Ablative therapy for people with localised prostate cancer: a systematic review and economic evaluation

To address the objectives outlined earlier, the research will have the following inter-related components: I. Development of care pathways for patients with localised prostate cancer (objective 1) The care pathways and related resource use for patients with localised prostate cancer will be mapped out, building upon previous work by our group. Core datasets will be developed for the systematic review, resources, costs and utilities (for health states and the outcomes and processes of care). This approach will ensure that the work conducted in the other elements of the project will generate the parameter estimates required for the economic model. II.Systematic review of the safety and effectiveness of ablative procedures compared to alternative therapies (objectives 2a-c) The three reviews may also be considered as one large review with three sub-questions i.e. low/intermediate and high risk localised disease, and local relapse following radiotherapy are all contained within a search for localised disease though the comparator treatments differ between reviews. Type of Included studies: Randomised controlled trials, non-randomised comparative studies and case series (for only the ablative procedures). Eligible patients will have stage T1 or T2 disease stratified into localised low/intermediate risk and localised high risk of progression based upon established criteria. Men with localised disease (confined to the prostate gland) will be included but we will exclude locally advanced disease (T3/T4). Comparator treatments: For the primary therapy review on low/intermediate risk localised prostate cancer (objective 2a), the ablative therapies being considered are cryotherapy, HIFU, PDT, RITA, laser ablation and brachytherapy. The comparators will be active surveillance, radical prostatectomy and EBRT. For the primary therapy review on high risk localised prostate cancer (objective 2b), the ablative therapies being considered are cryotherapy, HIFU, PDT, RITA, laser ablation and brachytherapy. The comparators will be radical prostatectomy and EBRT. For the salvage therapy review (objective 2c), the ablative therapies being considered are cryotherapy and HIFU. The comparator will be radical prostatectomy. Outcomes:(1) Cancer related: Biochemical (PSA) recurrence and/or treatment failure (as defined by trialists); Disease free and overall survival;re-intervention. (2) Functional: Sexual function; Urinary continence; Urinary symptoms; Bowel symptoms. (3) Patient-driven: Generic and disease-specific quality of life. (4) Procedural: Procedure time; Nature of anaesthetic; Length of hospital stay. (5) Adverse effects. Electronic searches: Databases searched will include Medline, Medline in Process, Embase, CINAHL, HTA amongst others, to identify reports of published studies. Highly sensitive search strategies will be designed, including appropriate subject headings and text word terms, interventions under consideration and included study designs. Reference lists of all included studies will be scanned. Ongoing studies will be through identified though Current Controlled Trials, NIHR Portfolio and WHO International Clinical Trials Registry. Websites of manufacturers, professional organisations, regulatory bodies and the HTA will be checked for unpublished reports. Manufactures of the ablative therapies will be contacted to contribute relevant data. Assessment of study risk of bias: Multiple quality assessment tools will be used reflecting the various study designs. The Cochrane Collaboration's risk of bias tool will be used to assess the risk of bias in randomised and (modified form) nonrandomised comparative studies. A 17-question checklist will be used to assess case series. Two reviewers will independently screen titles and abstracts, assess fulltext copies of potentially relevant reports and undertake data abstraction of included studies. Any disagreements will be resolved by consensus or arbitration by a third person. Data analysis: An indirect comparison (cross design) approach allowing inclusion of non-randomised comparative data and case series will be used. We will examine heterogeneity between and within different study designs using a Bayesian hierarchical random effects model enabling use of all available evidence. Differences between interventions will be assessed by the corresponding odds ratio and 95% credible interval. III. Economic evaluation (objectives 3 & 4) We will construct a discrete event simulation model to estimate the costs, long-term effects and relative efficiency of the alternative interventions (objective 3). The parameters will be derived from systematic reviews described above with cost and utilities informed by additional focused literature reviews. The results of the economic model will be presented as costs, quality-adjusted life years (QALYS) and as a cost-utility analysis (CUA). We will conduct extensive deterministic and probabilistic sensitivity analysis. Value of information analysis will identify priorities for research and assess the expected value of removing uncertainty surrounding specific parameters or groups of parameters (objective 4).