Beth Woods

Network meta-analysis on the log-hazard scale, combining count and hazard ratio statistics accounting for multi-arm trials: A tutorial

BackgroundData on survival endpoints are usually summarised using either hazard ratio, cumulative number of events, or median survival statistics. Network meta-analysis, an extension of traditional pairwise meta-analysis, is typically based on a single statistic. In this case, studies which do not report the chosen statistic are excluded from the analysis which may introduce bias.MethodsIn this paper we present a tutorial illustrating how network meta-analyses of survival endpoints can combine count and hazard ratio statistics in a single analysis on the hazard ratio scale. We also describe methods for accounting for the correlations in relative treatment effects (such as hazard ratios) that arise in trials with more than two arms. Combination of count and hazard ratio data in a single analysis is achieved by estimating the cumulative hazard for each trial arm reporting count data. Correlation in relative treatment effects in multi-arm trials is preserved by converting the relative treatment effect estimates (the hazard ratios) to arm-specific outcomes (hazards).ResultsA worked example of an analysis of mortality data in chronic obstructive pulmonary disease (COPD) is used to illustrate the methods. The data set and WinBUGS code for fixed and random effects models are provided.ConclusionsBy incorporating all data presentations in a single analysis, we avoid the potential selection bias associated with conducting an analysis for a single statistic and the potential difficulties of interpretation, misleading results and loss of available treatment comparisons associated with conducting separate analyses for different summary statistics.

Individual patient data network meta-analysis of mortality effects of implantable cardiac devices

Objective Implantable cardioverter defibrillators (ICD), cardiac resynchronisation therapy pacemakers (CRT-P) and the combination therapy (CRT-D) have been shown to reduce all-cause mortality compared with medical therapy alone in patients with heart failure and reduced EF. Our aim was to synthesise data from major randomised controlled trials to estimate the comparative mortality effects of these devices and how these vary according to patients’ characteristics. Methods Data from 13 randomised trials (12 638 patients) were provided by medical technology companies. Individual patient data were synthesised using network meta-analysis. Results Unadjusted analyses found CRT-D to be the most effective treatment (reduction in rate of death vs medical therapy: 42% (95% credible interval: 32–50%), followed by ICD (29% (20–37%)) and CRT-P (28% (15–40%)). CRT-D reduced mortality compared with CRT-P (19% (1–33%)) and ICD (18% (7–28%)). QRS duration, left bundle branch block (LBBB) morphology, age and gender were included as predictors of benefit in the final adjusted model. In this model, CRT-D reduced mortality in all subgroups (range: 53% (34–66%) to 28% (−1% to 49%)). Patients with QRS duration ≥150 ms, LBBB morphology and female gender benefited more from CRT-P and CRT-D. Men and those <60 years benefited more from ICD. Conclusions These data provide estimates for the mortality benefits of device therapy conditional upon multiple patient characteristics. They can be used to estimate an individual patients expected relative benefit and thus inform shared decision making. Clinical guidelines should discuss age and gender as predictors of device benefits.

No Study Left Behind: A Network Meta-Analysis in Non-Small-Cell Lung Cancer Demonstrating the Importance of Considering All Relevant Data

OBJECTIVE To demonstrate the importance of considering all relevant indirect data in a network meta-analysis of treatments for non-small-cell lung cancer (NSCLC). METHODS A recent National Institute for Health and Clinical Excellence appraisal focussed on the indirect comparison of docetaxel with erlotinib in second-line treatment of NSCLC based on trials including a common comparator. We compared the results of this analysis to a network meta-analysis including other trials that formed a network of evidence. We also examined the importance of allowing for the correlations between the estimated treatment effects that can arise when analysing such networks. RESULTS The analysis of the restricted network including only trials of docetaxel and erlotinib linked via the common placebo comparator produced an estimated mean hazard ratio (HR) for erlotinib compared with docetaxel of 1.55 (95% confidence interval [CI] 0.72-2.97). In contrast, the network meta-analysis produced an estimated HR for erlotinib compared with docetaxel of 0.83 (95% CI 0.65-1.06). Analyzing the wider network improved the precision of estimated treatment effects, altered their rankings and also allowed further treatments to be compared. Some of the estimated treatment effects from the wider network were highly correlated. CONCLUSIONS This empirical example shows the importance of considering all potentially relevant data when comparing treatments. Care should therefore be taken to consider all relevant information, including correlations induced by the network of trial data, when comparing treatments.

How far do you go? Efficient searching for indirect evidence.

Background. Indirect evidence is particularly valuable in health care decision making when direct trial evidence comparing relevant treatments is absent or limited. Current approaches using a predetermined set of comparators in the search query may fail to identify all relevant indirect evidence. Purpose. To present a framework for the efficient design of search strategies for identifying clinical trials providing indirect evidence for a treatment comparison. Findings. The authors present 2 search strategies that differ from traditional search strategies in using a series of iterative searches to identify the set of relevant comparators. In both, the comparators included in each search are determined by the results of previous searches. For a given number of searches, the strategies presented will find all indirect comparisons that include a certain number of comparators linking the treatments of interest. Methods of estimating the value of indirect evidence via a given number of comparators linking the treatments of interest are presented, thus allowing the burden of additional searching to be traded off against the likely impact of finding more distant comparisons. A practical illustration of the search strategies in the context of informing a network meta-analysis of second-line treatments for non-small cell lung cancer is presented. Conclusions. The iterative strategies presented offer a means of identifying such evidence and allow the researcher to determine the optimal scope of the search by estimating the value of additional indirect evidence.

Comparative efficacy of tyrosine kinase inhibitor treatments in the third-line setting, for chronic-phase chronic myelogenous leukemia after failure of second-generation tyrosine kinase inhibitors

We compared the efficacy of ponatinib and second-generation tyrosine kinase inhibitors (2G-TKIs: bosutinib, dasatinib, and nilotinib) in chronic phase CML resistant/intolerant to ≥1 prior 2G-TKI. Estimated probabilities of CCyR with 2G-TKI ranged from 22% to 26%, compared with 60% (95% CrI 52-68%) with ponatinib. The estimated probability of ponatinib providing higher response rate than all other included treatments was 99% (CCyR) and 97% (MCyR). Use of further 2G-TKI may provide limited benefit in CP-CML patients resistant/intolerant to prior 2G-TKI treatment. Compared with 2G-TKIs, ponatinib is estimated to provide substantially higher probability of achieving CCyR and MCyR; safety was not compared.

Bendamustine Versus Chlorambucil for the First-Line Treatment of Chronic Lymphocytic Leukemia in England and Wales: A Cost-Utility Analysis

OBJECTIVES To evaluate the cost-effectiveness of bendamustine compared with chlorambucil as first-line treatment for patients with chronic lymphocytic leukemia who would be considered unsuitable for treatment with fludarabine combination chemotherapy regimens. METHODS A semi-Markov approach was used to estimate time in each health state. The model was parameterized primarily by using data from a phase III randomized, open-label trial comparing bendamustine with chlorambucil. It captured the increased progression-free survival and improved response rates with bendamustine, and the cost and quality of life impacts of postprogression treatments. The analysis was conducted from the perspective of the National Health Service in England and Wales. A lifetime (35-year) time horizon was used. Deterministic sensitivity analyses, probabilistic sensitivity analyses, and subgroup analyses in older patients and patients with poor performance status were carried out. RESULTS The estimated incremental cost-effectiveness ratio was £ 11,960 per quality-adjusted life-year. None of the deterministic sensitivity analyses increased the incremental cost-effectiveness ratio by more than £ 2000. Subgroup analyses showed that bendamustine remained cost-effective across different patient groups. Probabilistic sensitivity analysis showed that at the £ 20,000 threshold, bendamustine has a 90% probability of being cost-effective. CONCLUSIONS Bendamustine represents good value for first-line treatment of patients with chronic lymphocytic leukemia who are unsuitable for treatment with fludarabine combination chemotherapy. The incremental cost-effectiveness ratio is below the thresholds commonly applied in England and Wales (£ 20,000-£ 30,000 per quality-adjusted life-year).

Raltitrexed plus cisplatin is cost-effective compared with pemetrexed plus cisplatin in patients with malignant pleural mesothelioma

INTRODUCTION The National Institute for Health and Clinical Excellence (NICE) has previously recommended pemetrexed plus cisplatin for the treatment of patients with advanced malignant pleural mesothelioma (MPM) and WHO performance status 0-1. Subsequent to this appraisal, randomised controlled trial (RCT) data for raltitrexed plus cisplatin and comparing chemotherapy to active symptom control (ASC) has become available, allowing a more complete analysis of the comparative efficacy and cost-effectiveness of first-line chemotherapy in MPM. METHODS An adjusted indirect comparison is used to estimate the relative efficacy of raltitrexed plus cisplatin and pemetrexed plus cisplatin. A cost-effectiveness model is used to assess the lifetime costs and health outcomes associated with these comparators and ASC. Patient level data from the EORTC 08983 trial are used to estimate baseline progression and survival rates. Relative treatment effects are taken from RCTs; cost and utility data from the literature. RESULTS Raltitrexed plus cisplatin and pemetrexed plus cisplatin were not found to be statistically significantly different with respect to overall response, progression free survival or overall survival. The cost-effectiveness analysis found raltitrexed plus cisplatin to be cost-effective at a cost per quality adjusted life year of £13,454 compared to cisplatin and £27,360 compared to ASC. Pemetrexed plus cisplatin is dominated by raltitrexed plus cisplatin as the raltitrexed combination offers marginally higher quality adjusted life years (QALYs) and life years (LYs) at a substantially lower total cost. CONCLUSION Raltitrexed plus cisplatin is a cost-effective first-line treatment for MPM. This conclusion was maintained across a number of sensitivity analyses.

Prioritizing Pharmacogenetic Research: A Value of Information Analysis of CYP2D6 Testing to Guide Breast Cancer Treatment

OBJECTIVES To demonstrate how value of information (VOI) analysis can be used to establish research priorities regarding the use of pharmacogenetic tests using CYP2D6 testing to select adjuvant hormonal therapy in early stage breast cancer as a case study. METHODS The following four treatment pathways are compared in a Markov model: tamoxifen treatment; CYP2D6 test and treat homozygous and heterozygous wild type patients (wt/wt; wt/*4) with tamoxifen and *4/*4 patients with anastrozole (HetTam); CYP2D6 test and treat homozygous wild type patients with tamoxifen and others with anastrozole (HomTam); and anastrozole treatment. Pharmacogenetic testing efficacy is estimated by synthesizing randomized controlled trial data comparing tamoxifen to anastrozole with observational data linking CYP2D6 genotype to tamoxifen outcomes. RESULTS In order of increasing effectiveness the comparators are tamoxifen, HetTam, HomTam, anastrozole. Health outcomes for test and treatment strategies are highly uncertain. Differences in comparator costs depend on assumptions made regarding anastrozole patent expiry. The expected value of a decision taken with perfect information is £69 to £106 million (pound sterling) for the United Kingdom depending on patent expiry assumptions and the acceptable cost-effectiveness threshold. The most valuable research (VOI £53-£82 million) elucidates the relationship between CYP2D6 genotype and tamoxifen effectiveness. It is uncertain whether values of other research designs would exceed their costs. CONCLUSIONS Retrospective analysis of one of the large adjuvant aromatase inhibitor trials is warranted to better understand any association between CYP2D6 genotype and tamoxifen outcomes. VOI approaches may be helpful for prioritising evidence needs and structuring coverage with evidence development agreements for pharmacogenetics.

Cost-effectiveness of adjunct non-pharmacological interventions for osteoarthritis of the knee

Background There is limited information on the costs and benefits of alternative adjunct non-pharmacological treatments for knee osteoarthritis and little guidance on which should be prioritised for commissioning within the NHS. This study estimates the costs and benefits of acupuncture, braces, heat treatment, insoles, interferential therapy, laser/light therapy, manual therapy, neuromuscular electrical stimulation, pulsed electrical stimulation, pulsed electromagnetic fields, static magnets and transcutaneous electrical nerve Stimulation (TENS), based on all relevant data, to facilitate a more complete assessment of value. Methods Data from 88 randomised controlled trials including 7,507 patients were obtained from a systematic review. The studies reported a wide range of outcomes. These were converted into EQ-5D index values using prediction models, and synthesised using network meta-analysis. Analyses were conducted including firstly all trials and secondly only trials with low risk of selection bias. Resource use was estimated from trials, expert opinion and the literature. A decision analytic model synthesised all evidence to assess interventions over a typical treatment period (constant benefit over eight weeks or linear increase in effect over weeks zero to eight and dissipation over weeks eight to 16). Results When all trials are considered, TENS is cost-effective at thresholds of £20–30,000 per QALY with an incremental cost-effectiveness ratio of £2,690 per QALY vs. usual care. When trials with a low risk of selection bias are considered, acupuncture is cost-effective with an incremental cost-effectiveness ratio of £13,502 per QALY vs. TENS. The results of the analysis were sensitive to varying the intensity, with which interventions were delivered, and the magnitude and duration of intervention effects on EQ-5D. Conclusions Using the £20,000 per QALY NICE threshold results in TENS being cost-effective if all trials are considered. If only higher quality trials are considered, acupuncture is cost-effective at this threshold, and thresholds down to £14,000 per QALY.

Bendamustine-rituximab: a cost-utility analysis in first-line treatment of indolent non-Hodgkin’s lymphoma in England and Wales

Abstract Objective: To evaluate the cost-effectiveness of bendamustine-rituximab (B-R) compared with CHOP-R (cyclophosphamide, doxorubicin, vincristine, prednisone, rituximab) and CVP-R (cyclophosphamide, vincristine, prednisone, rituximab) as first-line treatment for patients with advanced indolent non-Hodgkin’s lymphoma (NHL). Methods: A patient-level simulation was adapted from the model used by the University of Sheffield School of Health and Related Research (ScHARR) in a health technology appraisal of rituximab for first-line treatment of follicular lymphoma. This approach allowed modelling of the complex treatment pathways in indolent NHL. Data from a Phase 3 randomized, open-label trial were used to compare B-R with CHOP-R. The relative efficacy of CHOP-R and CVP-R was estimated using an indirect treatment comparison similar to the original ScHARR approach. The analysis was conducted from the perspective of the National Health Service in England and Wales, using a lifetime time horizon. A number of one-way sensitivity and scenario analyses were conducted, including one using recently published data comparing CVP-R with CHOP-R. Results: The deterministic incremental cost-effectiveness ratio (ICER) was £5249 per quality adjusted life year (QALY) for B-R vs CHOP-R, and £8092 per QALY for B-R vs CVP-R. The alternative scenario using direct data comparing CVP-R with CHOP-R approximately halved the ICER for B-R vs CVP-R to £4733. Owing to its better toxicity profile, B-R reduced the cost of treating adverse events by over £1000 per patient vs CHOP-R. Limitations: The main limitations were: immaturity of overall survival data from the Phase 3 trial; reliance on quality-of-life data from previous health technology appraisals (as this was not collected in the trial); and a lack of direct evidence or a network of connected evidence comparing B-R with CVP-R. Conclusions: The ICERs for B-R vs CHOP-R and CVP-R were considerably below the thresholds normally regarded as cost-effective in England and Wales (£20,000–30,000 per QALY).