Intended for healthcare professionals

Rapid response to:

Feature Informed Consent

WHO’s malaria vaccine study represents a “serious breach of international ethical standards”

BMJ 2020; 368 doi: (Published 26 February 2020) Cite this as: BMJ 2020;368:m734

Linked Analysis

WHO’s rollout of malaria vaccine in Africa: can safety questions be answered after only 24 months?

Rapid Response:

The WHO Malaria Vaccine Implementation Program: the lack of informed consent is not the only ethical challenge

Dear Editor

In WHO’s response to Peter Doshi’s feature article, the WHO authors emphasize that the RTS,S malaria vaccine has an acceptable safety profile, is not an “experimental vaccine” and is undergoing “phased introduction”. These statements seem at odds not only with prior WHO statements, like “Other questions that should be addressed as part of pilot implementations include the extent to which RTS,S/AS01 vaccination impacts all-cause mortality (including gender-specific mortality)”(1), but also what EMA writes about the pilot implementation study: “The study will address the following safety concerns listed in the safety specification: febrile convulsions, meningitis, rebound effect, cerebral malaria, gender-specific mortality and vaccine effectiveness/impact. It will also evaluate the feasibility of implementing a fourth dose” (2).

If the task is to evaluate whether the 2-fold higher female mortality observed in the Phase III trials (3) represents a real phenomenon, then it should be carried out with due diligence. However, the pilot implementation study is designed in ways that make it possible to overlook if the RTS,S vaccine truly increases female mortality.

Despite numerous request WHO has not provided the protocol for the pilot implementation study, but based on what has been written in other documents, the pilot implementation study suffers from the following design weaknesses:

1) The study does not register individual participants. Hence, there is no follow-up of neither vaccine exposure nor outcome at an individual child level in the areas receiving or not receiving RTS,S vaccine;

2) Deaths are captured through village key informants. Such a system is unlikely to be perfect. Hence, mortality is likely to be underestimated;

3) The study does not have exact information about the number of children living in the areas receiving or not receiving RTS,S. Hence, nobody knows the exact denominator when calculating mortality risks;

4) The RTS,S vaccine coverage in the pilot study is now reported to be very low, down to 25% with the 3rd dose (4). If the study is comparing areas where <25% received the vaccine with areas where nobody received the vaccine, a real negative effect on female mortality would be diluted;

5) WHO has now changed the original plan to follow participants for 4-5 years for mortality and decided to make a preliminary assessment of the effect on mortality based on a surrogate outcome, severe malaria, after 24 months. There are two major problems with that decision: first, severe malaria was not associated with overall mortality in the phase III trials (5) and severe malaria can therefore not serve as a marker of overall mortality; second, as the negative effects or RTS,S in females increased with increasing follow-up time, an assessment 24 months after study start will limit the possibility of observing negative effects (5).

These five design decisions work in the same direction: they increase the likelihood of overlooking a true negative effect of RTS,S on female mortality.

The worrying prospect is that based on flawed data, the WHO may wrongly conclude after 24 months that there was no safety concern in relation to female mortality (measured as severe malaria), and then roll-out the vaccine in the unvaccinated regions of Ghana, Kenya and Malawi and the rest of Sub-Saharan Africa. Once rolled out, we may only discover potential harms after years of accumulation, since very few places have routine data collection to capture an increase in female mortality.

In the Phase III trials, RTS,S was associated with 2-fold higher female mortality between 5 months and 4-5 years of age (3). If this is a true finding and if it is overlooked in the pilot implementation study, and the vaccine is then rolled out over all of Sub-Saharan Africa, it may result in half a million excess female deaths/year if the vaccine coverage reaches 70% (assumptions from UNICEF’s State of the World’s children 2019: females vaccinated/year=18 million; female mortality from 5 months-5 years=40/1000).

According to us, the pilot implementation study is not a “phased introduction” but a very important safety research project. We hope the WHO will share the study protocol and make it possible to suggest if and how the study can reliably assess sex-differences in mortality.

1. World Health Organization. Malaria vaccine: WHO position paper, January 2016 - Recommendations. Vaccine. 2018;36(25):3576-7.
2. European Medicines Agency. Summary of risk management plan for Mosquirix (Plasmodium falciparumand hepatitis B vaccine (recombinant, adjuvanted)). [Available from: Accessed 9 March 2020.
3. Klein SL, Shann F, Moss WJ, Benn CS, Aaby P. RTS,S Malaria Vaccine and Increased Mortality in Girls. MBio. 2016;7(2).
4. Associated Press. First ever malaria vaccine tried out in babies in 3 African nations. New York Post. 2020 January 17, 2020. Accessed 9 March 2020.
5. Aaby P, Fisker AB, Bjorkman A, Benn CS. WHO's rollout of malaria vaccine in Africa: can safety questions be answered after only 24 months? BMJ. 2020;368:l6920.

Competing interests: Christine S. Benn was quoted in the article that WHO is criticizing.

10 March 2020
Christine S. Benn
Ane B. Fisker, Anders Björkman, Peter Aaby
University of Southern Denmark
Studiestræde 6, 1455 Copenhagen K, Denmark