The study compared the lifetime cost-effectiveness of radical prostatectomy with watchful waiting in subgroups of localised prostate cancer patients with different risk profiles. The authors concluded that the cost-effectiveness of radical prostatectomy varied with patient risk profile such as age, Gleason score, and prostate-specific antigen value. Generally, favourable cost-effectiveness estimates were associated with younger patients. The study used a conventional cost-effectiveness framework and the authors’ conclusions appear valid.

Cost-utility analysis

The lifetime cost-effectiveness of radical prostatectomy was compared with watchful waiting in subgroups of localised prostate cancer patients with different risk profiles.

Radical prostatectomy involved of pelvic lymphadenectomy followed by radical prostatectomy if the histopathological evaluations of the removed lymph nodes were negative. With watchful waiting, no immediate treatment was given but infravesical (bladder) obstruction symptom relief was treated with transurethral resection.

Sweden/hospital.

Analytical approach:

The analysis was based on a semi-Markov model with a lifetime horizon. The authors stated that the perspective of the Swedish health care system was adopted.

Effectiveness data:

Most clinical inputs were taken from the 695 patients enrolled in a published randomised controlled trial (RCT) of the Scandinavian Prostatic Cancer Group Study Number 4 (SPCG-4, Bill-Axelson, et al. 2008, see 'Other Publications of Related Interest' below for bibliographic details). This trial directly compared radical prostatectomy with watchful waiting. Enrolled patients had a median follow-up (life expectancy) of longer than 10 years. An extensive statistical analysis was performed to identify baseline risk factors for prostate cancer progression including Gleason score, age and prostate-specific antigen values. All-cause mortality data were based on Swedish age-specific life tables for men.

Monetary benefit and utility valuations:

Utility valuations were based on data from Swedish patients enrolled in the SPCG-4 trial who completed a 77-item questionnaire with a visual analogue scale.

Measure of benefit:

Quality-adjusted life-years (QALYs) were used as the summary benefit measure and were discounted at an annual rate of 3.5%.

Cost data:

The costs were related to prostate cancer treatment but a breakdown of items was not given. Data on resource quantities and costs were based on evidence from the SPCG-4 trial using a retrospective approach. A generalised linear model was applied to associate resources to health states of the model. Costs were in Swedish kronor (SEK). The price year was 2007. A 3.5% discount rate per annum was applied.

Analysis of uncertainty:

An extensive probabilistic approach was used to investigate uncertainty around cost-utility ratios. The age of patients, Gleason score, and prostate-specific antigen values were the most important factors; their impact on cost-utility ratios was considered.

Incremental cost-utility ratios (the additional cost per QALY gained) substantially varied depending on the patients' subgroup risk profile. Generally, with higher age there was a higher incremental cost-utility ratio, which ranged from SEK 21,026 to SEK 858,703.

For example, in a patient aged 70 years with a Gleason score of 5 to 6 and a prostate-specific antigen value range of 11 to 20, radical prostatectomy led to an incremental cost of SEK 61,211 and a gain of 0.60 QALYs over watchful waiting, at an incremental cost-utility ratio of SEK 101,957. The probability of being cost-effective was 49% at a threshold of SEK 100,000, 81% at a threshold of SEK 200,000, and 87% at a threshold of SEK 300,000.

Assuming a threshold of SEK 200,000 per QALY, radical prostatectomy was always cost-effective for patients aged 65 years or less, while it was not cost-effective in patients aged 75 years with a low Gleason score and low prostate-specific antigen value.

The authors concluded that the cost-effectiveness of radical prostatectomy varied with patient age, Gleason score, and prostate-specific antigen value. Generally, favourable cost-effectiveness estimates were associated with younger patients.

Interventions:

The rationale for the selection of the comparators was clear. Two commonly used strategies for the management of patients with localised prostate cancer were considered. It was unclear whether other options would have been comparable.

Effectiveness/benefits:

The use of a large clinical trial to obtain treatment effect and patient characteristics should have ensured high internal validity (as stated by the authors). The use of a large sample size and long follow-up represented a strength of the analysis. Appropriate statistical analyses were conducted to take account of key patient characteristics that had an impact on clinical effect of treatments. Extensive sensitivity analyses were conducted. The clinical analysis was performed satisfactorily. QALYs were a valid benefit measure given the impact of prostate cancer on health-related quality of life. Utility valuations were taken from the large sample of patients enrolled in the trial and at several time-points; this should have ensured a detailed collection of estimates.

Costs:

The authors did not provide an extensive description of the economic side of the study. Specifically, the cost categories included in the analysis were not reported, although the model focused on the direct medical costs associated with prostate cancer care. Data on unit costs and resource quantities were not reported. This reduced the transparency of the cost analysis. The use of data only taken from the clinical trial should have ensured a detailed collection in the Swedish setting. However, data were taken retrospectively rather than prospectively and it was unclear whether all relevant items were considered. The price year was reported, which would allow reflation exercises in other time periods. A description of the statistical analyses used to estimate costs in the various subgroups of patients was given. Appropriate tests appear to have been used.

Analysis and results:

An incremental approach was used to synthesise the costs and benefits of the two strategies. A probabilistic approach was used to deal with uncertainty and the results were clearly reported. The study results were presented in detail for all patient risk subgroups. The authors acknowledged that the clinical trial was conducted several years before the publication of this study, and that some changes might have occurred in radical prostatectomy surgery techniques over this period. Study results were specific to the Swedish context where both the clinical and the economic data were retrieved, so it might be difficult to transfer these findings to other settings.

Concluding remarks:

The study used a conventional cost-effectiveness framework and the authors’ conclusions appear valid.

Not stated.

Bill-Axelson A, Holmberg L, Filen F, Ruutu M, Garmo H, Busch C, Nordling S, Haggman M, Andersson SO, Bratell S, Spangberg A, Palmgren J, Adami HO, Johansson JE, Scandinavian Prostate Cancer Group Study Number 4. Radical prostatectomy versus watchful waiting in localized prostate cancer: the Scandinavian prostate cancer group-4 randomized trial. Journal of the National Cancer Institute 2008;100(16):1144-1154.