BT, %3DCRT, %EBRT+BT, %
12 months 27 (20; 34)57 (43; 71)27 (20; 34)41 (31; 51)
24 months42 (32; 53)43 (32; 54)42 (32; 53)51 (38; 64)
Proportion of patients with urinary incontinence at:RARPORPLRP 
  1. GI, gastrointestinal; GU, genitourinary. For ED, urinary incontinence, and late RT-related toxicity, the first number is the base-case estimate, and the numbers in parenthesis are the lower and upper bounds of the ranges tested in sensitivity analyses.
3 months 19 (14; 24)32 (24; 40)41 (31; 51) 
6 months9 (7; 11)24 (18; 30)28 (21; 35) 
12 months9 (7; 11)11 (8; 14)10 (8; 13) 
Late RT-related toxicityIMRT, %BT, %3DCRT, %EBRT+BT
GI grade ≥ 2 (annual probability)1.6 (0.8; 2.4)1.3 (0.6; 1.9)6.3 (3.1; 9.4)2.3 (1.13; 3.4)
GU grade ≥ 2 (annual probability)2.3 (1.2; 3.4)4.2 (2.1; 6.4)4.1 (2.0; 6.1)3.4 (1.7; 5.10)

The decision tree for men undergoing 3DCRT, IMRT, BT, or EBRT+BT was similar, but incontinence was replaced with grade ≥2 gastrointestinal and/or genitourinary toxicity per Radiation Therapy Oncology Group (RTOG) criteria (Table 1) [14]. Patients receiving a treatment including EBRT were assumed not to receive concurrent ADT if they were low-risk, 50% likely to receive 6 months of treatment if they were intermediate-risk and 75% likely to receive 18 months of treatment if they were high-risk [15]. In all, 25% of BT patients were assumed to receive a short course of neoadjuvant ADT for prostate downsizing [15]. RT patients in recurrence had the possibility of salvage and return to the remission state with surgery, or of secondary treatment with ADT alone. In all, 44% of RT patients were assumed to receive salvage therapy, 4% with RP and 96% with ADT only [4, 13].

Outcomes

Short-term outcomes (surgical complications and acute radiation toxicity) could only accrue once. ED, incontinence, and delayed radiation toxicity could persist for multiple cycles, with a probability of resolution. Perioperative mortality was assumed to be 0.2% for RRP and 0.1% for LRP and RARP [16]. Parameter estimates for other complications and adverse events are listed in Table 1.

Over 150 different definitions of biochemical recurrence have been proposed [6]. We included studies reporting the most common: for surgery patients, a PSA level of ≥0.2 ng/mL with or without verification, a PSA level of >0.3 ng/mL, or a PSA level of ≥0.4 ng/mL, also allowing for secondary treatment to define failure. For RT patients, we included studies reporting outcomes using the American Society for Therapeutic Radiology and Oncology (ASTRO) or Phoenix definitions, two PSA level rises above the nadir to ≥1.0 ng/mL, or a PSA level of ≥0.4 ng/mL after nadir [6]. The parameter estimates used for biochemical recurrence derived from publications are listed in Table 2. For both surgery and RT, success rates for salvage local therapy in returning the patient to the remission state were 70%, 60%, and 50% for low-, intermediate-, and high-risk disease, respectively [17-21].

Table 2. Estimates of biochemical recurrence (BCR).
 RARPORPLRPIMRTBT3DCRTEBRT+BT
  1. β dist, β distribution.
Low-risk patients       
Weighted average annual probability of BCR, %1.06 (β dist; se = 0.007)1.06 (β dist; se = 0.007)1.06 (β dist; se = 0.007)1.11 (β dist; se = 0.003)1.09 (β dist; se = 0.004)2.17 (β dist; se = 0.003)1.18 (β dist; se = 0.006)
10-year BCR-free survival, %89.989.9%89.9%89.489.680.3%88.8
Intermediate-risk patients       
Weighted average annual probability of BCR, %3.74 (β dist; se = 0.020)3.74 (β dist; se = 0.020)3.74 (β dist; se = 0.020)3.04 (β dist; se = 0.005)3.67 (β dist; se = 0.010)3.04 (β dist; se = 0.008)2.11 (β dist; se = 0.009)
10-year BCR-free survival, %68.368.368.3%73.468.873.5%80.8
High-risk patients       
Weighted average annual probability of BCR, %7.05 (β dist; se = 0.034)7.05% (β dist; se = 0.034)7.05 (β dist; se = 0.034)5.63 (β dist; se = 0.012)7.83 (β dist; se = 0.016)6.50 (β dist; se = 0.005)3.29 (β dist; se = 0.012)
10-year BCR-free survival, %48.148.148.156.044.251.171.6

Biochemical recurrence is itself an important endpoint, to the extent that it leads to additional testing and treatment, and causes anxiety. However, definitions of recurrence after surgery and RT are not comparable, by the nature of their calculation, reflecting the different biological effects of RT and surgery, the RT definitions shift the survival curves substantially to the right, and thus may introduce bias in favour of RT [22, 23]. Moreover, recurrence by no means uniformly predicts progression to metastasis and prostate cancer-specific mortality (CSM) [24].

Therefore, estimates for time to metastasis from recurrence for RP [24, 25] and RT patients [26] were determined based on publications to account for these variances. The median times used in the model for surgery and RT patients were 10 and 6 years, respectively for low-risk patients, 8 and 4 years for intermediate-risk patients, and 6 and 2 years for high-risk patients. These times were further varied in sensitivity analyses. Time to CSM after first onset of metastasis was assumed to be 3.5 years for all patients [4]. Mortality from non-prostate cancer causes was based on National Center for Health Statistics actuarial data [27]. Use of ADT was assumed to increase the risk of non-prostate cancer mortality by 1% annually. RT was assumed to be associated with an annual probability of bladder or rectal cancer of 0.16% starting 5 years after treatment [28]; mortality from these secondary pelvic malignancies was assumed to be 12.9% annually [29].

Each of the health states was assigned a utility weight, determined from publications and the Cost-Effectiveness Analysis Registry (http://www.cearegistry.org). These utilities, listed in Table 3, were validated by the expert panel and extensively tested in sensitivity analyses. Disutility values for short- and long-term complications of surgery or RT were subtracted from the health state utilities. Use of ADT was also assigned a fixed disutility value. For each cycle, the final utility score was multiplied by 1 month and discounted by 3% annually. These quality-adjusted life-months were summed over the lifetime to determine the QALYs.

Table 3. Utilities and disutilities for health states and side-effects.
ParameterUtility value (duration)
  1. β dist, β distribution; GI, gastrointestinal; GU, genitourinary. For each health state, the base-case utility or disutility value is presented along with the distribution or range tested in sensitivity analysis.
On-going health states: 
Remission0.92 [β dist; se = 0.020]
Biochemical failure without hormone therapy0.84 [β dist; se = 0.031]
Biochemical failure with hormone therapy0.78 [β dist; se = 0.031]
Metastasis0.45 [β dist; se = 0.015]
Death (prostate cancer or all-cause)0.00
Secondary malignancy0.40
Late toxicities (disutilities; subtracted from current health state): 
GU grade ≥ 20.15 (1 year) [0.075; 0.225]
GI grade ≥ 20.20 (1 year) [0.1; 0.3]
Both GU and GI0.25 (1 year) [0.175; 0.375]
Functional outcomes (disutilities; subtracted from current health state): 
ED0.10 [0.05; 0.15] (2 years) [5 years; lifetime]
Urinary incontinence0.20 [0.1; 0.3] (2 years) [5 years; lifetime]
Both ED and urinary incontinence0.25 [0.125; 0.375] (2 years) [5 years; lifetime]

Costs

To determine costs, medical resource utilisation (office visits, procedures, hospitalisations, medications, imaging and laboratory tests, etc.) was assigned to each treatment, and subsequently to each health state, reflecting complications of treatment where relevant. All services and products were described using coding taxonomies applicable to the Medicare fee-for-service payment system and validated by a certified coding expert. Costs associated with the resources were derived from the Fiscal Year 2009 National Medicare Fee Schedules and, in the case of medications, the 2009 Drug Topics Redbook. Costs were validated by clinical experts. In the case of BT+EBRT, two-thirds of the EBRT treatments was assumed to be IMRT and one-third was assumed to be 3DCRT. In either case, the cost of salvage EBRT was assumed to be two-thirds the cost of EBRT given as primary monotherapy. Costs were determined from the payer perspective; thus capital and maintenance costs for equipment were not separately included, as these are purported to be reflected in aggregate payment to providers. However, time spent by patients in treatment and recovery was estimated by the expert panel, and indirect costs were assessed by associating these times with wage losses based on 2008 Bureau of Labor Statistics hourly rates weighted by employment status and age cohort size and inflated by 2%.

Statistical Analyses

QALY outcomes and cost differences among treatments were assessed using anova; adjustment for multiple comparisons across treatments was made using the Tukey test. The study used a cost-utility analysis, in which the marginal cost for a treatment with improved outcomes is determined in terms of cost per QALY gained. In the event that one treatment was found to be dominant, that is, more efficacious and less costly, then cost-minimisation analysis was used in lieu of cost-utility analysis.

Probabilistic Monte Carlo simulation was used to follow hypothetical patients with prostate cancer undergoing the treatment alternatives. For critical variables, parameter distributions were used rather than fixed point estimates. A normal distribution centred at age 65 years was assumed for age at first treatment, triangular distributions for treatment costs, and β distributions for utilities and biochemical failure probabilities. These are shown in eFigure 2. The probability distributions were sampled 250 times and 250 first-order simulations were performed with each parameter set.

An extensive set of one-way and multi-way sensitivity analyses were performed to determine the effects of varying the parameter estimates for various cost and outcome variables. Where method- and risk-specific comparisons allowed, validation of the model-based predictions of prostate cancer death with outcomes published from two large cancer centres [4] were conducted. The value ranges included for sensitivity analyses are included in Tables 1-4. The analyses were performed using TreeAge Pro 2009 (TreeAge Software, Williamstown, MA).

Table 4. Direct and indirect costs.
ParameterDirect medical cost (range), $Patient time cost, $ (sensitivity analysis only)
  1. aSummation of the unit cost of treating each complication (from costing algorithm) multiplied by the probability that a patient will have the complication. GI, gastrointestinal; GU, genitourinary.
Treatment methods and expected cost of first-year sequelae  
Treatment methods:  
RARP8547 (6410; 10684)2362
ORP8056 (6042; 10070)4099
LRP8547 (6410; 10684)2988
IMRT27084 (20313; 33855)1529
BT14106 (10580; 17633)973
3DCRT13013 (9827; 16379)1529
EBRT+BT29142 (21857; 36428)1997
Short-term surgical complications:  
RARPa709152
ORPa1518322
LRPa1019229
Acute GI and GU toxicities:  
IMRTa340171
BTa23066
3DCRTa638375
EBRT+BTa383157
Late toxicities:  
GI grade ≥ 2a1026 (513; 2052)897
GU grade ≥ 2a1387 (694; 2774)188
Functional outcomes:  
ED  
Year 1 (accounts for patients who choose more invasive, one-time treatment)1411 (706; 2822)167
Year 2+505 (253; 1010)
Urinary incontinence  
Year 1 (accounts for patients who choose more invasive, one-time treatment)946 (473; 1892)73
Year ≥ 2565 (283; 1130)
On-going health states:  
Remission (annual cost):  
Without ADT476 (238; 952)139
With neoadjuvant ADT1481
With adjuvant ADT2267
Biochemical recurrence (annual cost):  
Without salvage therapy1775 (888; 3550)278
With salvage therapy 
ADT (one-time/annual)2565 (0/5,130)/1791 (896/3582)
RT (one-time)27586 (20690; 34483)
Surgery (one-time)8547 (6410; 10684)
Metastasis:  
Annual management2212 (1106; 4424)1112
Evaluation (one-time)960 (480; 1920)
Treatment (one-time)15773 (7887; 31546)
Secondary malignancy11465 (5733; 22930)0
Prostate cancer death (last year of life)40807 (20404; 81614)0
All-cause mortality00

Results

The results from the base-case analysis are presented in Table 5. The likelihood of disease recurrence, progression, and mortality increased with increasing baseline disease risk, as did associated lifetime costs. QALYs for each of the methods studied were relatively similar within a given risk stratum, and fell with increasing levels of risk. The differences across methods were modest but statistically significant; among low-risk patients, 3DCRT was the least effective RT method (10.3 QALYs), and for intermediate- and high-risk patients EBRT+BT was the most effective RT method (10.1 and 9.1 QALYs, respectively, P < 0.001). There were no significant differences among the surgical methods for QALYs (11.3, 10.3–10.4, and 9.2–9.3 QALYs, respectively, for low-, intermediate-, and high-risk), and, in all comparisons except EBRT+BT vs ORP for high-risk patients, the surgical alternatives were statistically significantly more effective than the RT methods in terms of QALYs.

Table 5. Mean discounted costs, QALYs and undiscounted survival.
Treatment methodMean (sd)
Costs, $QALYsLife years% prostate cancer death
  1. No differences were found between surgical methods for cost or QALYs. *Significantly less expensive than each RT method (P < 0.001). Significantly more effective than each RT method (P = 0.008). Significantly less expensive than other RT methods (P < 0.001). §Significantly more effective than 3DCRT (P < 0.001). Significantly more effective than other RT methods (P < 0.001). **Significantly less expensive than EBRT+BT and IMRT (P < 0.001).
Low risk    
EBRT+BT40 588 (3573)10.7§ (0.5)16.2 (0.7)7.4 (3.9)
BT25 067 (2213)10.8§ (0.5)16.2 (0.7)6.9 (2.8)
3DCRT27 626 (1830)10.3 (0.4)15.5 (0.6)12.1 (2.4)
IMRT37 718 (3033)10.8§ (0.4)16.2 (0.7)6.9 (2.7)
ORP20 245* (2701)11.3 (0.4)16.7 (0.6)2.6 (2.1)
RARP19 901* (2684)11.3 (0.4)16.7 (0.6)2.7 (1.9)
LRP20 497* (2877)11.3 (0.4)16.7 (0.6)2.7 (2.1)
Intermediate risk    
EBRT+BT43 566 (4218)10.1 (0.5)15.3 (0.7)13.3 (5.0)
BT32 553 (3311)9.6 (0.5)14.6 (0.7)21.2 (5.1)
3DCRT30 838 (2699)9.7 (0.5)14.8 (0.7)18.5 (4.4)
IMRT44 639 (3096)9.6 (0.4)14.7 (0.6)19.0 (3.7)
ORP28 589* (5457)10.4 (0.6)15.6 (0.9)13.0 (6.0)
RARP28 017* (5453)10.5 (0.6)15.6 (0.8)12.8 (5.9)
LRP29 041* (5581)10.4 (0.6)15.6 (0.8)13.4 (6.3)
High risk    
EBRT+BT50 276 (4667)9.1 (0.6)14.1 (0.9)23.6 (7.1)
BT43 952** (3477)7.8 (0.5)12.1 (0.8)43.0 (5.9)
3DCRT42 397** (2348)7.9 (0.4)12.5 (0.6)38.2 (3.6)
IMRT53 539 (4013)8.2 (0.6)12.9 (0.8)34.2 (6.0)
ORP36 279* (5902)9.2 (0.8)13.9 (1.1)27.9 (8.1)
RARP35 014* (5895)9.3 (0.8)14.1 (1.1)26.8 (8.2)
LRP35 118* (6085)9.3 (0.7)14.2 (1.0)26.4 (8.2)

As a validation test of the oncologic outcomes resulting from our model, we compared rates of CSM derived from the model for IMRT and ORP patients to those published in a large, multi-centre academic series reported by Zelefsky et al. [4]. Assuming a starting age of 60 years for ORP patients and 69 for IMRT patients, as was reported in the Zelefsky et al. series, CSM rates at 8 years in our model were 0.9%, 3.2%, and 8.8% for low-, intermediate-, and high-risk IMRT patients, respectively, and 0.3%, 2.0%, and 5.0% for ORP patients. These results matched closely to the published rates of 0%, 4.5%, and 9.5% for IMRT and 0%, 1.9%, and 3.8% for ORP (Fig. 1) [4].

Figure 1.

Effects of varying assumptions of the interval between biochemical recurrence and metastasis. A correction factor of 4 years difference between surgical and RT methods was assumed in the base case, as detailed in the text, to reflect differences in definitions of biochemical recurrence across methods. Model-derived cancer-specific mortality estimates are shown at this base-case assumption, and with the difference varied from 0 to 6 years. The base-case assumptions at 4 years are compared with the outcomes published for IMRT vs ORP by Zelefsky et al. [4].

As summarised in Table 5, given similar biochemical outcomes and payer and patient costs across the surgical methods, lifetime costs were statistically and clinically similar within risk strata across the surgical methods (≈$20 000, $28 500, and $35 500, respectively, for low-, intermediate, and high-risk patients). Lifetime costs for RT, conversely, varied substantially across methods within risk strata. For low- and intermediate-risk patients, BT was less expensive than the other methods ($25 067 and $32 553 for low- and intermediate-risk); for high-risk patients, BT ($43 952) and 3DCRT ($42 397) were both less expensive than BT+EBRT ($50 376) or IMRT ($53 539, P < 0.001). Regardless of risk, the RT methods consistently entailed higher costs than the surgical methods in each risk stratum (P = 0.008).

We conducted an extensive set of sensitivity analyses, varying the key parameters in the model to determine which exert the most influence on the model outcomes (Table 6). Many of these analyses had no or minimal impact on the model. For example, varying the probabilities, duration, disutility penalties, and costs for functional outcomes, including incontinence, ED, and late radiation toxicity, had no substantial impact on the relative costs and benefits for any of the treatment methods. Incorporating patient time costs into the model increased total costs for all methods by roughly $4000–$7000, but again had little effect on the relative costs among methods; the same was true of varying probabilities of salvage therapy use, secondary malignancy rates and costs, other costs such as those assigned to salvage therapy, and the discount rate (including a non-discounted analysis).

Table 6. Sensitivity analyses with base-case default estimates and analysis range.
Analysis descriptionRTSurgeryComments
Base caseUpperLowerBase caseUpperLower
  1. GI, gastrointestinal; GU, genitourinary; UI, urinary incontinence.
Costs, $       
With patient time costsWithoutWithNAWithoutWithNA 
Annual on-going management costs for remission, $476952238476952$23850%, 200% of base-case estimate
Prostate cancer death costs, $40 80781 61420 40440 80781 61420 40450%, 200% of base-case estimate
Metastasis treatment costs, $960/15 7731 920/31 546480/7887960/15 7731 920/31 546480/7887Evaluation cost/treatment cost; 50%, 200% of base-case estimate
Annual on-going management costs for biochemical failure, $1 7753 5508881 7753 55088850%, 200% of base-case estimate
Annual on-going management costs for metastasis, $2 2124 4241 1062 2124 4241 10650%, 200% of base-case estimate
Salvage therapy costs, $Comment 1Comment 2Comment 3Comment 1Comment 2Comment 3
  1. one-time and annual ADT costs (2565/1791), RT (27 586), or surgery (8547)
  2. one-time and annual ADT costs (5130/3582), RT (34 483), or surgery (10 684)
  3. one-time and annual ADT costs (0/896), RT (20 690), or surgery (6410)
Functional outcomes costs, $Comment 1Comment 2Comment 3Comment 1Comment 2Comment 3
  1. Annual Year 1/Year 2+ ED costs (1411/505); annual Year 1/Year 2+ UI costs (946/565)
  2. Annual Year 1/Year 2+ ED costs (2822/1010); annual Year 1/Year 2+ UI costs (1892/1130)
  3. Annual Year 1/Year 2+ ED costs (706/253); annual Year 1/Year 2+ UI costs (473/283)
Late toxicity costs, $Comment 1Comment 2Comment 3NANANA
  1. Annual grade ≥ 2 GI cost (1026); annual grade ≥ 2 GU costs (1387)
  2. Annual grade ≥ 2 GI cost (2052); annual grade ≥ 2 GU costs (2774)
  3. Annual grade ≥ 2 GI cost (513); annual grade ≥ 2 GU costs (694)
Secondary malignancy costs, $11 46522 9305 73311 46522 9305 73350%, 200% of base-case estimate
Probabilities       
Discount rate, %350350 
Escalated and non-dose-escalated 3DCRT annual biochemical failure rate data      Data modified for 3DCRT only, not other RT methods
low risk0.0220.034NANANANA
intermediate risk0.0300.061NANANANA
high risk0.0650.074NANANANA
Metastasis-free survival after biochemical failure, years       
low risk69310137
intermediate risk4628106
high risk231675
Incremental metastasis-free survival period between methods after biochemical failure, yearsComment 1Comment 1Comment 1471
  1. To conduct analyses, it was necessary to hold estimates for RT methods constant and vary estimates for surgical methods
Time to prostate cancer death after metastases, years3.5523.552 
Probability of salvage therapy success, %       
low risk709050709050
intermediate risk608040608040
high risk507030507030
Probability of mortality related to ADT, %120120 
Probability of EDComment 1Comment 2Comment 3Comment 1Comment 2Comment 3
  1. 12/24 month probabilities of ED for EBRT+BT, BT, 3DCRT, and IMRT are 41%/51%, 57%/43%, 27%/42%, and 27%/42%, respectively; 3/6/12/24 month probabilities of ED for ORP, RALP, and LRP are 66%/63%/58%/49%, 66%/50%/42%/28%, and 75%/58%/53%/40%, respectively
  2. 12/24 month probabilities of ED for EBRT+BT, BT, 3D-CRT, and IMRT are 51%/64%, 71%/54%, 34%/53%, and 34%/53%, respectively; 3/6/12/24 month probabilities of ED for ORP, RALP, and LRP are 83%/79%/73%/61%, 83%/63%/53%/35%, and 94%/73%/66%/50%, respectively [125% of base-case estimates]
  3. 12/24 month probabilities of ED for EBRT+BT, BT, 3D-CRT, and IMRT are 31%/38%, 43%/32%, 20%/32%, and 20%/32%, respectively; 3/6/12/24 month probabilities of ED for ORP, RALP, and LRP are 50%/47%/44%/37%, 50%/38%/32%/21%, and 56%/44%/40%/30%, respectively [75% of base-case estimates]
Probability of UINANANAComment 1Comment 2Comment 3
  1. 3/6/12 month probabilities of UI for ORP, RALP, and LRP are 32%/24%/11%, 19%/9%/9%, and 41%/28%/10%, respectively
  2. 3/6/12 month probabilities of UI for ORP, RALP, and LRP are 40%/30%/14%, 24%/11%/11%, and 51%/35%/13%, respectively [125% of base-case estimates]
  3. 3/6/12 month probabilities of UI for ORP, RALP, and LRP are 24%/18%/8%, 14%/7%/7%, and 31%/21%/8%, respectively [75% of base-case estimates]
Annual probability of late GI/GU toxicityComment 1Comment 2Comment 3NANANA
  1. Annual probabilities of GI/GU toxicity for EBRT+BT, BT, 3D-CRT, and IMRT are -->0.023/0.034, 0.013/0.042, 0.063/0.041, and 0.016/0.023, respectively
  2. Annual probabilities of GI/GU toxicity for EBRT+BT, BT, 3D-CRT, and IMRT are 0.034/0.051, 0.019/0.064, 0.094/0.061, and 0.024/0.034, respectively [150% of base-case estimates]
  3. Annual probabilities of GI/GU toxicity for EBRT+BT, BT, 3D-CRT, and IMRT are 0.011/0.017, 0.006/0.021, 0.031/0.020, and 0.008/0.011, respectively [50% of base-case estimates]
Probability of salvage therapy use, %4460307610050 
Annual RT-related secondary malignancy rate0.0016NA0NANANA 
Utilities       
Functional outcomes disutilities0.1/0.2/0.250.15/0.3/0.3750.05/0.1/0.1250.1/0.2/0.250.15/0.3/0.3750.05/0.1/0.125Disutility of ED, UI, and both ED+UI, respectively; range represents 50%, 150% of base-case estimate
Late toxicity disutilities0.15/0.2/0.250.225/0.3/0.3750.075/0.1/0.1250.15/0.2/0.250.225/0.3/0.3750.075/0.1/0.125Disutility of GU grade ≥ 2, GI grade ≥ 2, and both GU grade ≥ 2 and GI grade ≥ 2, respectively; range represents 50%, 150% of base-case estimate
Other       
Mean (sd) starting age distribution, years65 (6)69 (6)NA65 (6)NA60 (6)Based on approximate age distributions in Zelefsky et al. [4]
Time horizon, yearsLifetime105Lifetime105Both alternative estimates represent reductions from the base-case estimate.
Duration of functional outcomes, years2Lifetime52Lifetime5Both alternative estimates represent increases from the base-case estimate.

In the base case, the median age for all patients was 65 years; if surgical patients were assumed to be younger and RT patients older, reflecting actual practice [30], the differences in costs and CSM between surgical and RT patients were reduced, but corresponding differences in QALYs and overall survival increased. Varying the costs of salvage therapy and management of biochemical failure and metastasis, as well as the probability of mortality attributable to ADT use, impacted the model for intermediate- and high-risk patients only. While varying the estimates for these parameters resulted in changes of ≈5–15% from the base-case results, the changes were not sufficiently different by treatment method to alter conclusions related to the relative costs and benefits of the methods. Including reported data on non-dose-escalated 3DCRT resulted in substantially worse survival and QALY outcomes for this method.

Varying the time from biochemical recurrence to metastasis had the greatest impact on clinical and economic outcomes. Varying the assumption of a 4-year differential in terms of time between recurrence and metastasis had a strong effect on CSM estimates for men at intermediate- and high-risk. The CSM rates cross at 0 years differential for intermediate-risk tumours and at 1 year for high-risk tumours [4] (Fig. 1). These changes resulted in differences in QALYs and costs as well.

Discussion

There is no consensus for optimal management of localised prostate cancer, and patterns of management vary tremendously [15, 31, 32]. To date, clinical trials of intervention vs conservative management have been completed [33, 34], but those comparing surgery with RT have not [35]. One such trial has now accrued, but results will not be available for several years [36]. In the interim, cost-effectiveness analyses may shed important light on the question of which method or methods offer the best value relative to cost. These analyses are notably scarce in prostate cancer, however; a recent systematic review identified only 22 studies published through 2007, compared with 86, for example, in breast cancer [37].

The present model found, in the context of the USA reimbursement system, statistically significant but relatively modest differences among treatment methods in terms of QALYs (Table 5). In general, surgery was preferred over RT for lower-risk men, whereas combined EBRT+BT compared favourably for high-risk men. However, across the risk spectrum, RT was consistently more expensive. Some treatment strategies are thus considered dominated: IMRT for low- and intermediate-risk men, for example, is no more effective than surgery or BT, and is substantially more expensive. These findings generally were robust to a wide range of sensitivity analyses. The assumption that led to the greatest change in outcome in sensitivity analysis was the differential in time from recurrence to metastasis between surgical and RT patients. These effects were most dramatic in intermediate- and high-risk patients and could lead to changes in conclusions related to the relative costs and benefits of RT and surgery for prostate cancer. Future research related to correction for different recurrence definitions is warranted.

The present findings also are consistent with other recent studies based on carefully risk-adjusted retrospective studies of prospectively collected cohorts, which have found consistent evidence for improved distal clinical outcomes after surgery compared with EBRT. A study from the community-based Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) registry found a roughly two-fold increase in CSM among men treated with various RT approaches compared with surgery [3]. The Zelefsky et al. [4] series similarly found a three-fold difference in CSM comparing RRP patients to those receiving high-dose IMRT. Notably, both studies found the greatest differences among men at relatively high levels of risk, and neither included men treated with BT. Another multicentre academic series, reached similar conclusions; that study did include BT patients, whose outcomes were better in some analyses than those of EBRT patients [38].

The present results are also generally consistent with other recently published studies on costs and outcomes of treatment. A recent Medicare study reported statistically significant but relatively modest benefits for IMRT over conventional RT in some but not all quality-of-life domains [39]. Another Medicare study found that while the marginal costs of robotic compared with ORP were relatively modest and declined over time through the middle part of the last decade, the costs of IMRT compared with conventional RT were very high, and relatively stable [40]. Neither of these studies included BT patients. The present analysis found relatively minor differences between ORP and RARP. Indeed, a recent meta-analysis found advantages for RARP for short-term perioperative outcomes and margin rates, but no large study has yet shown clear advantages for either approach in terms of longer-term oncological or HRQL outcomes [41]. In the context of a lifetime decision analysis, any impact of short-term outcomes will generally be limited.

As described above, the absolute rates of risk-stratified CSM in the present model corresponded fairly closely to those reported by Zelefsky et al. [4]. However, the relative difference in mortality between RT and surgery patients was lower in the present model than in either the Zelefsky et al. [4] study or the CaPSURE study [3], suggesting that our analysis is relatively conservative in its estimation of the life-year and QALY differences between the surgical and RT methods. Our cost assumptions are generally consistent with those recently determined by another CaPSURE study [8].

Several limitations to this analysis should be considered. Primary ADT monotherapy for localised disease is commonly used in practice [15], but outcomes of this approach in the USA are sparsely reported, and it is not included as a standard option in the practice guideline [1]. Active surveillance, conversely, is rapidly gaining acceptance, including endorsement in practice guidelines [1], as a viable option for men with low-risk disease [42, 43], and for carefully selected men with intermediate-risk disease [44]. A recent cost-effectiveness analysis in fact found slightly greater QALYs for surveillance compared with immediate treatment for low-risk disease [9]. That study did not include costs, but another did find cost savings for initial surveillance over treatment, depending in part on the likelihood and timing of delayed treatment among patients initially surveilled [45]. We agree entirely that for low-risk disease active surveillance may well be preferred to any of the methods included in the present analysis. However, neither long-term oncological outcomes nor HRQL outcomes have been reported to date. Therefore, to avoid adding additional layers of complexity, active surveillance was not included in our model, but will certainly be the subject of future modelling efforts.

Multiple assumptions underlie the present model. Utilities for various after treatment health states, for example, are based on the best available in the literature, but these have not been extensively validated. Our literature review began in 2002; thus not all studies used to derive probabilities reflected the most recent improvements in treatment methods. Other variables, such as increased mortality attributable to ADT or secondary malignancy, are the subject of significant on-going controversy. Fortunately, none of these factors proved to be strong determinants of overall QALYs or costs, and were tested in sensitivity analyses with only minor impacts. It is important to stress that the economic analysis was performed from the USA payer perspective, with the additional incorporation in sensitivity analysis of indirect patient time costs. This approach does not account for hospital investments in capital equipment, disposables, and maintenance. These costs are theoretically reflected in insurance payments, but in fact in the USA government and private payers reimburse at substantially higher levels for IMRT, for example, compared with 3DCRT, but do not do so for RARP vs ORP. Particularly germane to the question of the cost-effectiveness of RARP vs ORP, then, is the fact that the costs associated with the robotic platform that are absorbed by hospitals are not reflected.

These assumptions reflect the present situation in the USA, and will vary substantially across other healthcare systems. Despite these caveats, we believe that through incorporation of both QALYs and costs, consistent risk-stratification, inclusion of multiple methods within surgery and RT, and use of a lifetime horizon, the present analysis is the most comprehensive economic analysis yet undertaken for this disease. With the exception of the time to metastasis from recurrence, the findings are robust to sensitivity analyses, and may inform future policy discussions about strategies to improve efficiency and reduce variation in localised prostate cancer care.

Acknowledgements and Conflict of Interest

Avalere Health LLC and Veritas Health Economics Consulting were commissioned by Intuitive Surgical (Sunnyvale, CA, USA) to perform the cost effectiveness analysis. However, the study sponsor had no role whatsoever in the collection, analysis, or interpretation of the data; in writing or approving the manuscript; or in the decision to submit for publication. Dr Ramakrishna was compensated as a consultant to Avalere. None of the other non-Avalere/Veritas authors received any direct financial or other remuneration for their work on this study. Dr Cooperberg's effort was supported by National Institutes of Health/National Cancer Institute (5RC1CA146596), and by the Agency for Healthcare Research and Quality (1U01CA88160).

Ancillary