Placebo use in vaccine trials: Recommendations of a WHO expert panel

Abstract

HighlightsPlacebo controls may be acceptable even when an efficacious vaccine exists, in the following four possible situations:When developing a locally affordable vaccine.When evaluating the local safety and efficacy of an existing vaccine.When testing a new vaccine when an existing vaccine is not considered appropriate locally.When determining the local burden of disease.

Report from the World Health Organization's Product Development for Vaccines Advisory Committee (PDVAC) meeting, Geneva, 7–9th Sep 2015

“…PDVAC considered the issue of trials of second generation vaccines, and whether or not there would be a need for comparability studies with the existing vaccine. It was noted that there are circumstances in which randomised, placebo controlled studies may be justified even when an efficacious vaccine exists, depending on a number of factors [13]. Related to this, WHO is planning to initiate a process in 2016 on trial design considerations for second generation dengue vaccines and their path to licensure.…”

The Ebola Vaccine Team B: a model for promoting the rapid development of medical countermeasures for emerging infectious disease threats

“…In such situations, clinical trials would likely be done by comparing the new candidate vaccine against the fi rst-generation vaccine. 17,18 This might apply to the development of future vaccines for other diseases as well, such as malaria and dengue, where second-generation vaccines are compared with partly effi cacious fi rst-generation vaccines. 19…”

“…Vaccine is provided as a liquid or lyophilised product in mono-dose or low multi-dose (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) presentations § §, ¶ ¶ Multi-dose presentations should be formulated, managed, and discarded in compliance with multi-dose vial policies Lyophilised vaccine will need to be accompanied by paired separate vials of the appropriate diluent…”

“…Vaccine is provided as a liquid or lyophilised product in mono-dose or low multi-dose (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) presentations § §, ¶ ¶ Multi-dose presentations should be formulated, managed, and discarded in compliance with multi-dose vial policies Lyophilised vaccine will need to be accompanied by paired separate vials of the appropriate diluent Production Can be produced effi ciently and as expeditiously as possible after a validated scale-up that allows for maximum production yields; the dose of antigen required for protection allows for high production yield (which will aff ect cost and availability) 5 million doses can be produced by the end of 2015 (for the current epidemic) Ideally, production involves a single bulk-substance product (without requiring a separate booster product or diluent [needed for lyophilised vaccines]) If a booster with an alternative product is needed, that product also can be produced quickly and without substantial manufacturing barriers or supplychain issues If an adjuvant is needed, it can be formulated with the vaccine instead of combined at the time of use…”

Observational studies and the difficult quest for causality: lessons from vaccine effectiveness and impact studies

“…Even within a RCT, the effects of a vaccine on a subgroup defined after randomization (and therefore not protected by randomization against confounding and selection bias) may be of interest; such subgroup analyses are observational, despite being nested within a RCT. Once a vaccine is licensed and recommended for use, certain RCTs may face ethical challenges (1) , though there may remain VE questions of interest. Finally, RCTs randomized at the individual level are designed to measure only the direct protection the vaccine offers to vaccinated persons, but not the important overall effect of vaccination on disease in the population, including that achieved by indirect protection of unvaccinated people (herd immunity).…”