Intended for healthcare professionals

Rapid response to:

Editorials

The RTS,S malaria vaccine

BMJ 2011; 343 doi: https://doi.org/10.1136/bmj.d6986 (Published 27 October 2011) Cite this as: BMJ 2011;343:d6986

Rapid Response:

Re: The RTS,S malaria vaccine

Christopher J.A. Duncan and Adrian V.S. Hill

We read with great interest Whitty’s insightful editorial [1] placing the recent phase III RTS,S malaria vaccine trial results [2] in the wider context of malaria control efforts, and we agree that while vaccines are an important potential component, they are not the complete solution. As the editorial points out, RTS,S is clearly a big step in the right direction, although further incremental advances are needed towards the goal of a highly effective malaria vaccine [1].

The decision to publish interim efficacy data from the ongoing phase III study is rather unusual [3], and others have questioned the headline efficacy figure against first or all episodes of clinical malaria of around 50% in children aged 5-17 months [4]. Efficacy estimates will critically influence decisions on the role for RTS,S, and we wish to clarify some aspects of the published analysis.

A standard approach to vaccine efficacy analysis involves comparing risk ratios (the proportion of malaria in the intervention group over the control group), which is readily interpretable [5]. By this approach efficacy against clinical malaria in older children is more modest at 34% (intention to treat) or 36% (per protocol). Efficacy data on the younger age cohort is due next year, although the preliminary data indicate that protection against severe disease in 6-12 week olds is reduced (17% versus 45% in 5-17 month olds), and this is a concern.

The RTS,S analysis plan involved calculating the hazard ratio (for first episodes of malaria), the risk ratio (for severe disease), and the incidence rate ratio (for all episodes) [5]. Although the precise protective mechanism of RTS,S is unclear, it is thought to work by reducing the risk of infection from each exposure, rather than conferring “all or nothing” protection on a proportion of recipients [6]. By this hypothesis, everyone vaccinated should eventually experience malaria if transmission is high enough [5]. So the vaccine should have a greater effect on the incidence rate of the first or total episodes of clinical malaria than on the overall proportion of people experiencing it, a conclusion supported by the phase III data [2]. While analysis of incidence rates are completely valid, the risk ratio for clinical malaria provides additional highly relevant information to both policy makers and parents of immunised children, and we believe this should be reported.

References:

1. Whitty, C.J., The RTS,S malaria vaccine. BMJ, 2011. 343: p. d6986.

2. First Results of Phase 3 Trial of RTS,S/AS01 Malaria Vaccine in African Children. The New England journal of medicine, 2011.

3. White, N.J., A Vaccine for Malaria. The New England journal of medicine, 2011.

4. Butler, D., Malaria vaccine results face scrutiny. Nature, 2011. 478(7370): p. 439-40.

5. Lievens, M., et al., Statistical methodology for the evaluation of vaccine efficacy in a phase III multi-centre trial of the RTS, S/AS01 malaria vaccine in African children. Malaria journal, 2011. 10: p. 222.

6. Moorthy, V.S. and W.R. Ballou, Immunological mechanisms underlying protection mediated by RTS,S: a review of the available data. Malaria journal, 2009. 8: p. 312.

Competing interests: AVSH is a named inventor on patent applications covering malaria vectored vaccines and immunisation regimes

10 November 2011
Christopher J. Duncan
Research Fellow
Adrian VS Hill
The Jenner Institute
Oxford