Thread by @StabellBenn

A STORY ABOUT THE NEW #MALARIA #VACCINE.
We have written an analysis of the RTS,S malaria vaccine for @bmj_latest. In this thread, I summarise the history and describe the problems. I hope for a change of plans @WHO @WHOAFRO. bmj.com/content/368/bm…

In 2015, the RTS,S Clinical Trials Partnership published the results of a Phase-III trial of RTS,S malaria vaccine, including 6537 infants (6-12 wks) and 8922 children (5-17 mo). RTS,S gave 18% (infants)-36% (children) protection against clinical malaria. ncbi.nlm.nih.gov/pubmed/25913272

The authors wrote that there was no significant vaccine efficacy against allcause mortality". @EMA_News approved the vaccine. But based on the numbers in the appendix, we calculated that the vaccine was associated with 24% (95%CI=-3-58%) higher mortality. thelancet.com/journals/lance…

This was unexpected for a vaccine that reduced clinical malaria. We speculated that the higher mortality could be due to negative non-specific effects for girls, as has been seen for other non-live vaccines #NSEvac. We asked for the data by sex. Months later, we got access.

In 2016, we published that the vaccine was associated with a slight mortality reduction among boys, but girls had 2-fold higher mortality if they had received the new vaccine than if they had not. This was significant in its own right in both age groups. mbio.asm.org/content/7/2/e0…

There were additional danger signals: increased risk of meningitis and cerebral malaria. Nonetheless, @WHO decided to test the vaccine in a pilot implementation study among 720,000 African children, who should receive 4 malaria vaccines between 5-17 mo. who.int/wer/2016/WER91…

@WHO noted that the pilot study should evaluate the impact of the vaccine on child mortality, including the impact by sex, in addition to the impact on meningitis and cerebral malaria, also by sex. who.int/news-room/deta…

The pilot study was launched in Malawi in April 2019, Ghana and Kenya followed. @DrTedros stated that "The malaria vaccine has the potential to save tens of thousands of children’s lives”. Excited media praised the possibility to combat malaria. who.int/news-room/deta…

However, all available evidence suggests that if you are a girl, the vaccine may reduce your risk of malaria, but this benefit is offset by a negative non-specific effect #NSEvac (also seen for other non-live vaccines) that leads to increased OVERALL mortality.

The only justification for a large pilot implementation study would be if the increased mortality in girls was a chance finding. The probability is <0.001. But it could be. But given this situation, it is indispensable to follow children very closely for mortality.

Is this then being done? Is it ensured that every participant and each death in this pilot study is registered? Unfortunately no. The vaccine is added to the vaccination program in some regions, but not in others. There is no individual registration of participants.

Deaths are difficult to keep track of. E.g. in Malawi there are no official death registries, hospital staff have not routinely registered hospital deaths, and to report deaths outside hospital the study relies on town chiefs, many being illiterate. sciencemag.org/news/2019/11/f…

The latest development is worrying: The pilot study was to have 720,000 children and last 4 years to capture a negative effect of the vaccine on girls’ mortality. But now, after pressure from @GSK, WHO decided to make a decision already after 2 years. who.int/immunization/s…

This means that most study children will only have had 1-3 of the 4 vaccines before a decision is made. As we point out, the negative vaccine effects became worse with additional doses, e.g. girls had more than 3-fold higher mortality after more vaccines. bmj.com/content/bmj/36…

Adding to the problems, the first reports from the pilot study indicate that vaccine coverage is very low. So far in Malawi, the first dose reached about half of the children targeted, further diminishing the power to detect any mortality difference. nypost.com/2020/01/17/fir…?

The lack of participant registration, the potentially flawed death registration, the short follow-up and the low coverage all act in the same direction: making it difficult to document increased mortality among vaccinated girls, even if it should exist.

It is a real risk is that the pilot study will find no increased female mortality due to these shortcomings, and therefore the vaccine will be approved for use in Africa even though it has harmful non-specific effects for girls.

What will @WHO do to repair this situation?