Which methodologies were tested?

BRAT Framework PrOACT-URL MCDA (Deterministic) MCDA (Stochastic) w-NCB NNT & NNH BRR Direct Utility Elicitation

PrOACT-URL Framework

In this case study, the PrOACT-URL was streamlined against the steps in BRAT framework to identify common elements. PrOACT-URL was therefore not fully adopted (see Natalizumab case study report). The table below shows the steps involved in PrOACT-URL and their descriptions using the natalizumab example.

Table 1 Literature searching and data selection strategy

Step Description
1) Problem What is the benefit-risk balance of natalizumab following the occurrence of PML cases?
Decision 1: Should natalizumab be given marketing approval at the time of first registration?
  • This analysis used the incidence of PML observed at the time of CHMP re-evaluation and lead to a benefit-risk balance in favour of natalizumab compared to all the other treatments, which would be improved by removing the PML incidence. Hence natalizumab should have been given marketing approval at the time of first registration.

Decision 2: Should natalizumab be kept on the market given that episodes of PML are observed at the time that these episodes were observed (at time of CHMP re-evaluation)?
  • The incidence of PML at the time of CHMP re-evaluation still leads to a benefit-risk balance in favour of natalizumab. The sensitivity analysis shows that this balance is robust to changes in the weight given to PML or the incidence of PML.
2) Objectives Benefits: Reduction in relapse rate, slowdown in disability progression.
Risks: PML, reactivation of serious herpes viral infections, seizures, abortion or congenital abnormalities, transaminases elevation, infusion or injection site reactions, hypersensitivity reactions, flu-like reactions.
3) Alternatives Interferon beta-1a, glatiramer acetate, placebo. Which option to choose?
4) Consequences Data source: EPARs and literature search
5) Trade offs Benefit-risk comparison using (patient-) assigned weights and value functions (see MCDA (Deterministic))
6) Uncertainty Deterministic or stochastic sensitivity analysis on the weights and outcome measures (see MCDA (Deterministic) and MCDA (Stochastic))
7) Risk tolerance Not considered relevant for this case study as we consider this a standalone example to explore the methodology
8) Linked decisions Not considered relevant for this case study as we consider this a standalone example to explore the methodology