NETWORK META-ANALYSIS IN THE EVALUATION OF VACCINES
Author(s)
McGirr A1, Widenmaier R1, Burnett H2, Curran D3, Espie E3, Mrkvan T3, Oostvogels L3, Simone B3, McElhaney J4
1GSK Canada, Mississauga, ON, Canada, 2Mapi, Dundas, ON, Canada, 3GSK, Wavre, Belgium, 4Health Sciences North Research institute, Sudbury, ON, Canada
BACKGROUND: Head-to-head clinical efficacy randomized controlled trials (RCTs) for vaccines are often unfeasible due to logistical and/or ethical constraints and comparative real-world effectiveness studies are methodologically challenging. Without direct efficacy or effectiveness comparisons, relative risk/benefit of vaccines can be challenging to assess. OBJECTIVES: While network meta-analyses (NMAs) have been used extensively to estimate the comparative efficacy and safety of therapeutic interventions, the application of this method to compare vaccines is rare. To our knowledge, there is only one published vaccine NMA which compares the relative effectiveness of rotavirus vaccines. We discuss how NMAs can be used to evaluate the relative efficacy and safety of vaccines within a specific disease area; using vaccines to prevent herpes zoster (HZ) as another case example. METHODS: NMA methodology can be applied to indirectly compare vaccines; however, issues such as waning immunity, indirect vaccine effect, changing epidemiology over time, and seasonality of the disease of interest pose unique challenges. Detailed feasibility assessments should be undertaken to systematically determine if NMA is appropriate from a methodological and clinical perspective. This requires close collaboration between methodologists and clinical experts. Recently, a comparison of vaccines for HZ was deemed feasible, in particular due to availability of evidence studied in large, well-designed RCTs, and due to the epidemiology of HZ. Key challenges faced included variations in effect modifiers such as patient age, follow-up duration, and outcome definitions. It was possible to leverage existing NMA guidance to provide credible and reliable estimates of the relative effect of HZ vaccines that have not been studied in head-to-head efficacy trials. CONCLUSIONS: NMAs provide a method to indirectly compare different medical interventions, including vaccines, in the absence of head-to-head data. With RCT data and real world evidence, NMAs can contribute to inform vaccine recommendations and public health decision making.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PRM258
Topic
Methodological & Statistical Research
Topic Subcategory
Confounding, Selection Bias Correction, Causal Inference
Disease
Infectious Disease (non-vaccine)