Defining a value-based price corridor (orphan drugs)

A consistent definition of the value of rare disease treatment is lacking. Traditional economic efficiency approaches used for the value assessment of treatments intended for more common diseases have structural limitations in rare disease treatments. The value of orphan medicine products (OMPs) is generally perceived as empirical and as a function of the sole rarity of a specific disease. In this project, we investigated how the perceived values of OMPs, as expressed by a Multi-criteria Decision Analysis (MCDA) value score was linked to observed OMPs pricing decisions.

Non-oncology and first-to-market OMPs approved in Europe and USA were assigned a medicine value (MV) score by expert panels during MCDA workshops. The MCDA exercise combined 8 value attributes into a unique and weighted summary score. We retrieved the corresponding target population size (N) and price (P) for each OMP indication for the “big” 5 European countries and the US. P was calculated as the average annual treatment cost. Multiple regression analyses were performed using P as the dependent variable and MV and N as the 2 independent variables. As the relationship between P and the independent variables MV and N were not linear, we fitted non-linear regression models (exponential, power and mixed models). Analyses were carried out for each country separately and for the EU5.

Value to client: We found that the OMP prices were associated with a multidimensional value construct and the size of the target population. We formalized mathematically this association. This helped our client to better inform and substantiate future value-based pricing decisions on innovative OMPs, for which the prices are to be negotiated in increasingly tense budgetary environments.

 

Prioritizing vaccines subject to a budgetary constraint (childhood vaccination)

To guide healthcare decision, decision analytic modeling mainly focuses on cost-effectiveness analysis (CEA). CEA is primarily a measure of economic value, notably expressed in terms of cost per life years (LYG) or cost per quality-adjusted life years (QALYs) gained. As CEA does not document the impact of an intervention on healthcare budget, it is complemented with budget impact analysis (BIA). BIA provides financial projections but does not necessarily quantify the beneficial health effect an intervention would have on the targeted population.

Moreover, CEA and BIA usually compare two mutually exclusive interventions. For instance, a vaccine is compared against no vaccine or an alternative vaccine. Portfolio of vaccines or the sequential introduction of different vaccines for childhood immunization is generally not considered.

Decision makers operate within finite short-term budget and may have different preferences on what constitutes the population health value and priorities of an intervention. The objective of this project was to develop a multiple criteria optimization methodology to allocate constrained healthcare budget across a portfolio of vaccines.

We proposed to optimize the sequential introduction of different vaccines while taking into account multiple preferences of decision makers in terms of public health priorities. Our model explicitly pondered the decision maker preferences and annual budget constraint.

Value to client:  An Excel-based tool to support local affiliates and local vaccines payers in setting up childhood vaccination programs to reach specific public health goals with fixed budget constraints.

 

Forecasting the financial implications of a risk-sharing scheme (oncology)

The objective was to develop an operational modeling framework to help in the designing of a potential Performance-Based Risk- Sharing (PBRS) scheme for an oncology product. A time-to-event endpoint was used as a performance criterion (progression-free survival).

The framework was based on an open population model with a monthly cycle and a 3-year time horizon from launch i.e. when enrolment into the PBRS scheme would start.

The model could accommodate different treatment dosing schedules and performance levels (i.e. minimum progression-free survival times guaranteed). Multiple PBRS scenarios could be run and compared in terms of their operational and financial implications for both, the payer and the manufacturer.

The effect of potential revisions of a PBRS scheme terms and conditions could also be examined as real-life information will would become available following the scheme implementation (i.e. Bayesian updating).

Value to client:  This modeling framework provided both payer and manufacturer with valuable insight into the operational and financial implications of the potential PBRS schemes they may contemplate as they negotiate patient access conditions. Based on the simulations provided, both parties could better anticipate the implications of the schemes and better plan resources, logistics and financial arrangements accordingly.

 

Defining the clinical utility of a companion diagnostic (oncology)

Personalized medicine (PM) commonly involves the development of companion diagnostic tests to guide optimal treatment selection (selecting patients with higher likelihood of response for instance). PM thus has the potential to dramatically improve outcomes for specific patients and thereby optimize allocation of resources. However, very few attempts have been made that transparently include the diagnostic test performance (sensitivity and specificity) into cost-effectiveness or budget impact models.

The objective of this project was to design a model assessing the value of a companion diagnostic based PM strategy taking into consideration the diagnostic test performance.

Value to client: The designed analytical framework enabled the manufacturer to gauge the potential value of its PM strategy at all stage of development, based on the expected performance of the companion diagnostic, the expected effectiveness differential between the PM and the standard of care and the anticipated prevalence of the predictive biomarker in the target patients population.