The Application and Implications of Novel Deterministic Sensitivity Analysis Methods.

Deterministic sensitivity analyses (DSA) remain important to interpret the effect of uncertainties in individual parameters on results of cost-effectiveness analyses. Classic DSA methodologies may lead to wrong conclusions due to a lack of or misleading information regarding marginal effects, non-linearity, likelihood and correlations. In addition, tornado diagrams are misleading in some situations. Recent advances in DSA methods have the potential to provide decision makers with more reliable information regarding the effects of uncertainties in individual parameters. This practical application discusses advances to classic DSA methods and their implications. Three methods are discussed: stepwise DSA, distributional DSA and probabilistic DSA. For each method, the technical specifications, options for presenting results, and its implications for decision making are discussed. Options for visualizing DSA results in incremental cost-effectiveness ratios and in incremental net benefits are presented. The use of stepwise DSA increases interpretability of marginal effects and non-linearities in the model, which is especially relevant when arbitrary ranges are implemented. Using the probability distribution of each parameter in distributional DSA provides insight on the likelihood of model outcomes while probabilistic DSA also includes the effects of correlations between parameters.Probabilistic DSA, preferably expressed in incremental net benefit, is the most appropriate method for providing insight on the effect of uncertainty in individual parameters on the estimate of cost effectiveness. However, the opportunities provided by probabilistic DSA may not always be needed for decision making. Other DSA methods, in particular distributional DSA, can sometimes be sufficient depending on model features. Decision makers must determine to which extent they will accept and implement these new and improved DSA methodologies and adjust guidelines accordingly.

Increasing the information provided by probabilistic sensitivity analysis: The relative density plot.

BACKGROUND: Results of probabilistic sensitivity analyses (PSA) are frequently visualized as a scatterplot, which is limited through overdrawing and a lack of insight in relative density. To overcome these limitations, we have developed the Relative Density plot (PSA-ReD). METHODS: The PSA-ReD combines a density plot and a contour plot to visualize and quantify PSA results. Relative density, depicted using a color gradient, is transformed to a cumulative probability. Contours are then plotted over regions with a specific cumulative probability. We use two real-world case studies to demonstrate the value of the PSA-ReD plot. RESULTS: The PSA-ReD method demonstrates proof-of-concept and feasibility. In the real-world case-studies, PSA-ReD provided additional visual information that could not be understood from the traditional scatterplot. High density areas were identified by color-coding and the contour plot allowed for quantification of PSA iterations within areas of the cost-effectiveness plane, diminishing overdrawing and putting infrequent iterations in perspective. Critically, the PSA-ReD plot informs modellers about non-linearities within their model. CONCLUSIONS: The PSA-ReD plot is easy to implement, presents more of the information enclosed in PSA data, and prevents inappropriate interpretation of PSA results. It gives modelers additional insight in model functioning and the distribution of uncertainty around the cost-effectiveness estimate.

A novel method for predicting the budget impact of innovative medicines: validation study for oncolytics.

BACKGROUND: High budget impact (BI) estimates of new drugs have led to decision-making challenges potentially resulting in restrictions in patient access. However, current BI predictions are rather inaccurate and short term. We therefore developed a new approach for BI prediction. Here, we describe the validation of our BI prediction approach using oncology drugs as a case study. METHODS: We used Dutch population-level data to estimate BI where BI is defined as list price multiplied by volume. We included drugs in the antineoplastic agents ATC category which the European Medicines Agency (EMA) considered a New Active Substance and received EMA marketing authorization (MA) between 2000 and 2017. A mixed-effects model was used for prediction and included tumor site, orphan, first in class or conditional approval designation as covariates. Data from 2000 to 2012 were the training set. BI was predicted monthly from 0 to 45 months after MA. Cross-validation was performed using a rolling forecasting origin with e^|Ln(observed BI/predicted BI)| as outcome. RESULTS: The training set and validation set included 25 and 44 products, respectively. Mean error, composed of all validation outcomes, was 2.94 (median 1.57). Errors are higher with less available data and at more future predictions. Highest errors occur without any prior data. From 10 months onward, error remains constant. CONCLUSIONS: The validation shows that the method can relatively accurately predict BI. For payers or policymakers, this approach can yield a valuable addition to current BI predictions due to its ease of use, independence of indications and ability to update predictions to the most recent data.

Affordability of oncology drugs: accuracy of budget impact estimations.

Background: In many countries, Budget Impact (BI) informs reimbursement decisions. Evidence has shown that decision-makers have restricted access based on high BI estimates but studies show that BI estimates are often inaccurate. Objective: To assess the accuracy of BI estimations used for informing access decisions on oncology drugs in the Netherlands. Study Design: Oncology products for which European Medicines Agency Marketing Authorisation was granted between 1-1-2000 and 1-10-2017 were selected. Observed BI data were provided by FarmInform. BI estimates were extracted from the reimbursement dossiers of the Dutch Healthcare Institute. Products without an estimated BI in the reimbursement dossier were excluded. Accuracy is defined as the ratio observed BI/estimated BI. Setting: General community, the Netherlands. Results: Ten products were included in the base case analysis. Mean accuracy was 0.64 and observed BI deviated by more than 40% and 100% from the estimated BI for 4 and 5 products, respectively. For all products together, €141 million BI was estimated and €82 million BI was observed, a €59 million difference. Conclusions: The findings indicate that BI estimates for oncology drugs in the Netherlands are inaccurate. The role and use of BI in reimbursement decisions for these potentially life-saving drugs should therefore be considered carefully, as well as BI estimation methodology.

Phase I/II Clinical Trial-Based Early Economic Evaluation of Acalabrutinib for Relapsed Chronic Lymphocytic Leukaemia.

OBJECTIVES: The objective of this study was to construct an early economic evaluation for acalabrutinib for relapsed chronic lymphocytic leukaemia (CLL) to assist early reimbursement decision making. Scenarios were assessed to find the relative impact of critical parameters on incremental costs and quality-adjusted life-years (QALYs). METHODS: A partitioned survival model was constructed comparing acalabrutinib and ibrutinib from a UK national health service perspective. This model included states for progression-free survival (PFS), post-progression survival (PPS) and death. PFS and overall survival (OS) were parametrically extrapolated from ibrutinib publications and a preliminary hazard ratio based on phase I/II data was applied for acalabrutinib. Deterministic and probabilistic sensitivity analyses were performed, and 1296 scenarios were assessed. RESULTS: The base-case incremental cost-effectiveness ratio (ICER) was £61,941/QALY, with 3.44 incremental QALYs and incremental costs of £213,339. Deterministic sensitivity analysis indicated that survival estimates, utilities and treatment costs of ibrutinib and acalabrutinib and resource use during PFS have the greatest influence on the ICER. Probabilistic results under different development scenarios indicated that greater efficacy of acalabrutinib would decrease the likelihood of cost effectiveness (from 63% at no effect to 2% at maximum efficacy). Scenario analyses showed that a reduction in PFS did not lead to great QALY differences (- 8 to - 14% incremental QALYs) although it did greatly affect costs (- 47 to - 122% incremental pounds). For OS, the opposite was true (- 89 to - 93% QALYs and - 7 to - 39% pounds). CONCLUSIONS: Acalabrutinib is not likely to be cost effective compared with ibrutinib under current development scenarios. The conflicting effects of OS, PFS, drug costs and utility during PFS show that determining the cost effectiveness of acalabrutinib without insight into all parameters complicates health technology assessment decision making. Early assessment of the cost effectiveness of new products can support development choices and reimbursement processes through effective early dialogues between stakeholders.

Accuracy of budget impact estimations and impact on patient access: a hepatitis C case study.

BACKGROUND: High budget impact (BI) estimates of new drugs limit access to patients due to concerns regarding affordability and displacement effects. The accuracy and methodological quality of BI analyses are often low, potentially mis-informing reimbursement decision making. Using hepatitis C as a case study, we aim to quantify the accuracy of the BI predictions used in Dutch reimbursement decision-making and to characterize the influence of market-dynamics on actual BI. METHODS: We selected hepatitis C direct-acting antivirals (DAAs) that were introduced in the Netherlands between January 2014 and March 2018. Dutch National Health Care Institute (ZIN) BI estimates were derived from the reimbursement dossiers. Actual Dutch BI data were provided by FarmInform. BI prediction accuracy was assessed by comparing the ZIN BI estimates with the actual BI data. RESULTS: Actual BI, from 1 Jan 2014 to 1 March 2018, was €248 million whilst the BI estimates ranged from €388-€510 million. The latter figure represents the estimated BI for the reimbursement scenario that was adopted, implying a €275 million overestimation. Absent incorporation of timing of regulatory decisions and inadequate correction for the introduction of new products were main drivers of BI overestimation, as well as uncertainty regarding the patient population size and the impact of the final reimbursement decision. DISCUSSION: BI in reimbursement dossiers largely overestimated actual BI of hepatitis C DAAs. When BI analysis is performed according to existing guidelines, the resulting more accurate BI estimates may lead to better informed reimbursement decisions.

Whole-exome sequencing in intellectual disability; cost before and after a diagnosis.

Clinical application of whole-exome and whole-genome sequencing (WES and WGS) has led to an increasing interest in how it could drive healthcare decisions. As with any healthcare innovation, implementation of next-generation sequencing in the clinic raises questions on affordability and costing impact for society as a whole. We retrospectively analyzed medical records of 370 patients with ID who had undergone WES at various stages of their diagnostic trajectory. We collected all medical interventions performed on these patients at the University Medical Center Utrecht (UMCU), Utrecht, the Netherlands. We categorized the patients according to their WES-based preliminary diagnosis ("yes", "no", and "uncertain"), and assessed the per-patient healthcare activities and corresponding costs before (pre) and after (post) genetic diagnosis. The WES-specific diagnostic yield among the 370 patients was 35% (128 patients). Pre-WES costs were €7.225 on average. Highest average costs were observed for laboratory-based tests, including genetics, followed by consults. Compared to pre-WES costs, the post-WES costs were on average 80% lower per patient, irrespective of the WES-based diagnostic outcome. Application of WES results in a considerable reduction of healthcare costs, not just in current settings, but even more so when applied earlier in the diagnostic trajectory (genetics-first). In such context, WES may replace less cost-effective traditional technologies without compromising the diagnostic yield. Moreover, WES appears to harbor an intrinsic "end-of-trajectory" effect; regardless of the diagnosis, downstream medical interventions decrease substantially in both number and costs.

Cost of illness of metastatic prostate cancer: a perspective of costs for new treatment options in The Netherlands.

AIM: To assess the resource use and associated costs of treating patients with metastatic prostate cancer with a focus on skeletal-related events (SREs). METHODS: We performed a bottom-up cost of illness study in The Netherlands. RESULTS: A total of 136 patients were studied. The mean total costs were €17,931 per patient. SREs that required hospitalization (n = 53) were, at median costs of €2039-9346, depending on care. These SREs had median costs of €200-1912. CONCLUSION: Our data provide a basis to investigate the cost-effectiveness of novel treatment options for metastatic prostate cancer. The impact of SREs on total costs could justify policy aimed at actively preventing SREs, possibly resulting in better quality of life and cost-reduction.

Early health technology assessments in pharmacogenomics: a case example in cardiovascular drugs.

AIM: To assess the required characteristics (cost, sensitivity and specificity) of a pharmacogenomic test for being a cost-effective prevention of angiotensin-converting enzyme inhibitors induced angioedema. Furthermore, we assessed the influence of only testing high-risk populations. MATERIALS & METHODS: A decision tree was used. RESULTS: With a willingness-to-pay threshold of €20,000 and €80,000 per quality adjusted life year, a 100% sensitive and specific test may have a maximum cost of €1.30 and €1.95, respectively. When only genotyping high-risk populations, the maximum test price would be €5.03 and €7.55, respectively. CONCLUSION: This theoretical pharmacogenomic test is only cost-effective at high specificity, high sensitivity and a low price. Only testing high-risk populations yields more realistic maximum test prices for cost-effectiveness of the intervention.