Implementing multi-level network meta-regression for time-to-event outcomes: A case study in relapsed refractory multiple myeloma

OBJECTIVES

Multi-level network meta-regression (ML-NMR) leverages individual patient data (IPD) and aggregate data (AD) from a network of randomized controlled trials (RCTs) to assess the comparative efficacy of multiple treatments, while adjusting for between-study differences. We provide an overview of ML-NMR for time-to-event outcomes and apply it to an illustrative case study, including example R code.

METHODS

The case study evaluated the comparative efficacy of idecabtagene vicleucel (ide-cel), selinexor+dexamethasone (Sd), belantamab mafodotin (BM), and conventional care (CC) for patients with triple class exposed relapsed/refractory multiple myeloma in terms of overall survival (OS). Single-arm clinical trials and real-world data were naively combined to create an AD artificial RCT (aRCT) (MAMMOTH-CC versus DREAMM-2-BM verus STORM-2-Sd) and an IPD aRCT (KarMMa-ide-cel versus KarMMa-RW-CC). With some assumptions, we incorporated continuous covariates with skewed distributions, reported as median and range. The ML-NMR models adjusted for number of prior lines, triple-class refractory (TCR) status, and age and were compared via the leave-one-out information criterion (LOOIC). We summarized predicted hazard ratios and survival (95% credible intervals) in the IPD aRCT population.

RESULTS

The Weibull ML-NMR model had the lowest LOOIC. Ide-cel was more efficacious than Sd, BM, and CC in terms of OS. Effect modifiers had minimal impact on the model and only TCR was a prognostic factor.

CONCLUSIONS

We demonstrate an application of ML-NMR for time-to-event outcomes and introduce code that can be used to aid implementation. Given its benefits, we encourage practitioners to utilize ML-NMR when population adjustment is necessary for comparisons of multiple treatments.

Unanchored Population-Adjusted Indirect Comparison Methods for Time-to-Event Outcomes Using Inverse Odds Weighting, Regression Adjustment, and Doubly Robust Methods With Either Individual Patient or Aggregate Data

OBJECTIVES

Several methods for unanchored population-adjusted indirect comparisons (PAICs) are available. Exploring alternative adjustment methods, depending on the available individual patient data (IPD) and the aggregate data (AD) in the external study, may help minimize bias in unanchored indirect comparisons. However, methods for time-to-event outcomes are not well understood. This study provides an overview and comparison of methods using a case study to increase familiarity. A recent method is applied to marginalize conditional hazard ratios, which allows for the comparisons of methods, and a doubly robust method is proposed.

METHODS

The following PAIC methods were compared through a case study in third-line small cell lung cancer, comparing nivolumab with standard of care based on a single-arm phase II trial (CheckMate 032) and real-world study (Flatiron) in terms of overall survival: IPD-IPD analyses using inverse odds weighting, regression adjustment, and a doubly robust method; IPD-AD analyses using matching-adjusted indirect comparison, simulated treatment comparison, and a doubly robust method.

RESULTS

Nivolumab extended survival versus standard of care with hazard ratios ranging from 0.63 (95% CI 0.44-0.90) in naive comparisons (identical estimates for IPD-IPD and IPD-AD analyses) to 0.69 (95% CI 0.44-0.98) in the IPD-IPD analyses using regression adjustment. Regression-based and doubly robust estimates yielded slightly wider confidence intervals versus the propensity score-based analyses.

CONCLUSIONS

The proposed doubly robust approach for time-to-event outcomes may help to minimize bias due to model misspecification. However, all methods for unanchored PAIC rely on the strong assumption that all prognostic covariates have been included.

The Health Inequality Impact of a New Cancer Therapy Given Treatment and Disease Characteristics

OBJECTIVES

This study aimed to perform a simulation study to quantify the health inequality impact of a cancer therapy given cancer and treatment characteristics using the distributional cost-effectiveness framework.

METHODS

The following factors were varied in 10 000 simulations: lifetime risk of the disease, median overall survival (OS) with standard of care (SOC), difference in OS between non-Hispanic (NH)-Black and NH-White patients (prognostic effect), treatment effect of the new therapy relative to SOC, whether the treatment effect differs between NH-Black and NH-White patients (effect modification), health utility, drug costs, and preprogression and postprogression costs. Based on these characteristics, the incremental population net health benefits were calculated for the new therapy and applied to a US distribution of quality-adjusted life expectancy at birth. The health inequality impact was quantified as the difference in the degree of inequality in the "post-new therapy" versus "pre-new therapy" quality-adjusted life expectancy distributions.

RESULTS

For cancer types characterized by relatively large lifetime risk, large median OS with SOC, large treatment effect, and large effect modification, the direction of the impact of the new therapy on inequality is easy to predict. When effect modification is minor or absent, which is a realistic scenario, the direction of the inequality impact is difficult to predict. Larger incremental drug costs have a worsening effect on health inequality.

CONCLUSIONS

The findings provide a guide to help decision makers and other stakeholders make an initial assessment whether a new therapy with known treatment effects for a specific tumor type can have a positive or negative health inequality impact.

The Health Inequality Impact of Liquid Biopsy to Inform First-Line Treatment of Advanced Non-Small Cell Lung Cancer: A Distributional Cost-Effectiveness Analysis

OBJECTIVES

To perform a distributional cost-effectiveness analysis of liquid biopsy (LB) followed by, if needed, tissue biopsy (TB) (LB-first strategy) relative to a TB-only strategy to inform first-line treatment of advanced non-small cell lung cancer (aNSCLC) from a US payer perspective by which we quantify the impact of LB-first on population health inequality according to race and ethnicity.

METHODS

With a health economic model, quality-adjusted life-years (QALYs) and costs per patient were estimated for each subgroup. Given the lifetime risk of aNSCLC, and assuming equally distributed opportunity costs, the incremental net health benefits of LB-first were calculated, which were used to estimate general population quality-adjusted life expectancy at birth (QALE) by race and ethnicity with and without LB-first. The degree of QALYs and QALE differences with the strategies was expressed with inequality indices. Their differences were defined as the inequality impact of LB-first.

RESULTS

LB-first resulted in an additional 0.21 (95% uncertainty interval: 0.07-0.39) QALYs among treated patients, with the greatest gain observed among Asian patients (0.31 QALYs [0.09-0.61]). LB-first resulted in an increase in relative inequality in QALYs among patients, but a minor decrease in relative inequality in QALE.

CONCLUSIONS

LB-first to inform first-line aNSCLC therapy can improve health outcomes. With current diagnostic performance, the benefit is the greatest among Asian patients, thereby potentially widening racial and ethnic differences in survival among patients with aNSCLC. Assuming equally distributed opportunity costs and access, LB-first does not worsen and, in fact, may reduce inequality in general population health according to race and ethnicity.

Clinical validity and utility of circulating tumor DNA (ctDNA) testing in advanced non-small cell lung cancer (aNSCLC): a systematic literature review and meta-analysis

Purpose

Circulating tumor DNA (ctDNA) testing has become a promising tool to guide first-line (1L) targeted treatment for advanced non-small cell lung cancer (aNSCLC). This study aims to estimate the clinical validity (CV) and clinical utility (CU) of ctDNA-based next-generation sequencing (NGS) for oncogenic driver mutations to inform 1L treatment decisions in aNSCLC through a systematic literature review and meta-analysis.

Methods

A systematic literature search was conducted in PubMed/MEDLINE and Embase to identify randomized control trials or observational studies reporting CV/CU on ctDNA testing in patients with aNSCLC. Meta-analyses were performed using bivariate random-effects models to estimate pooled sensitivity and specificity. Progression-free/overall survival (PFS/OS) was summarized for CU studies.

Results

Eighteen studies were identified: 17 CV only, 2 CU only, and 1 both. Thirteen studies were included for the meta-analysis on multi-gene detection. The overall sensitivity and specificity for ctDNA detection of any mutation were 0.69 (95% CI, 0.63-0.74) and 0.99 (95% CI, 0.97-1.00) respectively. However, sensitivity varied greatly by driver gene, ranging from 0.29 (95% CI, 0.13-0.53) for ROS 1 to 0.77 (95% CI, 0.63-0.86) for KRAS . Two studies compared PFS with ctDNA versus tissue-based testing followed by 1L targeted therapy found no significant differences. One study reported OS curves on ctDNA-matched and tissue-matched therapies but no hazard ratios were provided.

Conclusion

ctDNA testing demonstrated an overall acceptable diagnostic accuracy in aNSCLC patients, however, sensitivity varied greatly by driver mutation. Further research is needed, especially for uncommon driver mutations, to better understand the CU of ctDNA testing in guiding targeted treatments for aNSCLC.

Multi-state network meta-analysis of progression and survival data

Multiple randomized controlled trials, each comparing a subset of competing interventions, can be synthesized by means of a network meta-analysis to estimate relative treatment effects between all interventions in the evidence base. Here we focus on estimating relative treatment effects for time-to-event outcomes. Cancer treatment effectiveness is frequently quantified by analyzing overall survival (OS) and progression-free survival (PFS). We introduce a method for the joint network meta-analysis of PFS and OS that is based on a time-inhomogeneous tri-state (stable, progression, and death) Markov model where time-varying transition rates and relative treatment effects are modeled with parametric survival functions or fractional polynomials. The data needed to run these analyses can be extracted directly from published survival curves. We demonstrate use by applying the methodology to a network of trials for the treatment of non-small-cell lung cancer. The proposed approach allows the joint synthesis of OS and PFS, relaxes the proportional hazards assumption, extends to a network of more than two treatments, and simplifies the parameterization of decision and cost-effectiveness analyses.

Indirect treatment comparison of lurbinectedin versus other second-line treatments for small-cell lung cancer

Aim: Compare lurbinectedin versus other second-line (2L) small-cell lung cancer (SCLC) treatments. Methods: An unanchored matching-adjusted indirect comparison connected the platinum-sensitive SCLC cohort of a single-arm lurbinectedin trial to a network of three randomized controlled trials (oral and intravenous [IV] topotecan, and platinum re-challenge) identified by systematic literature review. Network meta-analysis methods estimated relative treatment effects. Results: In platinum-sensitive patients, lurbinectedin demonstrated a survival benefit and favorable safety profile versus oral and IV topotecan and platinum re-challenge (overall survival, hazard ratio [HR]: 0.43; 95% credible interval [CrI]: 0.27, 0.67; HR: 0.43; 95% CrI: 0.26, 0.70; HR: 0.42; 95% CrI: 0.30, 0.58 respectively). Conclusion: Lurbinectedin showed a robust survival benefit and favorable safety versus other SCLC treatments in 2L platinum-sensitive SCLC.

A Comparison of Alternative Network Meta-Analysis Methods in the Presence of Nonproportional Hazards: A Case Study in First-Line Advanced or Metastatic Renal Cell Carcinoma

OBJECTIVES

Network meta-analysis (NMA) of time-to-event outcomes based on constant hazard ratios can result in biased findings when the proportional hazards (PHs) assumption does not hold in a subset of trials. We aimed to summarize the published non-PH NMA methods for time-to-event outcomes, demonstrate their application, and compare their results.

METHODS

The following non-PH NMA methods were compared through an illustrative case study in oncology of 4 randomized controlled trials in terms of progression-free survival and overall survival: (1) 1-step or (2) 2-step multivariate NMAs based on traditional survival distributions or fractional polynomials, (3) NMAs with restricted cubic splines for baseline hazard, and (4) restricted mean survival NMA.

RESULTS

For progression-free survival, the PH assumption did not hold across trials and non-PH NMA methods better reflected the relative treatment effects over time. The most flexible models (fractional polynomials and restricted cubic splines) fit better to the data than the other approaches. Estimated hazard ratios obtained with different non-PH NMA methods were similar at 5 years of follow-up but differed thereafter in the extrapolations. Although there was no strong evidence of PH violation for overall survival, non-PH NMA methods captured this uncertainty in the relative treatment effects over time.

CONCLUSIONS

When the PH assumption is questionable in a subset of the randomized controlled trials, we recommend assessing alternative non-PH NMA methods to estimate relative treatment effects for time-to-event outcomes. We propose a transparent and explicit stepwise model selection process considering model fit, external constraints, and clinical validity. Given inherent uncertainty, sensitivity analyses are suggested.

Matching-adjusted indirect comparison of axi-cel and liso-cel in relapsed or refractory large B-cell lymphoma

In the absence of a randomized head-to-head trial, an unanchored matching-adjusted indirect comparison was performed to estimate the relative treatment effects of axicabtagene ciloleucel (axi-cel; ZUMA-1) versus lisocabtagene maraleucel (liso-cel; TRANSCEND-NHL-001) for treatment of relapsed/refractory (R/R) large B-cell lymphoma (LBCL) after at least two lines of therapy. After matching, axi-cel and liso-cel had comparable objective response rates and duration. Compared to liso-cel, axi-cel was associated with improvements in overall survival (hazard ratio [HR]: 0.53 [95% CI: 0.34-0.82]) and progression-free survival (HR: 0.61 [95% CI: 0.40-0.92]). Axi-cel was associated with a higher rate of grade ≥3 cytokine release syndrome (odds ratio [OR]: 3.64 [95% CI: 1.04-12.76]) and neurological events (OR: 3.45 [95% CI: 1.65-7.19]), with smaller differences estimated in scenario analyses including ZUMA-1 safety management cohorts. Results suggest axi-cel improved survival compared to liso-cel but with increased odds of specific adverse events.

Network Meta-analysis on Disconnected Evidence Networks When Only Aggregate Data Are Available: Modified Methods to Include Disconnected Trials and Single-Arm Studies while Minimizing Bias

BACKGROUND

Network meta-analysis (NMA) requires a connected network of randomized controlled trials (RCTs) and cannot include single-arm studies. Regulators or academics often have only aggregate data. Two aggregate data methods for analyzing disconnected networks are random effects on baseline and aggregate-level matching (ALM). ALM has been used only for single-arm studies, and both methods may bias effect estimates.

METHODS

We modified random effects on baseline to separate RCTs connected to and disconnected from the reference and any single-arm studies, minimizing the introduction of bias. We term our modified method reference prediction. We similarly modified ALM and extended it to include RCTs disconnected from the reference. We tested these methods using constructed data and a simulation study.

RESULTS

In simulations, bias for connected treatments for ALM ranged from -0.0158 to 0.051 and for reference prediction from -0.0107 to 0.083. These were low compared with the true mean effect of 0.5. Coverage ranged from 0.92 to 1.00. In disconnected treatments, bias of ALM ranged from -0.16 to 0.392 and of reference prediction from -0.102 to 0.40, whereas coverage of ALM ranged from 0.30 to 0.82 and of reference prediction from 0.64 to 0.94. Under fixed study effects for disconnected evidence, bias was similar, but coverage was 0.81 to 1.00 for reference prediction and 0.18 to 0.76 for ALM. Trends of similar bias but greater coverage for reference prediction with random study effects were repeated in constructed data.

CONCLUSIONS

Both methods with random study effects seem to minimize bias in treatment connected to the reference. They can estimate treatment effects for disconnected treatments but may be biased. Reference prediction has greater coverage and may be recommended overall.

HIGHLIGHTS

Two methods were modified for network meta-analysis on disconnected networks and for including single-arm observational or interventional studies in network meta-analysis using only aggregate data and for minimizing the bias of effect estimates for treatments only in trials connected to the reference.Reference prediction was developed as a modification of random effects on baseline that keeps analyses of trials connected to the reference separately from those disconnected from the reference and from single-arm studies. The method was further modified to account for correlation in trials with more than 2 arms and, under random study effects, to estimate variance in heterogeneity separately in connected and disconnected evidence.Aggregate-level matching was extended to include trials disconnected from the reference, rather than only single-arm studies. The method was further modified to separately estimate treatment effects and heterogeneity variance in the connected and disconnected evidence and to account for the correlation between arms in trials with more than 2 arms.Performance was assessed using a constructed data example and simulation study.The methods were found to have similar, and sometimes low, bias when estimating the relative effects for disconnected treatments, but reference prediction with random study effects had the greatest coverage.The use of reference prediction with random study effects for disconnected networks is recommended if no individual patient data or alternative real-world evidence is available.

Assessments of the Value of New Interventions Should Include Health Equity Impact

A formal evaluation of the health equity impact of a new intervention is hardly ever performed as part of a health technology assessment to understand its value. This should change, in our view. An evidence-based quantitative assessment of the health equity impact can help decision makers develop coverage policies, programme designs, and quality initiatives focused on optimizing both total health and health equity given the treatment options available. We outline the conceptual basis of how a new intervention can impact health equity and adopt distributional cost-effectiveness analysis based on decision-analytic models to assess this quantitatively, using a newly US FDA-approved drug for Alzheimer's disease (aducanumab) as an example. We argue that gaps in the evidence base for the new intervention, for example, due to limited clinical research participation among racial and ethnic minority groups, do not preclude such an evaluation. Understanding these uncertainties has implications for fair pricing, decision making, and future research. If we are serious about population-level decision making that not only is focused on improving total health but also aims to improve health equity, we should consider routinely assessing the health equity impact of new interventions.

Comparative effectiveness and safety of pharmaceuticals assessed in observational studies compared with randomized controlled trials

BACKGROUND

There have been ongoing efforts to understand when and how data from observational studies can be applied to clinical and regulatory decision making. The objective of this review was to assess the comparability of relative treatment effects of pharmaceuticals from observational studies and randomized controlled trials (RCTs).

METHODS

We searched PubMed and Embase for systematic literature reviews published between January 1, 1990, and January 31, 2020, that reported relative treatment effects of pharmaceuticals from both observational studies and RCTs. We extracted pooled relative effect estimates from observational studies and RCTs for each outcome, intervention-comparator, or indication assessed in the reviews. We calculated the ratio of the relative effect estimate from observational studies over that from RCTs, along with the corresponding 95% confidence interval (CI) for each pair of pooled RCT and observational study estimates, and we evaluated the consistency in relative treatment effects.

RESULTS

Thirty systematic reviews across 7 therapeutic areas were identified from the literature. We analyzed 74 pairs of pooled relative effect estimates from RCTs and observational studies from 29 reviews. There was no statistically significant difference (based on the 95% CI) in relative effect estimates between RCTs and observational studies in 79.7% of pairs. There was an extreme difference (ratio < 0.7 or > 1.43) in 43.2% of pairs, and, in 17.6% of pairs, there was a significant difference and the estimates pointed in opposite directions.

CONCLUSIONS

Overall, our review shows that while there is no significant difference in the relative risk ratios between the majority of RCTs and observational studies compared, there is significant variation in about 20% of comparisons. The source of this variation should be the subject of further inquiry to elucidate how much of the variation is due to differences in patient populations versus biased estimates arising from issues with study design or analytical/statistical methods.

Matching adjusted indirect comparisons of efficacy outcomes for idecabtagene vicleucel (ide-cel, bb2121) versus selinexor + dexamethasone and belantamab mafodotin in relapsed and refractory multiple myeloma

Idecabtagene vicleucel (ide-cel, bb2121), a chimeric antigen receptor (CAR) T cell therapy, has been investigated in patients with relapsed and refractory multiple myeloma (RRMM) who have received an immunomodulatory drug, proteasome inhibitor, and anti-CD38 antibody in the single-arm phase 2 KarMMa clinical trial. Two therapies with distinct mechanisms of action - selinexor plus dexamethasone (Sd) and belantamab mafodotin (BM) - are currently approved in the United States for heavily pretreated patients, including those who are triple-class refractory. To compare ide-cel versus Sd and ide-cel versus BM, matching-adjusted indirect comparisons were performed. Ide-cel extended progression-free survival (PFS) and overall survival (OS) versus both Sd and BM (hazard ratio (HR); 95% confidence interval (CI)). PFS: ide-cel versus Sd, 0.46; 0.28-0.75; ide-cel versus BM, 0.45; 0.27-0.77. OS: ide-cel versus Sd, 0.23; 0.13-0.42; ide-cel versus BM, 0.35; 0.14-0.87. These results suggest ide-cel offers clinically meaningful improvements over currently approved regimens for patients with heavily pretreated RRMM.

Complicated legacies: The human genome at 20

Millions of people today have access to their personal genomic information. Direct-to-consumer services and integration with other “big data” increasingly commoditize what was rightly celebrated as a singular achievement in February 2001 when the first draft human genomes were published. But such remarkable technical and scientific progress has not been without its share of missteps and growing pains. Science invited the experts below to help explore how we got here and where we should (or ought not) be going.

Aligning text mining and machine learning algorithms with best practices for study selection in systematic literature reviews

BACKGROUND

Despite existing research on text mining and machine learning for title and abstract screening, the role of machine learning within systematic literature reviews (SLRs) for health technology assessment (HTA) remains unclear given lack of extensive testing and of guidance from HTA agencies. We sought to address two knowledge gaps: to extend ML algorithms to provide a reason for exclusion-to align with current practices-and to determine optimal parameter settings for feature-set generation and ML algorithms.

METHODS

We used abstract and full-text selection data from five large SLRs (n = 3089 to 12,769 abstracts) across a variety of disease areas. Each SLR was split into training and test sets. We developed a multi-step algorithm to categorize each citation into the following categories: included; excluded for each PICOS criterion; or unclassified. We used a bag-of-words approach for feature-set generation and compared machine learning algorithms using support vector machines (SVMs), naïve Bayes (NB), and bagged classification and regression trees (CART) for classification. We also compared alternative training set strategies: using full data versus downsampling (i.e., reducing excludes to balance includes/excludes because machine learning algorithms perform better with balanced data), and using inclusion/exclusion decisions from abstract versus full-text screening. Performance comparisons were in terms of specificity, sensitivity, accuracy, and matching the reason for exclusion.

RESULTS

The best-fitting model (optimized sensitivity and specificity) was based on the SVM algorithm using training data based on full-text decisions, downsampling, and excluding words occurring fewer than five times. The sensitivity and specificity of this model ranged from 94 to 100%, and 54 to 89%, respectively, across the five SLRs. On average, 75% of excluded citations were excluded with a reason and 83% of these citations matched the reviewers' original reason for exclusion. Sensitivity significantly improved when both downsampling and abstract decisions were used.

CONCLUSIONS

ML algorithms can improve the efficiency of the SLR process and the proposed algorithms could reduce the workload of a second reviewer by identifying exclusions with a relevant PICOS reason, thus aligning with HTA guidance. Downsampling can be used to improve study selection, and improvements using full-text exclusions have implications for a learn-as-you-go approach.

Comparative effectiveness of nivolumab versus standard of care for third-line patients with small-cell lung cancer

Aim: To estimate the comparative effectiveness of nivolumab versus standard of care (SOC) in terms of overall survival (OS) for small-cell lung cancer patients treated with two prior lines of chemotherapy, in other words, third line in the USA. Materials & methods: Data were from CheckMate 032, a single-arm trial of nivolumab, and real-world electronic patient records. Comparisons of OS were conducted using three different methods to adjust for differences (regression, weighting and doubly robust) between the populations. Results: Nivolumab was associated with longer survival compared with SOC (hazard ratio for OS: 0.58-0.70) across all methods for adjustment. Conclusion: Nivolumab was more efficacious in terms of OS as third-line treatment for small-cell lung cancer compared with current SOC in the USA.

The Effect of Dose Escalation on the Cost-Effectiveness of Etanercept and Adalimumab with Methotrexate Among Patients with Moderate to Severe Rheumatoid Arthritis

BACKGROUND

Patients with moderate to severe rheumatoid arthritis (RA) occasionally increase their doses of tumor necrosis factor (TNF) inhibitors, especially the monoclonal antibody origin drugs such as adalimumab and infliximab, after inadequate response to the initial dose. Previous studies have evaluated the cost-effectiveness of various sequences of treatment for RA in the United States but have not considered the effect of dose escalation.

OBJECTIVE

To assess the cost-effectiveness of etanercept and adalimumab by incorporating the effect of dose escalation in moderate to severe RA patients.

METHODS

We adapted the open-source Innovation and Value Initiative - Rheumatoid Arthritis model, version 1.0 to separately simulate the magnitude and time to dose escalation among RA patients taking adalimumab plus methotrexate or etanercept plus methotrexate from a societal perspective and lifetime horizon. An important assumption in the model was that dose escalation would increase treatment costs through its effect on the number of doses but would have no effect on effectiveness. We estimated the dose escalation parameters using the IBM MarketScan Commercial and Medicare Supplemental Databases. We fit competing parametric survival models to model time to dose escalation and used model diagnostics to compare the fit of the competing models. We measured the magnitude of dose escalation as the percentage increase in the number of doses conditional on dose escalation. Finally, we used the parameterized model to simulate treatment sequences beginning with a TNF inhibitor (adalimumab, etanercept) followed by nonbiologic treatment.

RESULTS

In baseline models without dose escalation, the incremental cost per quality-adjusted life-year of the etanercept treatment sequence relative to the adalimumab treatment sequence was $85,593. Incorporating dose escalation increased treatment costs for each sequence, but costs increased more with adalimumab, lowering the incremental cost-effectiveness ratio to $9,001. At willingness-to-pay levels of $100,000, the etanercept sequence was more cost-effective compared with the adalimumab sequence, with probability 0.55 and 0.85 in models with and without dose escalation, respectively.

CONCLUSIONS

Dose escalation has important effects on cost-effectiveness and should be considered when comparing biologic medications for the treatment of RA.

DISCLOSURES

Funding for this study was contributed by Amgen. When this work was conducted, Incerti and Jansen were employees of Precision Health Economics, which received financial support from Amgen. Maksabedian Hernandez, Collier, Gharaibeh, and Stolshek were employees and stockholders of Amgen, and Tkacz and Moore-Schiltz were employees of IBM Watson Health, which received financial support from Amgen. Some of the results of this work were previously presented as a poster at the 2019 AMCP Managed Care & Specialty Pharmacy Annual Meeting, March 25-28, 2019, in San Diego, CA.

Agalsidase beta treatment slows estimated glomerular filtration rate loss in classic Fabry disease patients: results from an individual patient data meta-analysis

Background

Fabry disease is a rare, X-linked genetic disorder that, if untreated in patients with the Classic phenotype, often progresses to end-stage kidney disease. This meta-analysis determined the effect of agalsidase beta on loss of estimated glomerular filtration rate (eGFR) in the Classic phenotype using an expansive evidence base of individual patient-level data.

Methods

The evidence base included four Sanofi-Genzyme studies and six studies from a systematic literature review. These were restricted to Classic Fabry patients meeting the eligibility criteria from Phases III and IV agalsidase beta trials, including 315 patients (161 treated). Linear regression was first used to model annual change in eGFR for each patient and the resulting annualized eGFR slopes were modelled with treatment and covariates using quantile regression. These results were then used to estimate median annualized eGFR change in agalsidase beta treated versus untreated groups.

Results

Imbalances across treatment groups were found in baseline age, sex and proteinuria, but not in the use of renin–angiotensin system blockers. The adjusted model suggests that treated (agalsidase beta) patients experienced a slower median eGFR decrease [2.46 mL/min/1.73 m²/year slower; 95% confidence interval (CI) 0.63–4.29; P = 0.0087] than comparable untreated patients. The median eGFR decrease was 2.64 mL/min/1.73 m²/year slower (95% CI 0.53–4.78; P = 0.0141) in treated Classic males.

Conclusions

Using an expansive evidence base and robust modelling approach, these data indicate that agalsidase beta-treated patients with the Classic phenotype conserve their renal function better than untreated patients.

The economic burden of kidney graft failure in the United States

Despite improvements in outcomes for kidney transplant recipients in the past decade, graft failure continues to impose substantial burden on patients. However, the population-wide economic burden of graft failure has not been quantified. This study aims to fill that gap by comparing outcomes from a simulation model of kidney transplant patients in which patients are at risk for graft failure with an alternative simulation in which the risk of graft failure is assumed to be zero. Transitions through the model were estimated using Scientific Registry of Transplant Recipients data from 1987 to 2017. We estimated lifetime costs, overall survival, and quality-adjusted life-years (QALYs) for both scenarios and calculated the difference between them to obtain the burden of graft failure. We find that for the average patient, graft failure will impose additional medical costs of $78 079 (95% confidence interval [CI] $41 074, $112 409) and a loss of 1.66 QALYs (95% CI 1.15, 2.18). Given 17 644 kidney transplants in 2017, the total incremental lifetime medical costs associated with graft failure is $1.38B (95% CI $725M, $1.98B) and the total QALY loss is 29 289 (95% CI 20 291, 38 464). Efforts to reduce the incidence of graft failure or to mitigate its impact are urgently needed.

Developing Open-Source Models for the US Health System: Practical Experiences and Challenges to Date with the Open-Source Value Project

The Innovation and Value Initiative started the Open-Source Value Project with the aim to improve the credibility and relevance of model-based value assessment in the context of the US healthcare environment. As a core activity of the Open-Source Value Project, the Innovation and Value Initiative develops and provides access to flexible open-source economic models that are developed iteratively based on public feedback and input. In this article, we describe our experience to date with the development of two currently released, Open-Source Value Project models, one in rheumatoid arthritis and one in epidermal growth factor receptor-positive non-small-cell lung cancer. We developed both Open-Source Value Project models using the statistical programming language R instead of spreadsheet software (i.e., Excel), which allows the models to capture multiple model structures, model sequential treatment with individual patient simulations, and improve integration with formal evidence synthesis. By developing the models in R, we were also able to use version control systems to manage changes to the source code, which is needed for iterative and collaborative model development. Similarly, Open-Source Value Project models are freely available to the public to provide maximum transparency and facilitate collaboration. Development of the rheumatoid arthritis and non-small-cell lung cancer model platforms has presented multiple challenges. The development of multiple components of the model platform tailored to different audiences, including web interfaces, required more resources than a cost-effectiveness analysis for a publication would. Furthermore, we faced methodological hurdles, in particular related to the incorporation of multiple competing model structures and novel elements of value. The iterative development based on public feedback also posed some challenges during the review phase, where methodological experts did not always understand feedback from clinicians and vice versa. Response to the Open-Source Value Project by the modeling community and patient organizations has been positive, but feedback from US decision makers has been limited to date. As we progress with this project, we hope to learn more about the feasibility, benefits, and challenges of an open-source and collaborative approach to model development for value assessment.

Integrating expert opinion with clinical trial data to extrapolate long-term survival: a case study of CAR-T therapy for children and young adults with relapsed or refractory acute lymphoblastic leukemia

BACKGROUND

Long-term clinical outcomes are necessary to assess the cost-effectiveness of new treatments over a lifetime horizon. Without long-term clinical trial data, current practice to extrapolate survival beyond the trial period involves fitting alternative parametric models to the observed survival. Choosing the most appropriate model is based on how well each model fits to the observed data. Supplementing trial data with feedback from experts may improve the plausibility of survival extrapolations. We demonstrate the feasibility of formally integrating long-term survival estimates from experts with empirical clinical trial data to provide more credible extrapolated survival curves.

METHODS

The case study involved relapsed or refractory B-cell pediatric and young adult acute lymphoblastic leukemia (r/r pALL) regarding long-term survival for tisagenlecleucel (chimeric antigen receptor T-cell [CAR-T]) with evidence from the phase II ELIANA trial. Seven pediatric oncologists and hematologists experienced with CAR-T therapies were recruited. Relevant evidence regarding r/r pALL and tisagenlecleucel provided a common basis for expert judgments. Survival rates and related uncertainty at 2, 3, 4, and 5 years were elicited from experts using a web-based application adapted from Sheffield Elicitation Framework. Estimates from each expert were combined with observed data using time-to-event parametric models that accounted for experts' uncertainty, producing an overall distribution of survival over time. These results were validated based on longer term follow-up (median duration 24.2 months) from ELIANA following the elicitation.

RESULTS

Extrapolated survival curves based on ELIANA trial without expert information were highly uncertain, differing substantially depending on the model choice. Survival estimates between 2 to 5 years from individual experts varied with a fair amount of uncertainty. However, incorporating expert estimates improved the precision in the extrapolated survival curves. Predictions from a Gompertz model, which experts believed was most appropriate, suggested that more than half of the ELIANA patients treated with tisagenlecleucel will survive up to 5 years. Expert estimates at 24 months were validated by longer follow-up.

CONCLUSIONS

This study provides an example of how expert opinion can be elicited and synthesized with observed survival data using a transparent and formal procedure, capturing expert uncertainty, and ensuring projected long-term survival is clinically plausible.

Indirect comparison of nivolumab ± ipilimumab (CheckMate 032) versus other treatments for recurrent small-cell lung cancer

Aim: To estimate the comparative efficacy of nivolumab ± ipilimumab versus alternative treatments for small-cell lung cancer after at least one prior line of chemotherapy. Materials & m ethods: A systematic literature review identified six randomized controlled trials (RCTs) that could be connected in a network. The Kaplan-Meier survival curves from these RCTs were synthesized using network meta-analysis models. Aggregate-level matching was used to connect CheckMate 032 to the RCTs. Results: CheckMate 032 was connected to the network by Amrubicin Clinical Trial-1. Nivolumab ± ipilimumab had a more durable tumor response and more favorable long-term survival versus topotecan via intravenous and versus amrubicin. Conclusion: Compared with chemotherapies for recurrent small-cell lung cancer, nivolumab ± ipilimumab improves response duration, which may translate to long-term survival benefits.

Forced vital capacity and cross-domain late-onset Pompe disease outcomes: an individual patient-level data meta-analysis

Background

Late-onset Pompe disease (LOPD) is a rare, metabolic disease primarily affecting the musculoskeletal and respiratory systems. Forced vital capacity (FVC) is commonly used to measure pulmonary function; however, associations between FVC and other LOPD outcomes remain unclear.

Methods

A systematic literature review was conducted on November 2015, updated September 2016 and supplemented with clinical trial data from the sponsor. Outcomes included: 6-min walk test distance (6MWT), FVC, maximal inspiratory/expiratory pressure (MIP/MEP), Medical Research Council-skeletal muscle strength score (MRC), 36-item short-form survey-physical component score (SF-36), Rotterdam Handicap Scale (RHS), Fatigue Severity Scale (FSS) and survival. Individual patient data meta-analysis was used for cross-sectional analyses and longitudinal analyses to determine associations between percent of predicted FVC and LOPD measures and outcomes.

Results

Fifteen studies were selected. From cross-sectional analyses, FVC and MRC were most strongly associated. Specifically, patients with 10% higher FVC (a round number for illustrative purposes only) were associated with a 4.72% (95% confidence interval [CI]: 3.37, 6.07) higher MRC score, indicating a positive association. Similarly, slopes for the 6MWT and SF-36 relative to a 10% higher FVC were estimated at 33.2 meters (95% CI 24.0, 42.4) and 1.2% (95% CI 0.24, 2.16%), respectively. From longitudinal analyses, a 10% incremental increase in predicted FVC was associated with an average increase of 4.12% in MRC score (95% CI 1.29, 6.95), 35.6 m in the 6MWT (95% CI 19.9, 51.6), and 1.34% in SF-36 (95% CI 0.08, 2.60). There was insufficient data to conduct analyses for RHS, FSS and survival.

Conclusions

FVC is positively associated with LOPD measures and outcomes across multiple domains. Additionally, longitudinal changes in FVC are positively associated with changes in the 6MWT, MRC and SF-36.

Electronic supplementary material

The online version of this article (10.1007/s00415-019-09401-1) contains supplementary material, which is available to authorized users.

A Flexible Open-Source Decision Model for Value Assessment of Biologic Treatment for Rheumatoid Arthritis

OBJECTIVE

The nature of model-based cost-effectiveness analysis can lead to disputes in the scientific community. We propose an iterative and collaborative approach to model development by presenting a flexible open-source simulation model for rheumatoid arthritis (RA), accessible to both technical and non-technical end-users.

METHODS

The RA model is a discrete-time individual patient simulation with 6-month cycles. Model input parameters were estimated based on currently available evidence and treatment effects were obtained with Bayesian network meta-analysis techniques. The model contains 384 possible model structures informed by previously published models. The model consists of the following components: (i) modifiable R and C++ source code available in a GitHub repository; (ii) an R package to run the model for custom analyses; (iii) detailed model documentation; (iv) a web-based user interface for full control over the model without the need to be well-versed in the programming languages; and (v) a general audience web-application allowing those who are not experts in modeling or health economics to interact with the model and contribute to value assessment discussions.

RESULTS

A primary function of the initial version of RA model is to help understand and quantify the impact of parameter uncertainty (with probabilistic sensitivity analysis), structural uncertainty (with multiple competing model structures), the decision framework (cost-effectiveness analysis or multi-criteria decision analysis), and perspective (healthcare or limited societal) on estimates of value.

CONCLUSION

In order for a decision model to remain relevant over time it needs to evolve along with its supporting body of clinical evidence and scientific insight. Multiple clinical and methodological experts can modify or contribute to the RA model at any time due to its open-source nature.

R You Still Using Excel? The Advantages of Modern Software Tools for Health Technology Assessment

Economic models are used in health technology assessments (HTAs) to evaluate the cost-effectiveness of competing medical technologies and inform the efficient use of healthcare resources. Historically, these models have been developed with specialized commercial software (such as TreeAge) or more commonly with spreadsheet software (almost always Microsoft Excel). Although these tools may be sufficient for relatively simple analyses, they put unnecessary constraints on the analysis that may ultimately limit its credibility and relevance. In contrast, modern programming languages such as R, Python, Matlab, and Julia facilitate the development of models that are (i) clinically realistic, (ii) capable of quantifying decision uncertainty, (iii) transparent and reproducible, and (iv) reusable and adaptable. An HTA environment that encourages use of modern software can therefore help ensure that coverage and pricing decisions confer greatest possible benefit and capture all scientific uncertainty, thus enabling correct prioritization of future research.

Toward Relevant and Credible Cost-Effectiveness Analyses for Value Assessment in the Decentralized U.S. Health Care System

In the United States, there is an increased interest to understand the value of health technologies. Cost-effectiveness analysis is arguably the most appropriate framework to quantify value and to inform reimbursement decision making regarding medical interventions; however, a thorough analysis is resource intensive and complex. In many countries, the cost-effectiveness of medical interventions is evaluated by expert agencies at the national level, but in the United States, reimbursement decision making occurs at the local level. This raises the question of how we can provide a means to transparent cost-effectiveness analysis that reflects the local context and patient population and is based on the latest evidence and scientific insights. In other words, how can we maximize the relevance and credibility of cost-effectiveness evaluations in the context of a decentralized decision-making environment? Published cost-effectiveness analyses typically fail on these dimensions. Access to transparent open-source models that can be adapted to reflect the local setting in a relatively straightforward manner is an essential step toward such a goal. However, no model for cost-effectiveness analysis is ever truly "right" or "complete," and it must evolve along with clinical evidence and improvements in scientific methodology to ensure that its credibility remains. We propose a transparent approach of iterative development and collaboration between content and methodology experts to produce up-to-date, open-source consensus-based cost-effectiveness models that account for parameter and structural uncertainty to help local decision makers understand the confidence with which they might make a decision. Our proposed approach provides a way to adapt formal assessments of value-long the province of centralized health care systems-into the decentralized U.S. health care landscape. DISCLOSURES: This research was funded through the Innovation and Value Initiative, a nonprofit multistakeholder research organization. The Innovation and Value Initiative contracted with Precision Medicine Group for research activities related to this article. Jansen and Incerti are salaried employees and shareholders of Precision Medicine Group. Curtis is a paid consultant for the Innovation and Value Initiative. Curtis also reports consulting fees and grants from Amgen, AbbVie, BMS, Corrona, Janssen, Lilly, Myriad, Pfizer, Roche/Genentech, Radius, and UCB, unrelated to this article.

Incorporating single-arm evidence into a network meta-analysis using aggregate level matching: Assessing the impact

Increasingly, single-armed evidence is included in health technology assessment submissions when companies are seeking reimbursement for new drugs. While it is recognized that randomized controlled trials provide a higher standard of evidence, these are not available for many new agents that have been granted licenses in recent years. Therefore, it is important to examine whether alternative strategies for assessing this evidence may be used. In this work, we examine approaches to incorporating single-armed evidence formally in the evaluation process. We consider matching aggregate level covariates to comparator arms or trials and including this evidence in a network meta-analysis. We consider two methods of matching: (i) we include the chosen matched arm in the data set itself as a comparator for the single-arm trial; (ii) we use the baseline odds of an event in a chosen matched trial to use as a plug-in estimator for the single-arm trial. We illustrate that the synthesis of evidence resulting from such a setup is sensitive to the between-study variability, formulation of the prior for the between-design effect, weight given to the single-arm evidence, and extent of the bias in single-armed evidence. We provide a flowchart for the process involved in such a synthesis and highlight additional sensitivity analyses that should be carried out. This work was motivated by a hepatitis C data set, where many agents have only been examined in single-arm studies. We present the results of our methods applied to this data set.

Survival, Motor Function, and Motor Milestones: Comparison of AVXS-101 Relative to Nusinersen for the Treatment of Infants with Spinal Muscular Atrophy Type 1

Introduction

Infants with spinal muscular atrophy (SMA) type 1 typically face a decline in motor function and a severely shortened life expectancy. Clinical trials for SMA type 1 therapies, onasemnogene abeparvovec (AVXS-101) and nusinersen, demonstrated meaningful improvements in efficacy (e.g., overall survival) but there were no head-to-head clinical trials assessing comparative efficacy. This study estimated the treatment effects of AVXS-101 relative to nusinersen for the treatment of SMA type 1.

Methods

Overall survival, event-free survival (no death or need to use permanent assisted ventilation), improvement in motor function [increase of ≥ 4 points in Children’s Hospital of Philadelphia Infant Test of Neuromuscular Disorders (CHOP-INTEND) score from baseline], and motor milestone achievements (head control, rolling over, and sitting unassisted) reported in the onasemnogene abeparvovec (AVXS-101-CL-101; NCT02122952) and nusinersen (ENDEAR; NCT02193074) clinical trials were indirectly compared using frequentist and Bayesian approaches.

Results

Among symptomatic infants with SMA type 1, the number needed to treat (NNT) to prevent one more death with AVXS-101 instead of nusinersen was 6.2 [95% confidence intervals (CI) = 4.1–12.2], and the probability of preventing death was 20% higher for patients treated with AVXS-101 than nusinersen [risk ratio (RR) = 1.2, 95% CI 1.1–1.3]. For event-free survival, the NNT to prevent one more event was 2.6 (95% CI 2.0–3.6) and RR was 1.6 (95% CI 1.4–1.9). For improvement in motor function, NNT was 3.5 (95% CI 2.6–5.3) and RR was 1.4 (95% CI 1.2–1.6). For milestone achievements, the NNTs were 1.4 (95% CI 1.1–1.9), 1.5 (95% CI 1.1–2.5), and 1.2 (95% CI 1.0–1.5); RRs 4.2 (95% CI 2.6–6.7), 7.8 (95% CI 3.6–17.0), and 11.2 (95% CI 5.1–24.5) for head control, rolling over, and sitting unassisted, respectively. Results were similar using the Bayesian approach.

Conclusion

This indirect comparison (AVXS-101-CL-101 vs. ENDEAR) among symptomatic SMA type 1 infants suggests that AVXS-101 may have an efficacy advantage relative to nusinersen for overall survival, independence from permanent assisted ventilation, motor function, and motor milestones.

Funding

AveXis.

The number needed to treat and relevant between-trial comparisons of competing interventions

The number needed to treat (NNT) is considered an intuitive as well as popular effect measure. The aims of this review were to 1) explain why we cannot compare trial-specific NNT estimates for the competing treatments evaluated in different randomized controlled trials (RCTs) and 2) outline the principles of how relative treatment effects of different trials can be compared and results can be presented as NNT, without violating the principles of valid between-trial comparisons. Our premise is that ratio measures for relative treatment effects of response outcomes are less prone to effect modification than absolute difference measures of response outcomes. Accordingly, any between-trial comparisons of the efficacy of competing interventions using the study-specific ORs are less likely to be invalid or biased than comparisons based on the study-specific NNT estimates. However, treatment-specific ORs obtained from a meta-analysis or taken directly from an individual study can be transformed into consistent treatment-specific NNT estimates that allow for credible comparisons of treatments when these ratio measures are applied to the same reference response estimate. The theoretical discussion is illustrated with a relevant indirect comparison of biologics for the treatment of ulcerative colitis. Between-trial comparisons directly based on the NNT of individual trials may result in erroneous conclusions and should be avoided. Treatment-specific NNT estimates need to be based on the same probability of response with the common reference treatment against which the interventions are compared.

An indirect treatment comparison of the efficacy of insulin degludec/liraglutide (IDegLira) and insulin glargine/lixisenatide (iGlarLixi) in patients with type 2 diabetes uncontrolled on basal insulin

AIMS

To obtain estimates of the relative treatment effects between insulin degludec/liraglutide (IDegLira) and insulin glargine U100/lixisenatide (iGlarLixi) in patients with type 2 diabetes mellitus (T2DM) uncontrolled on basal insulin therapy.

MATERIALS AND METHODS

Data from phase 3 trials providing evidence for estimating the relative efficacy and safety of IDegLira vs iGlarLixi in patients uncontrolled on basal insulin-only regimens were used in this analysis. Outcomes of interest were changes in HbA_1c, body weight and insulin dose, and rate ratio of hypoglycemia. The indirect comparison of the reported trial findings followed the principles of Bucher et al.

RESULTS

IDegLira was estimated to provide a 0.44 [95% CI = 0.17-0.71] %-point reduction in HbA_1c compared with iGlarLixi. Body weight was reduced by 1.42 [95% CI = 0.35-2.50] kg with IDegLira compared with iGlarLixi. Insulin dose was comparable between the two interventions. The rate of severe or blood glucose-confirmed (self-measured plasma glucose [SMPG] ≤ 3.1 mmol/L) hypoglycemia with IDegLira was approximately half that of iGlarLixi (rate ratio = 0.51 [95% CI = 0.29-0.90]). However, using the American Diabetes Association definition of documented symptomatic hypoglycemia (SMPG ≤3.9 mmol/L) the rate was comparable between the two treatments (rate ratio = 1.07 [95% CI = 0.90-1.28]).

LIMITATIONS

The assumptions made in the indirect comparison and differences between the included trials in baseline HbA_1c levels, previous use of sulfonylureas, definitions of hypoglycemia, presence or absence of run-in period, the different duration of the trials, and the cross-over design of one of the trials.

CONCLUSIONS

The results of this indirect treatment comparison demonstrate that, among patients with T2DM uncontrolled on basal insulin, treatment with IDegLira results in a greater reduction of HbA_1c and a greater reduction in body weight compared with iGlarLixi at similar insulin doses.

Cost-effectiveness of sequenced treatment of rheumatoid arthritis with targeted immune modulators

AIMS

To determine the cost-effectiveness of treatment sequences of biologic disease-modifying anti-rheumatic drugs or Janus kinase/STAT pathway inhibitors (collectively referred to as bDMARDs) vs conventional DMARDs (cDMARDs) from the US societal perspective for treatment of patients with moderately to severely active rheumatoid arthritis (RA) with inadequate responses to cDMARDs.

MATERIALS AND METHODS

An individual patient simulation model was developed that assesses the impact of treatments on disease based on clinical trial data and real-world evidence. Treatment strategies included sequences starting with etanercept, adalimumab, certolizumab, or abatacept. Each of these treatment strategies was compared with cDMARDs. Incremental cost, incremental quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs) were calculated for each treatment sequence relative to cDMARDs. The cost-effectiveness of each strategy was determined using a US willingness-to-pay (WTP) threshold of $150,000/QALY.

RESULTS

For the base-case scenario, bDMARD treatment sequences were associated with greater treatment benefit (i.e. more QALYs), lower lost productivity costs, and greater treatment-related costs than cDMARDs. The expected ICERs for bDMARD sequences ranged from ∼$126,000 to $140,000 per QALY gained, which is below the US-specific WTP. Alternative scenarios examining the effects of homogeneous patients, dose increases, increased costs of hospitalization for severely physically impaired patients, and a lower baseline Health Assessment Questionnaire (HAQ) Disability Index score resulted in similar ICERs.

CONCLUSIONS

bDMARD treatment sequences are cost-effective from a US societal perspective.

Treatment modifying factors of biologics for psoriatic arthritis: a systematic review and Bayesian meta-regression

OBJECTIVES

The aim of this study was to explore factors that modify treatment effects of non-conventional biologics versus placebo in patients with psoriatic arthritis.

METHODS

A systematic literature review and meta-regression was conducted. The biologics included etanercept, infliximab, adalimumab, golimumab, certolizumab, ustekinumab, tocilizumab, anakinra, abatacept, rituximab, and secukinumab. Outcomes included American College of Rheumatology (ACR) 20 and 50, Psoriasis Area Severity Index (PASI) 75, and 36-Item Short Form Health Survey (SF-36) Physical and Mental Component Summaries (PCS and MCS).

RESULTS

Twelve RCTs were eligible for meta-regression. Treatment effects for ACR-20 at 12 weeks were higher in trials with longer disease durations (OR=2.94), and lower in trials enrolling older patients (OR=0.48), and those recently published (OR=0.49). Treatment effects for ACR-50 at 12 weeks were higher in trials with more males (OR=2.27), but lower in trials with high prior anti-TNF use (OR=0.28) and recently published trials (OR=0.37). For PASI-75, trials with more male patients (24 weeks: OR=2.56), and with higher swollen and tender joint counts (12 weeks: OR=8.33; 24 weeks: OR=14.44) showed higher treatment effects, and trials with high prior anti-TNF use had lower effects (OR=0.41). Treatment effects for SF-36 PCS at 24 weeks were higher in trials with longer psoriasis disease durations (OR=2.95) and PsA disease durations (OR=4.76), and those published earlier (OR=4.19).

CONCLUSIONS

Our analyses show that differences in baseline characteristics may explain some of the differences in response to biologics versus placebo across different trials. Accounting for these factors in future studies will likely be important.

Estimated burden of cardiovascular disease and value-based price range for evolocumab in a high-risk, secondary-prevention population in the US payer context

AIM

To estimate real-world cardiovascular disease (CVD) burden and value-based price range of evolocumab for a US-context, high-risk, secondary-prevention population.

MATERIALS AND METHODS

Burden of CVD was assessed using the UK-based Clinical Practice Research Datalink (CPRD) in order to capture complete CV burden including CV mortality. Patients on standard of care (SOC; high-intensity statins) in CPRD were selected based on eligibility criteria of FOURIER, a phase 3 CV outcomes trial of evolocumab, and categorized into four cohorts: high-risk prevalent atherosclerotic CVD (ASCVD) cohort (n = 1448), acute coronary syndrome (ACS) (n = 602), ischemic stroke (IS) (n = 151), and heart failure (HF) (n = 291) incident cohorts. The value-based price range for evolocumab was assessed using a previously published economic model. The model incorporated CPRD CV event rates and considered CV event reduction rate ratios per 1 mmol/L reduction in low-density lipoprotein-cholesterol (LDL-C) from a meta-analysis of statin trials by the Cholesterol Treatment Trialists Collaboration (CTTC), i.e. CTTC relationship.

RESULTS

Multiple-event rates of composite CV events (ACS, IS, or coronary revascularization) per 100 patient-years were 12.3 for the high-risk prevalent ASCVD cohort, and 25.7, 13.3, and 23.3, respectively, for incident ACS, IS, and HF cohorts. Approximately one-half (42%) of the high-risk ASCVD patients with a new CV event during follow-up had a subsequent CV event. Combining these real-world event rates and the CTTC relationship in the economic model, the value-based price range (credible interval) under a willingness-to-pay threshold of $150,000/quality-adjusted life-year gained for evolocumab was $11,990 ($9,341-$14,833) to $16,856 ($12,903-$20,678) in ASCVD patients with baseline LDL-C levels ≥70 mg/dL and ≥100 mg/dL, respectively.

CONCLUSION

Real-world CVD burden is substantial. Using the observed CVD burden in CPRD and the CTTC relationship, the cost-effectiveness analysis showed that, accounting for uncertainties, the expected value-based price for evolocumab is higher than its current annual cost, as long as the payer discount off list price is greater than 20%.

Tofacitinib Versus Biologic Treatments in Patients With Active Rheumatoid Arthritis Who Have Had an Inadequate Response to Tumor Necrosis Factor Inhibitors: Results From a Network Meta-analysis

PURPOSE

Tofacitinib is an oral Janus kinase inhibitor for the treatment of rheumatoid arthritis (RA). This analysis compared the efficacy and safety of tofacitinib with biologic disease-modifying antirheumatic drugs in patients with RA and a prior inadequate response (IR) to tumor necrosis factor inhibitors (TNFi).

METHODS

A systematic literature review identified 5 randomized placebo-controlled trials that evaluated tofacitinib or biologic disease-modifying antirheumatic drugs (bDMARDs) against placebo in patient populations with RA with a prior IR to TNFi. The definition of TNFi-IR varied across studies, and included patients with an IR or who had failed treatment with TNFi for any reason. A network meta-analysis was conducted comparing study data with regard to American College of Rheumatology response rates and Health Assessment Questionnaire-Disability Index improvement at weeks 12 and 24, rates of treatment withdrawal due to all causes; adverse events (AEs) and lack of efficacy; and rates of AEs, serious AEs, and serious infections.

FINDINGS

The 5 trials included a total of 2136 patients. Tofacitinib 5 mg twice daily combined with methotrexate was found to have relative risk estimates of American College of Rheumatology responses and change from baseline in Health Assessment Questionnaire-Disability Index score comparable with abatacept, golimumab, rituximab, and tocilizumab combined with conventional synthetic disease-modifying antirheumatic drugs. Withdrawal rates from trials due to all causes and AEs were comparable between treatments, and tofacitinib had a lower rate of withdrawals due to lack of efficacy. Rates of AEs and HAQ-DI were comparable between tofacitinib, other active treatments, and placebo. No serious infections were reported with tofacitinib during the placebo-controlled period (up to week 12) in this study population; rates of serious infection with other active treatments were generally low and similar to placebo.

IMPLICATIONS

During a 24-week period, tofacitinib had efficacy and rates of AEs comparable with currently available bDMARDs in the treatment of patients with RA who had a prior IR to TNFi. ClinicalTrials.gov identifiers: ORAL Step, NCT00960440; ATTAIN, NCT00124982; GO-AFTER, NCT00299546; RADIATE, NCT00106522; REFLEX, NCT00462345.

Meta-analysis of long-term joint structural deterioration in minimally treated patients with rheumatoid arthritis

BACKGROUND

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by inflammation and joint structural deterioration. Driven by recent expectations that patients in clinical trials randomized to placebo should be 'rescued' with active therapy within 6 months of starting treatment, the relative benefit of arresting joint damage with biologic agents beyond this period is unclear. With longer-term evidence of the rate of joint deterioration with minimal treatment, the efficacy of biologic agents and novel treatments might be projected beyond the placebo-controlled phase observed in clinical trials. The aim of this study was to estimate radiographic structural deterioration over time in patients with moderate-to-severe RA minimally treated with DMARDs.

METHODS

A literature review identified evidence of joint structural deterioration in patients with (DMARD-IR population) and without (non-DMARD-IR population) a history of inadequate response to DMARDs. Patients were minimally treated with one non-biologic DMARD or palliative care (non-DMARD-IR population only). Outcomes of interest were the (modified) Total Sharp Score (TSS) and subscales (Erosion Subscore [ES] and Joint Space Narrowing [JSN] Subscore), and Larsen score. Pooled joint-deterioration curves over time were obtained with meta-analysis models.

RESULTS

Mean change from baseline in TSS increased in the DMARD-IR population from 1.14 (95 % credible interval [CrI] 0.66, 1.67) to 9.84 (5.68, 14.46) at Weeks 12 and 104, respectively, and a non-linear increase of 1.56 (0.79, 2.34) and 5.13 (-1.35, 11.67) in the non-DMARD-IR population. At the same time points, mean changes (95 % CrI) were 0.51 (0.27, 0.83) and 4.43 (2.38, 7.21) for ES and 0.36 (0.09, 0.67) and 3.14 (0.80, 5.78) for JSN in the DMARD-IR population, whereas corresponding changes in the non-DMARD-IR population were 0.69 (0.31, 1.12) and 2.93 (0.92, 5.02), and 0.29 (0.17, 0.44) and 2.55 (1.45, 3.80), respectively. Larsen scores were only available for the non-DMARD-IR population, with mean changes (95 % CrI) of 0.08 (0.04, 0.11) and 0.65 (0.36, 0.96) at Weeks 12 and 104, respectively.

CONCLUSION

Minimal treatment of RA with one non-biologic DMARD results in deterioration of joint structure in patients with or without a history of inadequate response to non-biologic DMARDs.

[Thoracic endometriosis with aspecific manifestation]

BACKGROUND

Endometriosis is a common disease, especially in subfertile women. The most common location is in the pelvis, but extragenital locations are also possible. This far less common condition has been described in almost all tissues in the body. Symptoms occurring cyclically are characteristic of endometriosis.

CASE DESCRIPTION

A 37-year-old woman was discovered by chance to have ascites and pleural effusion. She had no symptoms of this. Thoracoscopy showed an image consistent with thoracic endometriosis. After initial drug therapy was unsuccessful, surgical intervention was performed.

CONCLUSION

Thoracic endometriosis is a rare disease, in which the cyclical nature of the symptoms often leads to correct diagnosis. Drug therapy is the preferred treatment for patients.

Estimating the value of medical treatments to patients using probabilistic multi criteria decision analysis

Background

Estimating the value of medical treatments to patients is an essential part of healthcare decision making, but is mostly done implicitly and without consulting patients. Multi criteria decision analysis (MCDA) has been proposed for the valuation task, while stated preference studies are increasingly used to measure patient preferences. In this study we propose a methodology for using stated preferences to weigh clinical evidence in an MCDA model that includes uncertainty in both patient preferences and clinical evidence explicitly.

Methods

A probabilistic MCDA model with an additive value function was developed and illustrated using a case on hypothetical treatments for depression. The patient-weighted values were approximated with Monte Carlo simulations and compared to expert-weighted results. Decision uncertainty was calculated as the probability of rank reversal for the first rank. Furthermore, scenario analyses were done to assess the relative impact of uncertainty in preferences and clinical evidence, and of assuming uniform preference distributions.

Results

The patient-weighted values for drug A, drug B, drug C, and placebo were 0.51 (95 % CI: 0.48 to 0.54), 0.51 (95 % CI: 0.48 to 0.54), 0.54 (0.49 to 0.58), and 0.15 (95 % CI: 0.13 to 0.17), respectively. Drug C was the most preferred treatment and the rank reversal probability for first rank was 27 %. This probability decreased to 18 % when uncertainty in performances was not included and increased to 41 % when uncertainty in criterion weights was not included. With uniform preference distributions, the first rank reversal probability increased to 61 %. The expert-weighted values for drug A, drug B, drug C, and placebo were 0.67 (95 % CI: 0.65 to 0.68), 0.57 (95 % CI: 0.56 to 0.59), 0.67 (95 % CI: 0.61 to 0.71), and 0.19 (95 % CI: 0.17 to 0.21). The rank reversal probability for the first rank according to experts was 49 %.

Conclusions

Preferences elicited from patients can be used to weigh clinical evidence in a probabilistic MCDA model. The resulting treatment values can be contrasted to results from experts, and the impact of uncertainty can be quantified using rank probabilities. Future research should focus on integrating the model with regulatory decision frameworks and on including other types of uncertainty.

Ranibizumab vs. aflibercept for wet age-related macular degeneration: network meta-analysis to understand the value of reduced frequency dosing

OBJECTIVE

Although a reduced aflibercept (2.0 mg) injection frequency relative to the approved dosing posology is included in national treatment guidelines for wet age-related macular degeneration (AMD), there is limited evidence of its comparative efficacy. The objective was to compare the efficacy and safety of reduced frequency dosing for aflibercept, relative to other approved and marketed vascular endothelial growth factor inhibitors for wet AMD, over 12 months.

RESEARCH DESIGN AND METHODS

Based on a systematic literature review performed according to a pre-specified protocol, a Bayesian network meta-analysis (NMA) was conducted to indirectly compare posologies of aflibercept and ranibizumab (0.5 mg). The efficacy outcome, mean change from baseline in best-corrected visual acuity (BCVA) on the ETDRS chart, was evaluated at 3 and 12 months; and safety data at 12 months. Standard NMA models were used to analyze change at 3 months, and fractional polynomial regression over 12 months. Safety data were analyzed using binomial models with a logistic link function.

RESULTS

Five trials formed a complete evidence network. At 3 months, all posologies of aflibercept and ranibizumab resulted in similar changes in BCVA. Over 12 months, approved posologies of aflibercept and ranibizumab resulted in similar changes from baseline, between 6.7 (95% credible interval [CrI], 5.5, 7.8) and 9.1 (8.1, 10.1) ETDRS letters; however, reduced frequency aflibercept was associated with a smaller change (1.8 letters, [-25.9, 29.2]). There was a trend towards a greater change in BCVA, with increasing frequency of dosing. All posologies performed similarly with respect to safety, and CrIs were wide.

CONCLUSIONS

Approved posologies of ranibizumab and aflibercept are similarly effective treatments for wet AMD. Reduced frequency aflibercept was associated with the poorest visual outcomes, and sample sizes were small. Findings from these analyses provide novel evidence of the comparative efficacy and safety of aflibercept and ranibizumab for wet AMD.

A Novel Dynamic Model for Health Economic Analysis of Influenza Vaccination in the Elderly

Introduction

New vaccines are being developed to improve the efficacy of seasonal influenza immunization in elderly persons aged ≥65 years. These products require clinical and economic evaluation to aid policy decisions.

Methods

To address this need, a two-part model has been developed, which we have applied to examine the potential clinical and economic impact of vaccinating elderly persons with adjuvanted trivalent inactivated influenza vaccine (aTIV) relative to conventional trivalent (TIV) and quadrivalent (QIV) vaccines. We compared outcomes in the US population for (1) aTIV in persons aged ≥65 years and QIV in all other age cohorts; (2) QIV in all cohorts; (3) TIV in all cohorts. Low, average, and high intensity seasons with low, average, and high vaccine match scenarios were compared. Probabilistic sensitivity analysis was conducted within each discrete scenario to explore the impact of variation in model inputs on potential outcomes.

Results

Assuming current vaccination coverage rates in the US population with (a) 25% better efficacy of adjuvanted versus non-adjuvanted vaccine against any strain and (b) 35% better efficacy of non-adjuvanted vaccine against matched B versus mismatched B strains, use of aTIV in persons aged ≥65 years and QIV in persons <65 years could reduce influenza cases by 11,166–1,329,200, hospitalizations by 1365–43,674, and deaths by 421–11,320 versus use of QIV in all cohorts. These outcomes are reflected in a corresponding increase in quality-adjusted life-years (QALYs) of 3003–94,084. If the prevalence of mismatched influenza B was >54.5% of all circulating strains, use of QIV in all cohorts would offset the clinical benefits of aTIV. Elderly aTIV or QIV vaccination was associated with improved outcomes over non-adjuvanted TIV in many of the scenarios, particularly in low match seasons of any intensity. Total cost savings (including direct and indirect healthcare costs plus productivity impacts) with aTIV in the elderly versus QIV in the whole population ranged from $27 million (low intensity, low match) to $934 million (high intensity, high match). Univariate sensitivity analysis of relative vaccine prices in the average intensity, average match scenario indicated that aTIV could be marginally cost saving relative to QIV at the currently published Medicare price for influenza vaccines offering enhanced efficacy in the elderly. Elderly vaccination with aTIV was associated with a higher overall cost compared with TIV in only two scenarios (low intensity with average or high match); the incremental cost/QALY relative to TIV was $9980 in the average match scenario and $28,800 in the high match scenario.

Conclusions

Vaccination of persons aged ≥65 years with aTIV has the potential to provide clinical and economic benefit relative to QIV and TIV. The new model allows the assessment of various alternative strategies for available influenza vaccines.

Funding

Novartis Vaccines.

Electronic supplementary material

The online version of this article (doi:10.1007/s40121-015-0076-8) contains supplementary material, which is available to authorized users.

Accounting for correlation in network meta-analysis with multi-arm trials

Multi-arm trials (trials with more than two arms) are particularly valuable forms of evidence for network meta-analysis (NMA). Trial results are available either as arm-level summaries, where effect measures are reported for each arm, or as contrast-level summaries, where the differences in effect between arms compare with the control arm chosen for the trial. We show that likelihood-based inference in both contrast-level and arm-level formats is identical if there are only two-arm trials, but that if there are multi-arm trials, results from the contrast-level format will be incorrect unless correlations are accounted for in the likelihood. We review Bayesian and frequentist software for NMA with multi-arm trials that can account for this correlation and give an illustrative example of the difference in estimates that can be introduced if the correlations are not incorporated. We discuss methods of imputing correlations when they cannot be derived from the reported results and urge trialists to report the standard error for the control arm even if only contrast-level summaries are reported. Copyright © 2012 John Wiley & Sons, Ltd.

The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations

The PRISMA statement is a reporting guideline designed to improve the completeness of reporting of systematic reviews and meta-analyses. Authors have used this guideline worldwide to prepare their reviews for publication. In the past, these reports typically compared 2 treatment alternatives. With the evolution of systematic reviews that compare multiple treatments, some of them only indirectly, authors face novel challenges for conducting and reporting their reviews. This extension of the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) statement was developed specifically to improve the reporting of systematic reviews incorporating network meta-analyses. A group of experts participated in a systematic review, Delphi survey, and face-to-face discussion and consensus meeting to establish new checklist items for this extension statement. Current PRISMA items were also clarified. A modified, 32-item PRISMA extension checklist was developed to address what the group considered to be immediately relevant to the reporting of network meta-analyses. This document presents the extension and provides examples of good reporting, as well as elaborations regarding the rationale for new checklist items and the modification of previously existing items from the PRISMA statement. It also highlights educational information related to key considerations in the practice of network meta-analysis. The target audience includes authors and readers of network meta-analyses, as well as journal editors and peer reviewers.

Cost per remission and cost per response with infliximab, adalimumab, and golimumab for the treatment of moderately-to-severely active ulcerative colitis

OBJECTIVE

To determine the short-term costs per sustained remission and sustained response of three tumor necrosis factor inhibitors (infliximab, adalimumab, and golimumab) in comparison to conventional therapy for the treatment of moderately-to-severely active ulcerative colitis.

METHODS

A probabilistic Markov model was developed. This included an 8-week induction period, and 22 subsequent 2-week cycles (up to 1 year). The model included three disease states: remission, response, and relapse. Costs were from a Canadian public payer perspective. Estimates for the additional cost per 1 year of sustained remission and sustained response were obtained.

RESULTS

Golimumab 100 mg provided the lowest cost per additional remission ($935) and cost per additional response ($701) compared with conventional therapy. Golimumab 50 mg yielded slightly higher costs than golimumab 100 mg. Infliximab was associated with the largest additional number of estimated remissions and responses, but also higher cost at $1975 per remission and $1311 per response. Adalimumab was associated with the largest cost per remission ($7430) and cost per response ($2361). The cost per additional remission and cost per additional response associated with infliximab vs golimumab 100 mg was $14,659 and $4753, respectively.

CONCLUSIONS

The results suggest that the additional cost of 1 full year of remission and response are lowest with golimumab 100 mg, followed by golimumab 50 mg. Although infliximab has the highest efficacy, it did not exhibit the lowest cost per additional remission or response. Adalimumab produced the highest cost per additional remission and response.