Comparative effectiveness of ZUMA-5 (axi-cel) vs SCHOLAR-5 external control in relapsed/refractory follicular lymphoma.

In the pivotal ZUMA-5 trial, axicabtagene ciloleucel (axi-cel; an autologous anti-CD19 chimeric antigen receptor T-cell therapy) demonstrated high rates of durable response in relapsed/refractory (r/r) follicular lymphoma (FL) patients. Here, outcomes from ZUMA-5 are compared with the international SCHOLAR-5 cohort, which applied key ZUMA-5 trial eligibility criteria simulating randomized controlled trial conditions. SCHOLAR-5 data were extracted from institutions in 5 countries, and from 1 historical clinical trial, for r/r FL patients who initiated a third or higher line of therapy after July 2014. Patient characteristics were balanced through propensity scoring on prespecified prognostic factors using standardized mortality ratio (SMR) weighting. Time-to-event outcomes were evaluated using weighted Kaplan-Meier analysis. Overall response rate (ORR) and complete response (CR) rate were compared using weighted odds ratios. The 143 ScHOLAR-5 patients reduced to an effective sample of 85 patients after SMR weighting vs 86 patients in ZUMA-5. Median follow-up time was 25.4 and 23.3 months for SCHOLAR-5 and ZUMA-5. Median overall survival (OS) and progression-free survival (PFS) in SCHOLAR-5 were 59.8 months and 12.7 months and not reached in ZUMA-5. Hazard ratios for OS and PFS were 0.42 (95% confidence interval [CI], 0.21-0.83) and 0.30 (95% CI, 0.18-0.49). The ORR and CR rate were 49.9% and 29.9% in SCHOLAR-5 and 94.2% and 79.1% in ZUMA-5, for odds ratios of 16.2 (95% CI, 5.6-46.9) and 8.9 (95% CI, 4.3-18.3). Compared with available therapies, axi-cel demonstrated an improvement in meaningful clinical endpoints, suggesting axi-cel addresses an important unmet need for r/r FL patients. This trial was registered at www.clinicaltrials.gov as #NCT03105336.

An attribution of value framework for combination treatments.

OBJECTIVE: The use of cost-effectiveness methods to support policy decisions has become well established but difficulties can arise when evaluating a new treatment which is indicated to be used in combination with an established "backbone treatment." If the latter has been priced close to the decision maker's willingness to pay threshold, this may mean that there is no headroom for the new treatment to demonstrate value, at any price, even if the combination is clinically effective. Without a mechanism for attributing value to component treatments within a combination therapy, the health system risks generating negative funding decisions for combinations of proven clinical benefit to patients. The aim of this work was to define a value attribution methodology which could be used to allocate value between the components of any combination treatment. METHODS: The framework is grounded in the standard decision rules of cost-effectiveness analysis and provides solutions according to key features of the problem: perfect/imperfect information about component treatment monotherapy effects and balanced/unbalanced market power between their manufacturers. RESULTS: The share of incremental value varies depending on whether there is perfect/imperfect information and balance/imbalance of market power, with some scenarios requiring the manufacturers to negotiate a share of the incremental value within a range defined by the framework. CONCLUSIONS: It is possible to define a framework that is independent of price and focuses on benefits expressed as Quality-Adjusted Life-Year (QALY) gains (and/or QALY equivalents for cost-savings), a standard metric used by many HTA agencies to evaluate novel treatments.

Treatment patterns and outcomes in relapsed/refractory follicular lymphoma: results from the international SCHOLAR-5 study.

The SCHOLAR-5 study examines treatment patterns and outcomes of real-world follicular lymphoma (FL) patients on 3rd line of treatment (LoT) or higher, for whom existing data are limited. SCHOLAR-5 is a retrospective cohort study using data from adults (≥ 18 years) with grade 1-3a FL, initiating ≥3rd LoT after June 2014 at major lymphoma centers in the US and Europe. Objective response rate (ORR), complete response (CR), progression-free survival (PFS) and overall survival (OS) were analyzed by LoT. Time-to-event outcomes were assessed using Kaplan-Meier methods. Of 128 patients, 87 initiated 3rd LoT, 63 initiated 4th LoT, and 47 initiated 5th LoT. At 1st eligible LoT, 31% progressed within 24-months of 1st LoT anti-CD20 combination therapy, 28% had prior autologous stem cell transplantation, and 31% were refractory to the previous LoT. The most common regimen in each LoT was chemoimmunotherapy; however, experimental drugs were increasingly used at later LoT. In the US, anti-CD20 monotherapy was more common at ≥3rd LoT compared to Europe, where stem cell transplants were more common. ORR at 3rd LoT was 68% (CR 44%), but decreased after each LoT to 37% (CR 22%) in ≥5 LoT. Median OS and PFS at 3rd LoT were 68 and 11 months, respectively, and reduced to 43 and 4 months at ≥5 LoT. Treatments were heterogenous at each LoT in both the US and Europe. Few FL patients achieved CR in later LoT, and duration of response and survival diminished with each subsequent line.

Matching-adjusted indirect comparison of bleeding outcomes in severe haemophilia A: Comparing valoctocogene roxaparvovec gene therapy, emicizumab prophylaxis, and FVIII replacement prophylaxis.

INTRODUCTION: Head-to-head evaluation of valoctocogene roxaparvovec, the first gene therapy approved for haemophilia A, with emicizumab is not available. Therefore, phase 3 trial data were indirectly compared. AIM: To compare bleeding rates in trials evaluating 6 × 1013  vg/kg valoctocogene roxaparvovec (GENEr8-1; NCT03370913), 1.5 mg/kg emicizumab dosed every week (HAVEN 3; NCT02847637), and FVIII prophylaxis (270-902) in participants with severe haemophilia A (FVIII ≤1 IU/dL). METHODS: Valoctocogene roxaparvovec versus emicizumab and FVIII prophylaxis as used in 270-902 versus emicizumab cross-trial comparisons were performed using matching-adjusted indirect comparison (MAIC). Individual participant data from GENEr8-1 and 270-902 were weighted to equalise aggregate participant baseline characteristics from HAVEN 3. After MAIC weighting, annualised bleeding rates (ABR) and proportions of participants without bleeds were compared for treated bleeds, all bleeds, treated joint bleeds, and treated spontaneous bleeds. RESULTS: After MAIC weighting, ABR for all bleeds was statistically significantly lower with valoctocogene roxaparvovec than emicizumab (rate ratio [95% CI], .55 [.33-.93]). Additionally, significantly higher proportions of participants had no treated joint bleeds (odds ratio [95% CI], 2.75 [1.20-6.31]) and no treated bleeds (3.25 [1.53-6.90]) with valoctocogene roxaparvovec versus emicizumab. When compared with the mainly standard half-life FVIII prophylaxis regimens in 270-902, mean ABRs (except for all bleeds) were significantly lower, and significantly higher proportions reported 0 bleeds for all outcomes with emicizumab. CONCLUSION: Valoctocogene roxaparvovec provided generally lower bleeding rates and higher probability of no bleeds, including treated joint bleeds, than emicizumab. Emicizumab was more effective than FVIII prophylaxis regimens used in 270-902.

Incorporating Prior Beliefs Into Meta-Analyses of Health-State Utility Values Using the Bayesian Power Prior.

OBJECTIVES: Health-state utility values (HSUVs) directly affect estimates of Quality-Adjusted Life-Years and thus the cost-utility estimates. In practice a single preferred value (SPV) is often selected for HSUVs, despite meta-analysis being an option when multiple (credible) HSUVs are available. Nevertheless, the SPV approach is often reasonable because meta-analysis implicitly considers all HSUVs as equally relevant. This article presents a method for the incorporation of weights to HSUV synthesis, allowing more relevant studies to have greater influence. METHODS: Using 4 case studies in lung cancer, hemodialysis, compensated liver cirrhosis, and diabetic retinopathy blindness, a Bayesian Power Prior (BPP) approach is used to incorporate beliefs on study applicability, reflecting the authors' perceived suitability for UK decision making. Older studies, non-UK value sets, and vignette studies are thus downweighted (but not disregarded). BPP HSUV estimates were compared with a SPV, random effects meta-analysis, and fixed effects meta-analysis. Sensitivity analyses were conducted iteratively updating the case studies, using alternative weighting methods, and simulated data. RESULTS: Across all case studies, SPVs did not accord with meta-analyzed values, and fixed effects meta-analysis produced unrealistically narrow CIs. Point estimates from random effects meta-analysis and BPP models were similar in the final models, although BPP reflected additional uncertainty as wider credible intervals, particularly when fewer studies were available. Differences in point estimates were seen in iterative updating, weighting approaches, and simulated data. CONCLUSIONS: The concept of the BPP can be adapted for synthesizing HSUVs, incorporating expert opinion on relevance. Because of the downweighting of studies, the BPP reflected structural uncertainty as wider credible intervals, with all forms of synthesis showing meaningful differences compared with SPVs. These differences would have implications for both cost-utility point estimates and probabilistic analyses.

Health Utility Analysis of Tepotinib in Patients With Non-Small Cell Lung Cancer Harboring MET Exon 14 Skipping.

OBJECTIVES: The VISION trial showed durable activity of tepotinib in MET exon 14 (METex14) skipping non-small cell lung cancer. We analyzed health state utilities using patient-reported outcomes from VISION. METHODS: 5-level version of EQ-5D (EQ-5D-5L) and European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 responses were collected at baseline, every 6 to 12 weeks during treatment, and at the end of treatment and safety follow-up. EQ-5D-5L and European Organisation for Research and Treatment of Cancer Quality of Life Utility Measure-Core 10 Dimensions (QLU-C10D) utilities were derived using United States, Canada, United Kingdom, and Taiwan value sets, where available. Utilities were analyzed with linear mixed models including covariates for progression or time-to-death (TTD). RESULTS: Utilities were derived for 273/291 patients (EQ-5D-5L, 1545 observations; QLU-C10D, 1546 observations). Mean (± SD) US EQ-5D-5L utilities increased after tepotinib initiation, from 0.687 ± 0.287 at baseline to 0.754 ± 0.250 before independently assessed progression, and decreased post progression (0.704 ± 0.288). US QLU-C10D utilities showed similar trends (0.705 ± 0.215, 0.753 ± 0.195, and 0.708 ± 0.209, respectively). Progression-based models demonstrated a statistically significant impact of progression on utilities and predicted higher utilities pre versus post progression. TTD-based models showed statistically significant associations of TTD with utilities and predicted declining utilities as TTD decreased. Prior treatment (yes/no) did not significantly predict utilities in progression- or TTD-based models. Utilities for Canada, United Kingdom, and Taiwan showed comparable trends. CONCLUSIONS: In this first analysis of health state utilities in patients with METex14 skipping non-small cell lung cancer, who received tepotinib, utilities were significantly associated with progression and TTD, but not prior treatment.

Comparative efficacy of ravulizumab and eculizumab in the treatment of atypical hemolytic uremic syndrome: An indirect comparison using clinical trial data.

Ravulizumab and eculizumab are approved terminal complement inhibitor treatments for atypical hemolytic uremic syndrome (aHUS). Ravulizumab was engineered from eculizumab to have an increased half-life allowing for reduced dosing frequency (8-weekly vs. 2-weekly). To account for differences in respective clinical trials, a validated balancing technique was used to enable an indirect comparison of ravulizumab and eculizumab treatment efficacy in aHUS. Patient-level data from four eculizumab clinical trials were available for pooling and comparison with data from two ravulizumab trials. In the primary analysis, adult native kidney data were compared. Propensity scores were calculated from baseline characteristics (dialysis status, estimated glomerular filtration rate, platelet count, serum lactate dehydrogenase). Stabilized inverse probability weighting was used to balance groups. Changes in outcomes from baseline to 26 weeks were compared between treatment groups. Sensitivity and subgroup analyses were conducted to assess the robustness of findings. Overall, 85 patients (46 ravulizumab, 39 eculizumab) were included in the primary analysis. Demographic and clinical characteristics were well balanced after weighting at baseline. At 26 weeks, clinical outcomes (including renal function, hematological markers, and dialysis prevalence), and fatigue and quality of life measures were improved with eculizumab and ravulizumab treatment. No differences between treatment groups reached statistical significance, although confidence intervals were wide. Sensitivity and subgroup analysis results were consistent with those of the primary analysis. Using appropriate methodology for indirect comparison of studies, no differences in outcomes were seen between ravulizumab and eculizumab, although, owing to small sample sizes, confidence intervals were wide.

The Effects of Model Misspecification in Unanchored Matching-Adjusted Indirect Comparison: Results of a Simulation Study.

OBJECTIVES: To assess the performance of unanchored matching-adjusted indirect comparison (MAIC) by matching on first moments or higher moments in a cross-study comparisons under a variety of conditions. A secondary objective was to gauge the performance of the method relative to propensity score weighting (PSW). METHODS: A simulation study was designed based on an oncology example, where MAIC was used to account for differences between a contemporary trial in which patients had more favorable characteristics and a historical control. A variety of scenarios were then tested varying the setup of the simulation study, including violating the implicit or explicit assumptions of MAIC. RESULTS: Under ideal conditions and under a variety of scenarios, MAIC performed well (shown by a low mean absolute error [MAE]) and was unbiased (shown by a mean error [ME] of about zero). The performance of the method deteriorated where the matched characteristics had low explanatory power or there was poor overlap between studies. Only when important characteristics are not included in the matching did the method become biased (nonzero ME). Where the method showed poor performance, this was exaggerated if matching was also performed on the variance (ie, higher moments). Relative to PSW, MAIC provided similar results in most circumstances, although it exhibited slightly higher MAE and a higher chance of exaggerating bias. CONCLUSIONS: MAIC appears well suited to adjust for cross-trial comparisons provided the assumptions underpinning the model are met, with relatively little efficiency loss compared with PSW.

Evaluation of survival extrapolation in immuno-oncology using multiple pre-planned data cuts: learnings to aid in model selection.

BACKGROUND: Due to limited duration of follow up in clinical trials of cancer treatments, estimates of lifetime survival benefits are typically derived using statistical extrapolation methods. To justify the method used, a range of approaches have been proposed including statistical goodness-of-fit tests and comparing estimates against a previous data cut (i.e. interim data collected). In this study, we extend these approaches by presenting a range of extrapolations fitted to four pre-planned data cuts from the JAVELIN Merkel 200 (JM200) trial. By comparing different estimates of survival and goodness-of-fit as JM200 data mature, we undertook an iterative process of fitting and re-fitting survival models to retrospectively identify early indications of likely long-term survival. METHODS: Standard and spline-based parametric models were fitted to overall survival data from each JM200 data cut. Goodness-of-fit was determined using an assessment of the estimated hazard function, information theory-based methods and objective comparisons of estimation accuracy. Best-fitting extrapolations were compared to establish which one provided the most accurate estimation, and how statistical goodness-of-fit differed. RESULTS: Spline-based models provided the closest fit to the final JM200 data cut, though all extrapolation methods based on the earliest data cut underestimated the 'true' long-term survival (difference in restricted mean survival time [RMST] at 36 months: - 1.1 to - 0.5 months). Goodness-of-fit scores illustrated that an increasingly flexible model was favored as data matured. Given an early data cut, a more flexible model better aligned with clinical expectations could be reasonably justified using a range of metrics, including RMST and goodness-of-fit scores (which were typically within a 2-point range of the statistically 'best-fitting' model). CONCLUSIONS: Survival estimates from the spline-based models are more aligned with clinical expectation and provided a better fit to the JM200 data, despite not exhibiting the definitively 'best' statistical goodness-of-fit. Longer-term data are required to further validate extrapolations, though this study illustrates the importance of clinical plausibility when selecting the most appropriate model. In addition, hazard-based plots and goodness-of-fit tests from multiple data cuts present useful approaches to identify when a more flexible model may be advantageous. TRIAL REGISTRATION: JAVELIN Merkel 200 was registered with ClinicalTrials.gov as NCT02155647 on June 4, 2014.

Summarising salient information on historical controls: A structured assessment of validity and comparability across studies.

BACKGROUND: While placebo-controlled randomised controlled trials remain the standard way to evaluate drugs for efficacy, historical data are used extensively across the development cycle. This ranges from supplementing contemporary data to increase the power of trials to cross-trial comparisons in estimating comparative efficacy. In many cases, these approaches are performed without in-depth review of the context of data, which may lead to bias and incorrect conclusions. METHODS: We discuss the original 'Pocock' criteria for the use of historical data and how the use of historical data has evolved over time. Based on these factors and personal experience, we created a series of questions that may be asked of historical data, prior to their use. Based on the answers to these questions, various statistical approaches are recommended. The strategy is illustrated with a case study in colorectal cancer. RESULTS: A number of areas need to be considered with historical data, which we split into three categories: outcome measurement, study/patient characteristics (including setting and inclusion/exclusion criteria), and disease process/intervention effects. Each of these areas may introduce issues if not appropriately handled, while some may preclude the use of historical data entirely. We present a tool (in the form of a table) for highlighting any such issues. Application of the tool to a colorectal cancer data set demonstrates under what conditions historical data could be used and what the limitations of such an analysis would be. CONCLUSION: Historical data can be a powerful tool to augment or compare with contemporary trial data, though caution is required. We present some of the issues that may be considered when involving historical data and what (if any) statistical approaches may account for differences between studies. We recommend that, where historical data are to be used in analyses, potential differences between studies are addressed explicitly.

Real-world evidence use in assessments of cancer drugs by NICE.

OBJECTIVE: To establish how real-world evidence (RWE) has been used to inform single technology appraisals (STAs) of cancer drugs conducted by the National Institute for Health and Care Excellence (NICE). METHODS: STAs published by NICE from April 2011 to October 2018 that evaluated cancer treatments were reviewed. Information regarding the use of RWE to directly inform the company-submitted cost-effectiveness analysis was extracted and categorized by topic. Summary statistics were used to describe emergent themes, and a narrative summary was provided for key case studies. RESULTS: Materials for a total of 113 relevant STAs were identified and analyzed, of which nearly all (96 percent) included some form of RWE within the company-submitted cost-effectiveness analysis. The most common categories of RWE use concerned the health-related quality of life of patients (71 percent), costs (46 percent), and medical resource utilization (40 percent). While sources of RWE were routinely criticized as part of the appraisal process, we identified only two cases where the use of RWE was overtly rejected; hence, in the majority of cases, RWE was accepted in cancer drug submissions to NICE. DISCUSSION: RWE has been used extensively in cancer submissions to NICE. Key criticisms of RWE in submissions to NICE are seldom regarding the use of RWE in general; instead, these are typically concerned with specific data sources and the applicability of these to the decision problem. Within an appropriate context, RWE constitutes an extremely valuable source of information to inform decision making; yet the development of best practice guidelines may improve current reporting standards.

Migraine day frequency in migraine prevention: longitudinal modelling approaches.

BACKGROUND: Health economic models are critical tools to inform reimbursement agencies on health care interventions. Many clinical trials report outcomes using the frequency of an event over a set period of time, for example, the primary efficacy outcome in most clinical trials of migraine prevention is mean change in the frequency of migraine days (MDs) per 28 days (monthly MDs [MMD]) relative to baseline for active treatment versus placebo. Using these cohort-level endpoints in economic models, accounting for variation among patients is challenging. In this analysis, parametric models of change in MMD for migraine preventives were assessed using data from erenumab clinical studies. METHODS: MMD observations from the double-blind phases of two studies of erenumab were used: one in episodic migraine (EM) (NCT02456740) and one in chronic migraine (CM) (NCT02066415). For each trial, two longitudinal regression models were fitted: negative binomial and beta binomial. For a thorough comparison we also present the fitting from the standard multilevel Poisson and the zero inflated negative binomial. RESULTS: Using the erenumab study data, both the negative binomial and beta-binomial models provided unbiased estimates relative to observed trial data with well-fitting distribution at various time points. CONCLUSIONS: This proposed methodology, which has not been previously applied in migraine, has shown that these models may be suitable for estimating MMD frequency. Modelling MMD using negative binomial and beta-binomial distributions can be advantageous because these models can capture intra- and inter-patient variability so that trial observations can be modelled parametrically for the purposes of economic evaluation of migraine prevention. Such models have implications for use in a wide range of disease areas when assessing repeated measured utility values.

Longitudinal assessment of utilities in patients with migraine: an analysis of erenumab randomized controlled trials.

BACKGROUND: Cost-effectiveness analyses in patients with migraine require estimates of patients' utility values and how these relate to monthly migraine days (MMDs). This analysis examined four different modelling approaches to assess utility values as a function of MMDs. METHODS: Disease-specific patient-reported outcomes from three erenumab clinical studies (two in episodic migraine [NCT02456740 and NCT02483585] and one in chronic migraine [NCT02066415]) were mapped to the 5-dimension EuroQol questionnaire (EQ-5D) as a function of the Migraine-Specific Quality of Life Questionnaire (MSQ) and the Headache Impact Test (HIT-6™) using published algorithms. The mapped utility values were used to estimate generic, preference-based utility values suitable for use in economic models. Four models were assessed to explain utility values as a function of MMDs: a linear mixed effects model with restricted maximum likelihood (REML), a fractional response model with logit link, a fractional response model with probit link and a beta regression model. RESULTS: All models tested showed very similar fittings. Root mean squared errors were similar in the four models assessed (0.115, 0.114, 0.114 and 0.114, for the linear mixed effect model with REML, fractional response model with logit link, fractional response model with probit link and beta regression model respectively), when mapped from MSQ. Mean absolute errors for the four models tested were also similar when mapped from MSQ (0.085, 0.086, 0.085 and 0.085) and HIT-6 and (0.087, 0.088, 0.088 and 0.089) for the linear mixed effect model with REML, fractional response model with logit link, fractional response model with probit link and beta regression model, respectively. CONCLUSIONS: This analysis describes the assessment of longitudinal approaches in modelling utility values and the four models proposed fitted the observed data well. Mapped utility values for patients treated with erenumab were generally higher than those for individuals treated with placebo with equivalent number of MMDs. Linking patient utility values to MMDs allows utility estimates for different levels of MMD to be predicted, for use in economic evaluations of preventive therapies. TRIAL REGISTRATION: ClinicalTrials.gov numbers of the trials used in this study: STRIVE, NCT02456740 (registered May 14, 2015), ARISE, NCT02483585 (registered June 12, 2015) and NCT02066415 (registered Feb 17, 2014).

Frequentist and Bayesian meta-regression of health state utilities for multiple myeloma incorporating systematic review and analysis of individual patient data.

This analysis presents the results of a systematic review for health state utilities in multiple myeloma, as well as analysis of over 9,000 observations taken from registry and trial data. The 27 values identified from 13 papers are then synthesised in a frequentist nonparametric bootstrap model and a Bayesian meta-regression. Results were similar between the frequentist and Bayesian models with low utility on disease diagnosis (approximately 0.55), raising to approximately 0.65 on first line treatment and declining slightly with each subsequent line. Stem cell transplant was also found to be a significant predictor of health-related quality of life in both individual patient data and meta-regression, with an increased utility of approximately 0.06 across different models. The work presented demonstrates the feasibility of Bayesian methods for utility meta-regression, whilst also presenting an internally consistent set of data from the analysis of registry data. To facilitate easy updating of the data and model, data extraction tables and model code are provided as Data S1. The main limitations of the model relate to the low number of studies available, particularly in highly pretreated patients.

Estimating outcomes and cost effectiveness using a single-arm clinical trial: ofatumumab for double-refractory chronic lymphocytic leukemia.

BACKGROUND: Ofatumumab (Arzerra®, Novartis) is a treatment for chronic lymphocytic leukemia refractory to fludarabine and alemtuzumab [double refractory (DR-CLL)]. Ofatumumab was licensed on the basis of an uncontrolled Phase II study, Hx-CD20-406, in which patients receiving ofatumumab survived for a median of 13.9 months. However, the lack of an internal control arm presents an obstacle for the estimation of comparative effectiveness. METHODS: The objective of the study was to present a method to estimate the cost effectiveness of ofatumumab in the treatment of DR-CLL. As no suitable historical control was available for modelling, the outcomes from non-responders to ofatumumab were used to model the effect of best supportive care (BSC). This was done via a Cox regression to control for differences in baseline characteristics between groups. This analysis was included in a partitioned survival model built in Microsoft® Excel with utilities and costs taken from published sources, with costs and quality-adjusted life years (QALYs) were discounted at a rate of 3.5% per annum. RESULTS: Using the outcomes seen in non-responders, ofatumumab is expected to add approximately 0.62 life years (1.50 vs. 0.88). Using published utility values this translates to an additional 0.30 QALYs (0.77 vs. 0.47). At the list price, ofatumumab had a cost per QALY of £130,563, and a cost per life year of £63,542. The model was sensitive to changes in assumptions regarding overall survival estimates and utility values. CONCLUSIONS: This study demonstrates the potential of using data for non-responders to model outcomes for BSC in cost-effectiveness evaluations based on single-arm trials. Further research is needed on the estimation of comparative effectiveness using uncontrolled clinical studies.

The Cost of Costing Treatments Incorrectly: Errors in the Application of Drug Prices in Economic Evaluation Due to Failing to Account for the Distribution of Patient Weight.

BACKGROUND: The cost of pharmaceuticals dosed by weight or body surface area (BSA) can be estimated in several ways for economic evaluations. A review of 20 recent National Institute for Health and Care Excellence appraisals showed that 17 of them took the mean weight or BSA of patients, 2 costed the individual patient data from trials, and 2 fitted a distribution to patient-level data. OBJECTIVES: To investigate the estimated drug costs using different methodologies to account for patient characteristics for pharmaceuticals with a weight- or BSA-based posology. The secondary objective was to explore the suitability of general population data as a proxy for patient-level data. METHODS: Patient-level data were pooled from three clinical trials and used to calculate a hypothetical cost per administration of eight licensed pharmaceuticals, applying the three methods used in recent National Institute for Health and Care Excellence appraisals. The same analysis was performed using data from the Health Survey for England (in place of patient-level data) to investigate the validity of using general population data as a substitute for patient-level data. RESULTS: Compared with using patient-level data from clinical trials, the mean patient characteristics (weight or BSA) led to an underestimation of drug cost by 6.1% (range +1.5% to -25.5%). Fitting a distribution to patient-level data led to a mean difference of +0.04%. All estimates were consistent using general population data. CONCLUSIONS: Estimation of drug costs in health economic evaluation should account for the distribution in weight or BSA to produce accurate results. When patient data are not available, general population data may be used as an alternative.

Patient-reported utilities in advanced or metastatic melanoma, including analysis of utilities by time to death.

BACKGROUND: Health-related quality of life is often collected in clinical studies, and forms a cornerstone of economic evaluation. This study had two objectives, firstly to report and compare pre- and post-progression health state utilities in advanced melanoma when valued by different methods and secondly to explore the validity of progression-based health state utility modelling compared to modelling based upon time to death. METHODS: Utilities were generated from the ipilimumab MDX010-20 trial (Clinicaltrials.gov Identifier: NCT00094653) using the condition-specific EORTC QLQ-C30 (via the EORTC-8D) and generic SF-36v2 (via the SF-6D) preference-based measures. Analyses by progression status and time to death were conducted on the patient-level data from the MDX010-20 trial using generalised estimating equations fitted in Stata®, and the predictive abilities of the two approaches compared. RESULTS: Mean utility showed a decrease on disease progression in both the EORTC-8D (0.813 to 0.776) and the SF-6D (0.648 to 0.626). Whilst higher utilities were obtained using the EORTC-8D, the relative decrease in utility on progression was similar between measures. When analysed by time to death, both EORTC-8D and SF-6D showed a large decrease in utility in the 180 days prior to death (from 0.831 to 0.653 and from 0.667 to 0.544, respectively). Compared to progression status alone, the use of time to death gave similar or better estimates of the original data when used to predict patient utility in the MDX010-20 study. Including both progression status and time to death further improved model fit. Utilities seen in MDX010-20 were also broadly comparable with those seen in the literature. CONCLUSIONS: Patient-level utility data should be analysed prior to constructing economic models, as analysis solely by progression status may not capture all predictive factors of patient utility and time to death may, as death approaches, be as or more important. Additionally this study adds to the body of evidence showing that different scales lead to different health state values. Further research is needed on how different utility instruments (the SF-6D, EORTC-8D and EQ-5D) relate to each other in different disease areas.

Modelling Health State Utilities as a Transformation of Time to Death in Patients with Non-Small Cell Lung Cancer.

BACKGROUND: When utilities are analyzed by time to death (TTD), this has historically been implemented by 'grouping' observations as discrete time periods to create health state utilities. We extended the approach to use continuous functions, avoiding assumptions around groupings. The resulting models were used to test the concept with data from different regions and different country tariffs. METHODS: Five-year follow-up in advanced non-small cell lung cancer (NSCLC) was used to fit six continuous TTD models using generalized estimating equations, which were compared with progression-based utilities and previously published TTD groupings. Sensitivity analyses were performed using only patients with a confirmed death, the last year of life only, and artificially censoring data at 24 months. The statistically best-fitting model was then applied to data subsets by region and different EQ-5D-3L country tariffs. RESULTS: Continuous (natural) [Formula: see text] and [Formula: see text] models outperformed other continuous models, grouped TTD, and progression-based models in statistical fit (mean absolute error and Quasi Information Criterion). This held through sensitivity and scenario analyses. The pattern of reduced utility as a patient approaches death was consistent across regions and EQ-5D tariffs using the preferred [Formula: see text] model. CONCLUSIONS: The use of continuous models provides a statistically better fit than TTD groupings, without the need for strong assumptions about the health states experienced by patients. Where a TTD approach is merited for use in modelling, continuous functions should be considered, with the scope for further improvements in statistical fit by both widening the number of candidate models tested and the therapeutic areas investigated.

An Evaluation of an Algorithm for the Selection of Flexible Survival Models for Cancer Immunotherapies: Pass or Fail?

BACKGROUND AND OBJECTIVE: Accurately extrapolating survival beyond trial follow-up is essential in a health technology assessment where model choice often substantially impacts estimates of clinical and cost effectiveness. Evidence suggests standard parametric models often provide poor fits to long-term data from immuno-oncology trials. Palmer et al. developed an algorithm to aid the selection of more flexible survival models for these interventions. We assess the usability of the algorithm, identify areas for improvement and evaluate whether it effectively identifies models capable of accurate extrapolation. METHODS: We applied the Palmer algorithm to the CheckMate-649 trial, which investigated nivolumab plus chemotherapy versus chemotherapy alone in patients with gastroesophageal adenocarcinoma. We evaluated the algorithm's performance by comparing survival estimates from identified models using the 12-month data cut to survival observed in the 48-month data cut. RESULTS: The Palmer algorithm offers a systematic procedure for model selection, encouraging detailed analyses and ensuring that crucial stages in the selection process are not overlooked. In our study, a range of models were identified as potentially appropriate for extrapolating survival, but only flexible parametric non-mixture cure models provided extrapolations that were plausible and accurately predicted subsequently observed survival. The algorithm could be improved with minor additions around the specification of hazard plots and setting out plausibility criteria. CONCLUSIONS: The Palmer algorithm provides a systematic framework for identifying suitable survival models, and for defining plausibility criteria for extrapolation validity. Using the algorithm ensures that model selection is based on explicit justification and evidence, which could reduce discordance in health technology appraisals.