Rituximab for the first-line treatment of stage III-IV follicular lymphoma (review of Technology Appraisal No. 110): a systematic review and economic evaluation.

BACKGROUND: Follicular lymphoma (FL) is a non-Hodgkin's lymphoma which typically presents when the disease is at an advanced stage. The majority of patients receive first-line therapy of rituximab in combination with chemotherapy, with two-thirds receiving cyclophosphamide, vincristine and prednisolone. The clinical and cost-effectiveness of other chemotherapies in combination with rituximab in first-line therapy is not known. OBJECTIVE: To systematically evaluate and appraise the clinical effectiveness and cost-effectiveness of rituximab (MabThera(®), Roche Products) in combination with chemotherapy, compared with chemotherapy alone, for the first-line treatment of symptomatic stage III-IV FL. DATA SOURCES: A systematic review of literature and an economic evaluation were carried out. Key databases [including MEDLINE In-Process & Other Non-Indexed Citations; Cumulative Index to Nursing and Allied Health Literature (CINAHL); EMBASE; The Cochrane Library, including the Cochrane Database of Systematic Reviews (CDSR), Cochrane Central Register of Controlled Trials (CENTRAL), Database of Abstracts of Reviews of Effects (DARE), NHS Economic Evaluation Database (NHS EED) and Health Technology Assessment (HTA) databases; Science Citation Index (SCI); and BIOSIS], plus research registers and conference proceedings, were searched for relevant studies from inception up to October 2010. REVIEW METHODS: One reviewer assessed titles and abstracts of studies identified by the search strategy, obtained the full text of relevant papers and screened them against inclusion criteria. Data from included studies were extracted by one reviewer using a standardised data extraction form and checked by a second reviewer. The quality of included studies was assessed by one reviewer and checked by a second. A patient-level simulation model was developed to estimate the costs and quality-adjusted life-year (QALY) gains from the perspective of the UK NHS and Personal Social Services, with costs and benefits discounted at 3.5% annually. RESULTS: Four randomised controlled trials comparing rituximab plus chemotherapy (R-chemotherapy) with chemotherapy alone in untreated, symptomatic patients with stage III-IV FL were identified. R-chemotherapy compared with chemotherapy alone increased the likelihood of a response to treatment in all four trials, with no additional toxicity of clinical relevance. Overall response rates were significantly improved in all four trials, with a difference between the R-chemotherapy and chemotherapy arms of between 5% and 24%, respectively. Complete response rates were also improved, with a difference between the R-chemotherapy and chemotherapy arms of between 2% and 25%, respectively. Exploratory meta-analyses were conducted; the level of statistical heterogeneity was very high and thus we believe the response rates from the individual trials to be a more robust estimator of the efficacy of the specific R-chemotherapy regimens. Over a follow-up period of 4-5 years, R-chemotherapy significantly increased the overall survival rate compared with chemotherapy alone in three trials, although data for two trials were compromised owing to the use of additional treatments. The incremental cost-effectiveness ratio (ICER) for the addition of rituximab to CVP (cyclophosphamide, vincristine and prednisolone), CHOP (cyclophosphamide, doxorubicin/adriamycin, vincristine and prednisolone) and MCP [mitoxantrone, chlorambucil (Leukeran(®), Aspen) and prednisolone] was £7720, £10,834 and £9316 per QALY gained, respectively, when it was assumed that first-line rituximab maintenance was not used. A scenario analysis is also presented, assuming that responders to R-chemotherapy in first-line induction receive maintenance with rituximab, increasing the ICER to £14,959, £21,687 and £20,493 per QALY gained, respectively. LIMITATIONS: These relate to the sources of data used for the effectiveness in first and second line and the assumed utility values; there is uncertainty about the effect of salvage treatment on patients who had been previously treated with an anthracycline regimen. There is uncertainty whether or not rituximab is as effective in second-line treatment when patients have been previously treated with rituximab. CONCLUSIONS: The results from four randomised trials comparing R-chemotherapy with chemotherapy alone showed an improvement in clinical effectiveness outcomes, with minimal clinically relevant additional adverse events or toxicity. The cost per QALY gained is estimated to be < £25,000 for all three comparisons under our base-case assumption and is considerably lower if first-line rituximab maintenance is not assumed. More data on patients pre-treated with rituximab and on the effect of first-line maintenance with rituximab is required for future work. FUNDING: The National Institute for Health Research Health Technology Assessment programme.

Magnetic resonance for assessment of axillary lymph node status in early breast cancer: a systematic review and meta-analysis.

INTRODUCTION: Current methods of identifying axillary node metastases in breast cancer patients are highly accurate, but are associated with several adverse events. This review evaluates the diagnostic accuracy of magnetic resonance imaging (MRI) techniques for identification of axillary metastases in early stage newly diagnosed breast cancer patients. METHODS: Comprehensive searches were conducted in April 2009. Study quality was assessed. Sensitivity and specificity were meta-analysed using a bivariate random effects approach, utilising pathological diagnosis via node biopsy as the comparative gold standard. RESULTS: Based on the highest sensitivity and specificity reported in each of the nine studies evaluating MRI (n = 307 patients), mean sensitivity was 90% (95% CI: 78-96%; range 65-100%) and mean specificity 90% (95% CI: 75-96%; range 54-100%). Across five studies evaluating ultrasmall super-paramagnetic iron oxide (USPIO)-enhanced MRI (n = 93), mean sensitivity was 98% (95% CI: 61-100%) and mean specificity 96% (95% CI: 72-100%). Across three studies of gadolinium-enhanced MRI (n = 187), mean sensitivity was 88% (95% CI: 78-94%) and mean specificity 73% (95% CI: 63-81%). In the single study of in-vivo proton MR spectroscopy (n = 27), sensitivity was 65% (95% CI: 38-86%) and specificity 100% (95% CI: 69-100%). CONCLUSIONS: USPIO-enhanced MRI showed a trend towards higher sensitivity and specificity and may make a useful addition to the current diagnostic pathway. Additional larger studies with standardised methods and standardised criteria for classifying a node as positive are needed. Current estimates of sensitivity and specificity do not support replacement of SLNB with any current MRI technology in this patient group.

Trabectedin for the treatment of relapsed ovarian cancer.

The paper presents a summary of the evidence review group (ERG) report into the clinical effectiveness and cost-effectiveness of trabectedin for the treatment of relapsed platinum-sensitive ovarian cancer, based upon a review of the manufacturer's submission to the National Institute for Health and Clinical Excellence (NICE) as part of the single technology appraisal process. The submission addressed only part of the decision problem and did not provide evidence to compare trabectedin (Yondelis®, PharmaMar) and pegylated liposomal doxorubicin hydrochloride (PLDH) (Caelyx®, Schering-Plough) with key comparators. The submission's direct comparison evidence came from one reasonable-quality randomised controlled trial (RCT) of trabectedin and PLDH versus PLDH alone (ET743-OVA-301). The results of the RCT were subdivided into the entire platinum-sensitive population (> 6-month relapse after initial platinum-based chemotherapy) and partially platinum-sensitive (≥ 6- to 12-month relapse) and fully platinum-sensitive (> 12-month relapse) populations. The outcomes included were overall survival, progression-free survival measured by three types of assessor, response rates, adverse effects of treatment, health-related quality of life and cost per quality-adjusted-life-year (QALY) gained. A mixed treatment comparison (MTC) meta-analysis comparing trabectedin and PLDH with single-agent PLDH within the entire platinum-sensitive population, with paclitaxel or with topotecan also formed part of the submission. The RCT data showed that trabectedin plus PLDH compared with PLDH monotherapy had a significant effect on overall survival only within the partially platinum-sensitive subgroup. PFS results reported by the independent radiologists showed significant effects in favour of the trabectedin and PLDH arm for the entire and partially platinum-sensitive populations only. Rates of grade 3 and 4 adverse events were mostly higher in the trabectedin and PLDH arm than in the PLDH alone arm. There were several issues regarding the undertaking of the MTC, and thus the data were not considered robust. Furthermore, the ERG did not believe the MTC to be necessary to answer the decision problem. The manufacturer submitted a de novo cost-effectiveness model. The main analysis compared trabectedin in combination with PLDH versus paclitaxel, topotecan and PLDH (each as monotherapy) in the entire platinum-sensitive population, using results estimated from the MTC. Additional analyses were presented comparing trabectedin in combination with PLDH versus PLDH monotherapy using direct evidence from the OVA-301 trial for the fully, partially and entire platinum-sensitive populations. The cost per QALY gained for trabectedin in combination with PLDH versus PLDH monotherapy was estimated to be £ 70,076 in the main analysis. In the additional analyses, the cost per QALY gained for trabectedin in combination with PLDH versus PLDH monotherapy was £ 94,832, £ 43,996 and £ 31,092 for the entire, partially and fully platinum-sensitive populations, respectively. Additional work was undertaken by the ERG using patient-level data and amending some assumptions to provide a better statistical fit to the Kaplan-Meier data than the exponential distribution assumed by the manufacturer. The ERG base-case estimate of the cost per QALY of trabectedin in combination with PLDH ranged from £46,503 to £54,607 in the partially platinum-sensitive population. At the time of writing, trabectedin in combination with PLDH for the treatment of women with relapsed platinum-sensitive ovarian cancer is not recommended by NICE in the final appraisal determination.

Evidence Review Group approaches to the critical appraisal of manufacturer submissions for the NICE STA process: a mapping study and thematic analysis.

BACKGROUND: The National Institute for Health and Clinical Excellence (NICE) single technology appraisal (STA) process was set up as a rapid way to appraise new technologies for use within the NHS in England and Wales and has been in place since 2005. OBJECTIVES: This study had five primary objectives: (1) to provide a map of the STA process to date; (2) to identify current approaches to the critical appraisal of manufacturers' submissions (MSs) by Evidence Review Groups (ERGs); (3) to identify recurring themes in clarification letters sent to manufacturers; (4) to provide recommendations for possible alternative approaches to be used in the critical appraisal process; and (5) to revise the current ERG report template. DATA SOURCES: Data for the mapping of the STA process were obtained from the NICE website (www.nice.org.uk). Data for the analyses of the ERG reports and clarification letters were taken from the reports and letters themselves. REVIEW METHODS: For the mapping, a spreadsheet was developed to collect data on 22 predefined variables related to timings and outcomes. Simple descriptive statistics were used to analyse the data. For the thematic analysis, a documentary analysis of 30 ERG reports was undertaken. Data on key elements of the MSs, the processes undertaken by ERGs and the strengths and weaknesses of MSs were extracted. A framework of a priori themes was developed. Data were extracted, coded and analysed according to a framework approach. Twenty-one clarification letters were examined and data were extracted using a set of codes to cover report quality, systematic review methods and clinical/economic issues. The current ERG report template was modified and sent to the current ERG teams for comment. All comments were considered and formed the basis for further revisions to the template. RESULTS: Ninety-five STAs were included in the mapping exercise. Many STAs were subject to delay or cancellation for a variety of reasons. The ERG reports highlighted the strengths and weaknesses of MSs to the STA process. Thematic analysis of these data offered a means of clarifying and describing these aspects of the submissions. This analysis generated five themes: process, reporting, satisfaction of objectives, reliability and validity of findings, and content. Points from clarification letters were analysed and presented in four main categories: report quality, systematic review methods, clinical data analysis and economic data analysis. LIMITATIONS: Nearly all data were obtained from the NICE website; therefore, any errors in the data on the website will be reflected in the mapping analysis presented in this report. Missing data for the mapping exercise limit the generalisability of the findings. Analyses were limited to what was reported in the ERG reports and the clarification letters. CONCLUSIONS: Guidance suggested for manufacturers will help to ensure that more appropriate submissions are received in the future while recommendations provided for ERG teams will help to guide teams to ensure that reporting methods are transparent. FUNDING: The National Institute for Health Research Health Technology Assessment programme.

The clinical effectiveness and safety of prophylactic retinal interventions to reduce the risk of retinal detachment and subsequent vision loss in adults and children with Stickler syndrome: a systematic review.

BACKGROUND: Stickler syndrome, also known as hereditary progressive arthro-ophthalmopathy, is an inherited progressive disorder of the collagen connective tissues. Manifestations include short-sightedness, cataracts, retinal problems leading to retinal detachment and possible blindness. This is principally the case among individuals with type 1 Stickler Syndrome. It is the most commonly identified inherited cause of retinal detachment in childhood. However, there is no consensus regarding best practice and no current guidelines on prophylactic interventions for this population. OBJECTIVES: The aim of this systematic review was to assess the evidence for the clinical effectiveness and safety of primary prophylactic interventions for the prevention of retinal detachment in previously untreated eyes without retinal detachment in patients with Stickler syndrome. The primary outcome of interest was retinal detachment post prophylaxis. DATA SOURCES: A systematic search was made of 11 databases of published and unpublished literature, which included MEDLINE, MEDLINE In-Process & Other Non-Indexed Citations, EMBASE, the Cumulative Index to Nursing and Allied Health Literature and The Cochrane Library. There was no restriction by language or date. The references of all included studies were checked for further relevant citations and authors of studies with potentially relevant data were also contacted. REVIEW METHODS: Two reviewers double-screened all titles and abstracts of the citations retrieved by the search to identify studies that satisfied the inclusion criteria. Both reviewers also independently extracted and quality assessed all included studies. A narrative synthesis was performed. RESULTS: The literature search identified 1444 unique citations, of which four studies satisfied the inclusion criteria. The two principal studies were both retrospective cohort studies with control groups in populations with type 1 Stickler syndrome. One study evaluated 360° cryotherapy (n = 204) and the other focal or circumferential laser treatment (n = 22). Both studies reported a statistically significant difference in the rate of retinal detachment per eye between the groups receiving prophylaxis and the controls. However, both studies were subject to a high risk of bias. The results of the two supporting studies of Wagner-Stickler patients were either relatively inconsistent or unreliable. No study reported any major or long-term complications associated with the interventions. Despite the weaknesses of the evidence, the rate of retinal detachment in the intervention groups, especially the cryotherapy group, was lower than the rate either experienced in the study control groups or reported in other studies of untreated Stickler syndrome populations not exposed to prophylaxis. CONCLUSIONS: Only 360° cryotherapy and focal and circumferential laser treatment have been evaluated for the type 1 Stickler syndrome population, and then only by a single retrospective, controlled, cohort study in each case. Both of these studies report a significant difference between intervention and control groups (principally no treatment) and no major or long-term side effects or complications. However, there is a high risk of bias within these two studies, so the relative effectiveness of either intervention is uncertain. FUTURE WORK: A service priority is to determine reliably the prevalence of Stickler syndrome, i.e. how many individuals have type 1 or type 2 Stickler syndrome, and their risk of retinal detachment and subsequent blindness. A non-randomised, prospective cohort comparison study, in which eligible participants are treated, followed-up and analysed in one of three study arms, for no treatment, laser therapy or cryotherapy, would potentially offer further certainty in terms of the relative efficacy of both prophylaxis versus no prophylaxis and cryotherapy versus laser therapy than is possible with the currently available data. Alternatively, continued follow-up and analysis of existing study data, and data collection from relevant sample populations, are required to assess the long-term risks of blindness, retinal detachment and prophylaxis. FUNDING: This study was funded by the National Institute for Health Research Health Technology Assessment programme.

Positron emission tomography (PET) and magnetic resonance imaging (MRI) for the assessment of axillary lymph node metastases in early breast cancer: systematic review and economic evaluation.

BACKGROUND: Breast cancer is the most common type of cancer in women. Evaluation of axillary lymph node metastases is important for breast cancer staging and treatment planning. OBJECTIVES: To evaluate the diagnostic accuracy, cost-effectiveness and effect on patient outcomes of positron emission tomography (PET), with or without computed tomography (CT), and magnetic resonance imaging (MRI) in the evaluation of axillary lymph node metastases in patients with newly diagnosed early-stage breast cancer. DATA SOURCES: A systematic review of literature and an economic evaluation were carried out. Key databases (including MEDLINE, EMBASE and nine others) plus research registers and conference proceedings were searched for relevant studies up to April 2009. A decision-analytical model was developed to determine cost-effectiveness in the UK. REVIEW METHODS: One reviewer assessed titles and abstracts of studies identified by the search strategy, obtained the full text of relevant papers and screened them against inclusion criteria. Data from included studies were extracted by one reviewer using a standardised data extraction form and checked by a second reviewer. Discrepancies were resolved by discussion. Quality of included studies was assessed using the quality assessment of diagnostic accuracy studies (QUADAS) checklist, applied by one reviewer and checked by a second. RESULTS: Forty-five citations relating to 35 studies were included in the clinical effectiveness review: 26 studies of PET and nine studies of MRI. Two studies were included in the cost-effectiveness review: one of PET and one of MRI. Of the seven studies evaluating PET/CT (n = 862), the mean sensitivity was 56% [95% confidence interval (CI) 44% to 67%] and mean specificity 96% (95% CI 90% to 99%). Of the 19 studies evaluating PET only (n = 1729), the mean sensitivity was 66% (95% CI 50% to 79%) and mean specificity 93% (95% CI 89% to 96%). PET performed less well for small metastases; the mean sensitivity was 11% (95% CI 5% to 22%) for micrometastases (≤ 2 mm; five studies; n = 63), and 57% (95% CI 47% to 66%) for macrometastases (> 2 mm; four studies; n = 111). The smallest metastatic nodes detected by PET measured 3 mm, while PET failed to detect some nodes measuring > 15 mm. Studies in which all patients were clinically node negative showed a trend towards lower sensitivity of PET compared with studies with a mixed population. Across five studies evaluating ultrasmall super-paramagnetic iron oxide (USPIO)-enhanced MRI (n = 93), the mean sensitivity was 98% (95% CI 61% to 100%) and mean specificity 96% (95% CI 72% to 100%). Across three studies of gadolinium-enhanced MRI (n = 187), the mean sensitivity was 88% (95% CI 78% to 94%) and mean specificity 73% (95% CI 63% to 81%). In the single study of in vivo proton magnetic resonance spectroscopy (n = 27), the sensitivity was 65% (95% CI 38% to 86%) and specificity 100% (95% CI 69% to 100%). USPIO-enhanced MRI showed a trend towards higher sensitivity and specificity than gadolinium-enhanced MRI. Results of the decision modelling suggest that the MRI replacement strategy is the most cost-effective strategy and dominates the baseline 4-node sampling (4-NS) and sentinel lymph node biopsy (SLNB) strategies in most sensitivity analyses undertaken. The PET replacement strategy is not as robust as the MRI replacement strategy, as its cost-effectiveness is significantly affected by the utility decrement for lymphoedema and the probability of relapse for false-negative (FN) patients. LIMITATIONS: No included studies directly compared PET and MRI. CONCLUSIONS: Studies demonstrated that PET and MRI have lower sensitivity and specificity than SLNB and 4-NS but are associated with fewer adverse events. Included studies indicated a significantly higher mean sensitivity for MRI than for PET, with USPIO-enhanced MRI providing the highest sensitivity. However, sensitivity and specificity of PET and MRI varied widely between studies, and MRI studies were relatively small and varied in their methods; therefore, results should be interpreted with caution. Decision modelling based on these results suggests that the most cost-effective strategy may be MRI rather than SLNB or 4-NS. This strategy reduces costs and increases quality-adjusted life-years (QALYs) because there are fewer adverse events for the majority of patients. However, this strategy leads to more FN cases at higher risk of cancer recurrence and more false- positive (FP) cases who would undergo unnecessary axillary lymph node dissection. Adding MRI prior to SLNB or 4-NS has little effect on QALYs, though this analysis is limited by lack of available data. Future research should include large, well-conducted studies of MRI, particularly using USPIO; data on the long-term impacts of lymphoedema on cost and patient utility; studies of the comparative effectiveness and cost-effectiveness of SLNB and 4-NS; and more robust UK cost data for 4-NS and SLNB as well as the cost of MRI and PET techniques. FUNDING: This study was funded by the Health Technology Assessment programme of the National Institute of Health Research.

Positron emission tomography (PET) for assessment of axillary lymph node status in early breast cancer: A systematic review and meta-analysis.

PURPOSE: Sentinel lymph node biopsy (SLNB) and axillary lymph node dissection (ALND) are used to assess axillary nodal status in breast cancer, but are invasive procedures associated with morbidity, including lymphoedema. This systematic review evaluates the diagnostic accuracy of positron emission tomography (PET), with or without computed tomography (CT), for assessment of axillary nodes in early breast cancer. METHODS: Eleven databases including MEDLINE, EMBASE and the Cochrane Library, plus research registers and conference proceedings, were searched in April 2009. Study quality was assessed using the QUality Assessment of Diagnostic Accuracy Studies (QUADAS) checklist. Sensitivity and specificity were meta-analysed using a bivariate random effects approach. RESULTS: Across 26 studies evaluating PET or PET/CT (n = 2591 patients), mean sensitivity was 63% (95% CI: 52-74%; range 20-100%) and mean specificity 94% (95% CI: 91-96%; range 75-100%). Across 7 studies of PET/CT (n = 862), mean sensitivity was 56% (95% CI: 44-67%) and mean specificity 96% (90-99%). Across 19 studies of PET-only (n = 1729), mean sensitivity was 66% (50-79%) and mean specificity 93% (89-96%). Mean sensitivity was 11% (5-22%) for micrometastases (≤2 mm; five studies; n = 63), and 57% (47-66%) for macrometastases (>2 mm; four studies; n = 111). CONCLUSIONS: PET had lower sensitivity and specificity than SLNB. Therefore, replacing SLNB with PET would avoid the adverse effects of SLNB, but lead to more false negative patients at risk of recurrence and more false positive patients undergoing unnecessary ALND. The present evidence does not support the routine use of PET or PET-CT for the assessment of the clinically negative axilla.

Chemoprevention of colorectal cancer: systematic review and economic evaluation.

BACKGROUND: Colorectal cancer (CRC) is the third most common cancer in the UK: incidence increases with age, median age at diagnosis being over 70 years. Approximately 25% of cases occur in individuals with a family history of CRC, including 5% caused by familial adenomatous polyposis (FAP) or hereditary non-polyposis CRC (HNPCC). Most develop from adenomatous polyps arising from the intestine lining. Individuals with these polyps undergo polypectomy and are invited for endoscopic surveillance. Screening via faecal occult blood testing has been rolled out across the UK. OBJECTIVES: To evaluate the clinical effectiveness and cost-effectiveness of drug and micronutrient interventions for the prevention of CRC and/or adenomatous polyps. Interventions considered include: non-steroidal anti-inflammatory drugs (NSAIDs), including aspirin and cyclo-oxygenase-2 (COX-2) inhibitors; folic acid; calcium; vitamin D and antioxidants (including vitamin A, vitamin C, vitamin E, selenium and beta-carotene). Chemoprevention was assessed in the general population, in individuals at increased risk of CRC, and in individuals with FAP or HNPCC. DATA SOURCES: A systematic review identified randomised controlled trials (RCTs) assessing drug and nutritional agents for the prevention of CRC or adenomatous polyps. A separate search identified qualitative studies relating to individuals' views, attitudes and beliefs about chemoprevention. MEDLINE, MEDLINE In-Process & Other Non-Indexed Citations, EMBASE, CINAHL, the Cochrane Database of Systematic Reviews, Cochrane CENTRAL Register of Controlled Trials, DARE, NHS-EED (NHS Economic Evaluation Database), HTA database, Science Citation Index, BIOSIS previews and the Current Controlled Trials research register were searched in June 2008. Data were extracted by one reviewer and checked by a second. REVIEW METHODS: The synthesis methods used were systematic review and meta-analysis for RCTs and qualitative framework synthesis for qualitative studies. A health economic model was developed to assess the cost-effectiveness of chemoprevention for two populations with different levels of risk of developing CRC: the general population and an intermediate-risk population. RESULTS: The search identified 44 relevant RCTs and six ongoing studies. A small study of aspirin in FAP patients produced no statistically significant reduction in polyp number but a possible reduction in polyp size. There was a statistically significant 21% reduction in risk of adenoma recurrence [relative risk (RR) 0.79, 95% confidence interval (CI) 0.68 to 0.92] in an analysis of aspirin versus no aspirin in individuals with a history of adenomas or CRC. In the general population, a significant 26% reduction in CRC incidence was demonstrated in studies with a 23-year follow-up (RR 0.74, 95% CI 0.57 to 0.97). Non-aspirin NSAID use in FAP individuals produced a non-statistically significant reduction in adenoma incidence after 4 years of treatment and follow-up and reductions in polyp number and size. In individuals with a history of adenomas there was a statistically significant 34% reduction in adenoma recurrence risk (RR 0.66, 95% CI 0.60 to 0.72) and a statistically significant 55% reduction in advanced adenoma incidence (RR 0.45, 95% CI 0.35 to 0.58). No studies assessed the effect of non-aspirin NSAIDs in the general population. There were no studies of folic acid in individuals with FAP or HNPCC. There was no significant effect of folic acid versus placebo on adenoma recurrence (RR 1.16, 95% CI 0.97 to 1.39) or advanced adenoma incidence in individuals with a history of adenomas. In the general population there was no significant effect of folic acid on risk of CRC (RR 1.13, 95% CI 0.77 to 1.64), although studies were of relatively short duration. Calcium use by FAP patients produced no significant reduction in polyp number or disease progression. In individuals with a history of adenomas there was a statistically significant 18% reduction in risk of adenoma recurrence (RR 0.82, 95% CI 0.69 to 0.98) and a non-significant reduction in risk of advanced adenomas (RR 0.77, 95% CI 0.50 to 1.17). In the general population there was no significant effect of calcium on risk of CRC (RR 1.08, 95% CI 0.87 to 1.34), although studies were of relatively short duration. There were no studies of antioxidant use in individuals with FAP or HNPCC, and in individuals with a history of adenomas no statistically significant differences in relative risk of adenoma recurrence were found. In the general population there was no difference in incidence of CRC (RR 1.00, 95% CI 0.88 to 1.13) with antioxidant use compared with no antioxidant use. Twenty studies reported qualitative findings concerning chemoprevention. People are more likely to use NSAIDs if there is a strong perceived need. Perceptions of risk and benefit also influence decision-making and use. People have fewer concerns about using antioxidants or other supplements, but their perception of the benefits of these agents is less well-defined. The model analysis suggested that the most cost-effective age-range policy in the general population would be to provide chemoprevention to all individuals within the general population from age 50 to 60 years. The use of aspirin in addition to screening within the general population is likely to result in a discounted cost per life-year gained of around 10,000 pounds and a discounted cost per quality-adjusted life-year (QALY) gained of around 23,000 pounds compared with screening alone. In the intermediate-risk group the most economically viable age-range policy would be to provide chemoprevention to individuals following polypectomy aged 61 to 70 years. Calcium is likely to have a discounted cost per QALY gained of around 8000 pounds compared with screening alone. Although aspirin in addition to screening should be more effective and less costly than screening alone, under the current assumptions of benefits to harms of aspirin and calcium, aspirin is expected to be extendedly dominated by calcium. LIMITATIONS: Whilst a number of studies were included in the review, the duration of follow-up was generally insufficient to detect an effect on cancer incidence. Given the uncertainties and ambiguities in the evidence base, the results of the health economic analysis should be interpreted with caution. CONCLUSIONS: Aspirin and celecoxib may reduce recurrence of adenomas and incidence of advanced adenomas in individuals with an increased risk of CRC and calcium may reduce recurrence of adenomas in this group. COX-2 inhibitors may decrease polyp number in patients with FAP. There is some evidence for aspirin reducing the incidence of CRC in the general population. Both aspirin and NSAIDs are associated with adverse effects so it will be important to consider the risk-benefit ratio before recommending these agents for chemoprevention. The economic analysis suggests that chemoprevention has the potential to represent a cost-effective intervention, particularly when targeted at intermediate-risk populations following polypectomy.

Avoiding and identifying errors in health technology assessment models: qualitative study and methodological review.

BACKGROUND: Health policy decisions must be relevant, evidence-based and transparent. Decision-analytic modelling supports this process but its role is reliant on its credibility. Errors in mathematical decision models or simulation exercises are unavoidable but little attention has been paid to processes in model development. Numerous error avoidance/identification strategies could be adopted but it is difficult to evaluate the merits of strategies for improving the credibility of models without first developing an understanding of error types and causes. OBJECTIVES: The study aims to describe the current comprehension of errors in the HTA modelling community and generate a taxonomy of model errors. Four primary objectives are to: (1) describe the current understanding of errors in HTA modelling; (2) understand current processes applied by the technology assessment community for avoiding errors in development, debugging and critically appraising models for errors; (3) use HTA modellers' perceptions of model errors with the wider non-HTA literature to develop a taxonomy of model errors; and (4) explore potential methods and procedures to reduce the occurrence of errors in models. It also describes the model development process as perceived by practitioners working within the HTA community. DATA SOURCES: A methodological review was undertaken using an iterative search methodology. Exploratory searches informed the scope of interviews; later searches focused on issues arising from the interviews. Searches were undertaken in February 2008 and January 2009. In-depth qualitative interviews were performed with 12 HTA modellers from academic and commercial modelling sectors. REVIEW METHODS: All qualitative data were analysed using the Framework approach. Descriptive and explanatory accounts were used to interrogate the data within and across themes and subthemes: organisation, roles and communication; the model development process; definition of error; types of model error; strategies for avoiding errors; strategies for identifying errors; and barriers and facilitators. RESULTS: There was no common language in the discussion of modelling errors and there was inconsistency in the perceived boundaries of what constitutes an error. Asked about the definition of model error, there was a tendency for interviewees to exclude matters of judgement from being errors and focus on 'slips' and 'lapses', but discussion of slips and lapses comprised less than 20% of the discussion on types of errors. Interviewees devoted 70% of the discussion to softer elements of the process of defining the decision question and conceptual modelling, mostly the realms of judgement, skills, experience and training. The original focus concerned model errors, but it may be more useful to refer to modelling risks. Several interviewees discussed concepts of validation and verification, with notable consistency in interpretation: verification meaning the process of ensuring that the computer model correctly implemented the intended model, whereas validation means the process of ensuring that a model is fit for purpose. Methodological literature on verification and validation of models makes reference to the Hermeneutic philosophical position, highlighting that the concept of model validation should not be externalized from the decision-makers and the decision-making process. Interviewees demonstrated examples of all major error types identified in the literature: errors in the description of the decision problem, in model structure, in use of evidence, in implementation of the model, in operation of the model, and in presentation and understanding of results. The HTA error classifications were compared against existing classifications of model errors in the literature. A range of techniques and processes are currently used to avoid errors in HTA models: engaging with clinical experts, clients and decision-makers to ensure mutual understanding, producing written documentation of the proposed model, explicit conceptual modelling, stepping through skeleton models with experts, ensuring transparency in reporting, adopting standard housekeeping techniques, and ensuring that those parties involved in the model development process have sufficient and relevant training. Clarity and mutual understanding were identified as key issues. However, their current implementation is not framed within an overall strategy for structuring complex problems. LIMITATIONS: Some of the questioning may have biased interviewees responses but as all interviewees were represented in the analysis no rebalancing of the report was deemed necessary. A potential weakness of the literature review was its focus on spreadsheet and program development rather than specifically on model development. It should also be noted that the identified literature concerning programming errors was very narrow despite broad searches being undertaken. CONCLUSIONS: Published definitions of overall model validity comprising conceptual model validation, verification of the computer model, and operational validity of the use of the model in addressing the real-world problem are consistent with the views expressed by the HTA community and are therefore recommended as the basis for further discussions of model credibility. Such discussions should focus on risks, including errors of implementation, errors in matters of judgement and violations. Discussions of modelling risks should reflect the potentially complex network of cognitive breakdowns that lead to errors in models and existing research on the cognitive basis of human error should be included in an examination of modelling errors. There is a need to develop a better understanding of the skills requirements for the development, operation and use of HTA models. Interaction between modeller and client in developing mutual understanding of a model establishes that model's significance and its warranty. This highlights that model credibility is the central concern of decision-makers using models so it is crucial that the concept of model validation should not be externalized from the decision-makers and the decision-making process. Recommendations for future research would be studies of verification and validation; the model development process; and identification of modifications to the modelling process with the aim of preventing the occurrence of errors and improving the identification of errors in models.

Trabectedin for the treatment of advanced metastatic soft tissue sarcoma.

This paper presents a summary of the evidence review group (ERG) report into the clinical effectiveness and cost-effectiveness of trabectedin for the treatment of advanced metastatic soft tissue sarcoma, in accordance with the licensed indication, based on the evidence submission from the manufacturer to NICE as part of the single technology appraisal (STA) process. The outcomes stated in the manufacturer's definition of the decision problem were overall survival (OS), progression-free survival (PFS), response rates, adverse effects of treatment, health-related quality of life, and cost per quality-adjusted life-year (QALY) gained. The clinical evidence was derived from one randomised controlled trial (RCT), in which the licensed dose of trabectedin was compared with a different dose of trabectedin, and three phase II studies. In the RCT, the median OS was 13.9 months for the licensed dose of trabectedin, which was not significantly different from that for the comparator dose of trabectedin, which was 11.8 months. From the phase II uncontrolled trials, median OS was reported as 9.2 or 12.8 months. The RCT reported significantly superior PFS for the licensed dose of trabectedin (median 3.3 months) over the comparator trabectedin dose (median 2.3 months). One phase II uncontrolled trial reported median PFS as 1.9 months in the licensed dose of trabectedin. The RCT reported PFS rates at 6 months were 35.5% for the licensed dose of trabectedin, and 27.5% for the comparator dose of trabectedin. From the phase II uncontrolled trials, PFS rates at 6 months were 24.4% or 29%. For the RCT, deaths attributed to trabectedin occurred in 3.1% of the licensed dose, and 2.3% of the comparator group. The most common severe adverse events were neutropenia, although with a low rate of febrile neutropenia, thrombocytopenia, and aspartate aminotransferase and alanine aminotransferase elevation, although these were reported to be non-cumulative and reversible. Following dialogue iterations with the ERG team, the manufacturer revised the model twice. However, despite revisions, errors/inconsistencies were found in the latest version of the model and were corrected by the ERG (only for the base case). In the latest manufacturer's submission, the cost per QALY gained of trabectedin compared with best supportive care (BSC) was estimated to be 56,985 pounds for the base case using effectiveness from the STS (Soft Tissue Sarcomas)-201 trial for trabectedin and a pool analysis of the European Organisation for Research and Treatment of Cancer data set for BSC. This analysis was constrained to patients with L-sarcomas only. When the joint uncertainty between parameters was considered, the cost-effectiveness acceptability curve showed that trabectedin has a very low probability of being cost-effective at a threshold of 30,000 pounds per QALY gained compared with BSC for any scenario. The guidance has yet to be issued by NICE.

Vitamin K to prevent fractures in older women: systematic review and economic evaluation.

OBJECTIVE: To determine the clinical and cost-effectiveness of vitamin K in preventing osteoporotic fractures in postmenopausal women. DATA SOURCES: Searches were conducted in May 2007 in MEDLINE, MEDLINE In-Process, EMBASE, Cochrane Database of Systematic Reviews, Cochrane Controlled Trials Register, BIOSIS, CINAHL, DARE, NHS EED and HTA databases, AMED, NRR, Science Citation Index and Current Controlled Trials. The MEDLINE search was updated in March 2009. REVIEW METHODS: Selected studies were assessed and subjected to data extraction and quality assessment using standard methods. Where appropriate, meta-analysis was carried out. A mathematical model was constructed to estimate the cost-effectiveness of vitamin K1. RESULTS: The electronic literature searches identified 1078 potentially relevant articles. Of these, 14 articles relating to five trials that compared vitamin K with a relevant comparator in postmenopausal women with osteoporosis or osteopenia met the review inclusion criteria. The double-blind ECKO trial compared 5 mg of phylloquinone (vitamin K1) with placebo in Canadian women with osteopenia but without osteoporosis. Four open-label trials used 45 mg of menatetrenone (vitamin K2) in Japanese women with osteoporosis; the comparators were no treatment, etidronate or calcium. The methodological quality of the ECKO trial was good; however, all four menatetrenone trials were poorly reported and three were very small (n < 100 in each group). Phylloquinone was associated with a statistically significant reduction in the risk of clinical fractures relative to placebo [relative risk 0.46, 95% confidence interval (CI) 0.22 to 0.99]; morphometric vertebral fractures were not reported. The smaller menatetrenone trials found that menatetrenone was associated with a reduced risk of morphometric vertebral fractures relative to no treatment or calcium; however, the larger Osteoporosis Fracture (OF) study found no evidence of a reduction in vertebral fracture risk. The three smaller trials found no significant difference between treatment groups in non-vertebral fracture incidence. In the ECKO trial, phylloquinone was not associated with an increase in adverse events. In the menatetrenone trials, adverse event reporting was generally poor; however, in the OF study, menatetrenone was associated with a significantly higher incidence of skin and skin appendage lesions. No published economic evaluations of vitamin K were found and a mathematical model was thus constructed to estimate the cost-effectiveness of vitamin K1. Comparators were alendronate, risedronate and strontium ranelate. Vitamin K1 and alendronate were markedly more cost-effective than either risedronate or strontium ranelate. The base-case results favoured vitamin K1, but this relied on many assumptions, particularly on the efficacy of preventing hip and vertebral fractures. Calculation of the expected value of sampled information was conducted assuming a randomised controlled trial of 5 years' duration comparing alendronate with vitamin K1. The costs incurred in obtaining updated efficacy data from a trial with 2000 women per arm were estimated to be a cost-effective use of resources. CONCLUSIONS: There is currently large uncertainty over whether vitamin K1 is more cost-effective than alendronate; further research is required. It is unlikely that the present prescribing policy (i.e. alendronate as first-line treatment) would be altered.

Dabigatran etexilate for the prevention of venous thromboembolism in patients undergoing elective hip and knee surgery: a single technology appraisal.

This paper presents a summary of the evidence review group (ERG) report into the clinical effectiveness and cost-effectiveness of dabigatran etexilate (DBG) for the prevention of venous thromboembolism (VTE) in patients undergoing elective hip and knee surgery based upon a review of the manufacturer's submission to the NICE as part of the single technology appraisal (STA) process. The submission's evidence came from three reasonable-quality trials comparing DBG with enoxaparin, and a comparison of DBG with fondaparinux based on the relative efficacy and safety as derived from a mixed treatment comparison (MTC) meta-analysis. DBG (220 mg and 150 mg once daily) is not inferior to enoxaparin (40 mg once daily and 30 mg twice daily) in terms of major VTE or VTE-related events (secondary outcome). Meta-analysis shows that 220 mg DBG is not inferior to enoxaparin (40 mg once daily or 30 mg twice daily) in reducing total VTE and all-cause mortality (primary outcome) in total hip or knee replacement, whereas there is uncertainty around the clinical effectiveness of 150 mg DBG for this outcome. In the MTC analysis DBG compared favourably with the other interventions, with the exception of extended enoxaparin and fondaparinux. The adverse event profile was not significantly different in those receiving DBG and those receiving enoxaparin. The submitted two-phase economic model compares DBG with enoxaparin and fondaparinux in total hip and knee replacement. The model structure is appropriate and the model assumptions are reasonable. The health states, costs, utilities and recurrence rates used are considered to be appropriate for the required analysis. The model estimated that at the licensed dose of 220 mg once daily DBG dominates enoxaparin in both total hip replacement and total knee replacement and that at the lower dose of 150 mg once daily DBG dominates enoxaparin in total hip replacement and enoxaparin dominates DBG in total knee replacement. DBG is less cost-effective than fondaparinux in total hip replacement at both doses; the cost per quality-adjusted life-year of fondaparinux versus DBG is 11,111 pounds and 6857 pounds for the higher and lower doses of DBG respectively. In total knee replacement, both DBG doses are dominated by fondaparinux. For DBG versus all comparators in all cases the cost-effectiveness results are based on small incremental cost and health benefits. Weaknesses of the submitted evidence include that methods used for screening studies, data extraction and applying quality assessment criteria to included studies, as well as key details of trials included in the MTC, were not adequately described. In addition, some input parameters into the modelling process are incorrect. The ERG was unable to correct all of these mistakes and the impact on the model results is therefore unknown. The National Institute for Health and Clinical Excellence guidance issued as a result of the STA states that DBG is recommended as an option for the primary prevention of VTE events in adults who have undergone elective total hip or knee replacement surgery.

Spinal cord stimulation for chronic pain of neuropathic or ischaemic origin: systematic review and economic evaluation.

OBJECTIVES: This report addressed the question 'What is the clinical and cost-effectiveness of spinal cord stimulation (SCS) in the management of chronic neuropathic or ischaemic pain?' DATA SOURCES: Thirteen electronic databases [including MEDLINE (1950-2007), EMBASE (1980-2007) and the Cochrane Library (1991-2007)] were searched from inception; relevant journals were hand-searched; and appropriate websites for specific conditions causing chronic neuropathic/ischaemic pain were browsed. Literature searches were conducted from August 2007 to September 2007. REVIEW METHODS: A systematic review of the literature sought clinical and cost-effectiveness data for SCS in adults with chronic neuropathic or ischaemic pain with inadequate response to medical or surgical treatment other than SCS. Economic analyses were performed to model the cost-effectiveness and cost-utility of SCS in patients with neuropathic or ischaemic pain. RESULTS: From approximately 6000 citations identified, 11 randomised controlled trials (RCTs) were included in the clinical effectiveness review: three of neuropathic pain and eight of ischaemic pain. Trials were available for the neuropathic conditions failed back surgery syndrome (FBSS) and complex regional pain syndrome (CRPS) type I, and they suggested that SCS was more effective than conventional medical management (CMM) or reoperation in reducing pain. The ischaemic pain trials had small sample sizes, meaning that most may not have been adequately powered to detect clinically meaningful differences. Trial evidence failed to demonstrate that pain relief in critical limb ischaemia (CLI) was better for SCS than for CMM; however, it suggested that SCS was effective in delaying refractory angina pain onset during exercise at short-term follow-up, although not more so than coronary artery bypass grafting (CABG) for those patients eligible for that surgery. The results for the neuropathic pain model suggested that the cost-effectiveness estimates for SCS in patients with FBSS who had inadequate responses to medical or surgical treatment were below 20,000 pounds per quality-adjusted life-year (QALY) gained. In patients with CRPS who had had an inadequate response to medical treatment the incremental cost-effectiveness ratio (ICER) was 25,095 pounds per QALY gained. When the SCS device costs varied from 5000 pounds to 15,000 pounds, the ICERs ranged from 2563 pounds per QALY to 22,356 pounds per QALY for FBSS when compared with CMM and from 2283 pounds per QALY to 19,624 pounds per QALY for FBSS compared with reoperation. For CRPS the ICERs ranged from 9374 pounds per QALY to 66,646 pounds per QALY. If device longevity (1 to 14 years) and device average price (5000 pounds to 15,000 pounds) were varied simultaneously, ICERs were below or very close to 30,000 pounds per QALY when device longevity was 3 years and below or very close to 20,000 pounds per QALY when device longevity was 4 years. Sensitivity analyses were performed varying the costs of CMM, device longevity and average device cost, showing that ICERs for CRPS were higher. In the ischaemic model, it was difficult to determine whether SCS represented value for money when there was insufficient evidence to demonstrate its comparative efficacy. The threshold analysis suggested that the most favourable economic profiles for treatment with SCS were when compared to CABG in patients eligible for percutaneous coronary intervention (PCI), and in patients eligible for CABG and PCI. In these two cases, SCS dominated (it cost less and accrued more survival benefits) over CABG. CONCLUSIONS: The evidence suggested that SCS was effective in reducing the chronic neuropathic pain of FBSS and CRPS type I. For ischaemic pain, there may need to be selection criteria developed for CLI, and SCS may have clinical benefit for refractory angina short-term. Further trials of other types of neuropathic pain or subgroups of ischaemic pain, may be useful.

Thrombophilia testing in people with venous thromboembolism: systematic review and cost-effectiveness analysis.

OBJECTIVES: To assess whether thrombophilia testing following a venous thrombotic event is clinically effective and cost-effective in the management of thrombosis compared with no testing for thrombophilia. DATA SOURCES: Major electronic databases were searched from September to November 2006. REVIEW METHODS: A systematic review of the clinical effectiveness and cost-effectiveness literature was undertaken according to standard methods. A discrete event simulation model was constructed to assess the cost-effectiveness of changing the standard 3-month duration of warfarin treatment to 10 years, 20 years or lifelong. RESULTS: No clinical studies were identified that met the inclusion criteria for the systematic review. Further literature searches and clinical opinion were therefore used to inform the cost-effectiveness analysis. Thrombophilia testing in patients with pulmonary embolism (PE) had an estimated mean cost per quality-adjusted life-year (QALY) of below 20,000 pounds regardless of sex or age. In patients with a previous deep vein thrombosis (DVT), thrombophilia testing had an estimated mean cost per QALY of below 20,000 pounds in men aged 69 years or less and in women aged 49 years or less. The estimated duration of warfarin treatment (lifelong, 20 years, 10 years or no extended treatment) that was most cost-effective is presented for each age, sex, initial venous thromboembolism (VTE) event and type of thrombophilia. CONCLUSIONS: In terms of determining the duration of anticoagulation management, scenarios were found in which the cost per QALY of thrombophilia testing was below 20,000 pounds. However, these results are subject to great uncertainty, largely because of lack of knowledge about the increased risk of recurrence with each type of thrombophilia. Results are influenced by the fact that men have a greater risk of recurrence than women and by the fact that the frequency of adverse events associated with warfarin treatment increases with age. Further research, for example on the likely sensitivity and specificity of the tests for specific types of thrombophilia, is needed to reduce the uncertainty associated with these results. Studies comparing patients with VTE tested for thrombophilia with those whose risk assessment was based on personal and family history of thrombosis would also be beneficial.

Survival from childhood leukemia depending on socioeconomic status in Athens.

One hundred and twenty children first diagnosed as having acute leukemia between 1988 and 1992 in Athens, Greece, were followed until May 15, 1993. The socioeconomic status of the children's families was assessed by means of paternal occupation, paternal schooling, maternal schooling, ownership of a car, ability to choose a private medical facility and freedom in the choice of the attending physician. The analysis was done by proportional-hazards modelling, controlling for age and gender. All six socioeconomic indicators, alternatively evaluated, showed that fatality rates were higher in the lower socioeconomic groups, although nominal statistical significance was reached for only one of them. With respect to family ownership of a private car, the fatality rate ratio between children of families who own a car and children of families who do not was 0.29 with a 95% confidence interval of 0.13-0.62 (p = 0.002). These results suggest that in Greece, socially disadvantaged children have a less favorable survival from childhood leukemia.

Birth order sibship size and socio-economic factors in risk of schizophrenia in Greece.

A case-control study was undertaken to evaluate the effects, if any, of sibship size, birth order and parents' age at birth on the risk of a person's developing schizophrenia. Information was obtained, during an 18-month period, from 221 schizophrenic patients in the only mental-health hospital of the region of Peloponese, Greece, and from an equal number of matched patients. The data were analysed by modelling through logistic regression. No relationship was found between the occurrence of schizophrenia in a patient and either father's or mother's age at patient's birth, or sibship size, or birth order, when demographic and socio-economic variables were accounted for in the analysis. Although patients with schizophrenia were of similar socio-economic status to the comparison patients, they came from families with higher original socio-economic status, a fact which appears to reflect the decline of both schizophrenic and pre-schizophrenic people in the social ladder.