At a family compound in rural Niger, a line of kids waited their turn for a taste of banana-flavored mystery medicine. One by one, health workers measured the height of each child against a notched stick and poured the corresponding dose of medicine into a plastic cup.
“They drank it down like milkshakes,” says Dr. Thomas Lietman, a research ophthalmologist at the University of California, San Francisco. “They’d stand in line twice, if you let them.”
These kids were among 227,000 children in Malawi, Niger, and Tanzania enrolled in a groundbreaking study known as MORDOR (“Mortality Reduction After Oral Azithromycin”). The goal of the study was to see if twice yearly doses of azithromycin could reduce childhood mortality in sub-Saharan Africa, where one in 19 children dies before reaching the age of five.
“The number of children enrolled in the study was enormous,” says Dr. Rasa Izadnegahdar, deputy director of global health at the Bill & Melinda Gates Foundation. “The team made over 700,000 child contacts in two years. This was probably one of the largest placebo-controlled studies ever in an African setting.”
The results, published in The New England Journal of Medicine, found that the rate of childhood mortality in the azithromycin group dropped by a combined 14 percent across the three countries.
Critics warn, however, that a result this good is sure to have downsides.
* * *
Many years ago, long before Lietman was traveling between family compounds in rural Niger, he worked in Ethiopia with an international team of ophthalmologists treating outbreaks of trachoma. Over time, they noticed a curious thing: the number of early childhood deaths seemed to drop in communities where the antibiotic azithromycin was used for treating the eye disease. To find out if the effect was true, the team ran a small study from 2006 to 2007.
“In the Ethiopia study, they confirmed that azithromycin can save lives when given to children under the age of five,” says Izadnegahdar. “But the study was too small and over too short a period of time to be meaningful for child health policy.”
In 2012, the Gates Foundation awarded a $12 million grant for Lietman’s team to conduct a large-scale study.
“At the foundation,” says Izadnegahdar, “we act as technical partners for a study, engaging in the design and solving challenges that come up.”
A shortcoming of the Ethiopia study was that country’s already prevalent use of azithromycin for treating trachoma, which might have muddled the study’s findings.
“For the new study, we wanted to disentangle the results from the background use of azithromycin,” Izadnegahdar says. “We looked for sites where there had been no trachoma program for the past five years. That way, we knew the children weren’t previously exposed to azithromycin.”
The sub-Saharan countries of Tanzania, Malawi, and Niger were ideal candidates because they included communities which had not received mass treatments for trachoma in the preceding five years and were relatively azithromycin-free. They also represented a range in levels of child mortality: low (Tanzania), moderate (Malawi), and high (Niger).
The next challenge was obtaining a large amount of the study drug Zithromax(r), manufactured by Pfizer, and the corresponding amount of placebo. As part of the 2012 London Declaration, a pledge made by pharmaceutical companies and other stakeholders to fight neglected tropical diseases, the company had committed to supporting this trial. That commitment created a pathway for the MORDOR team to apply for study drug through Pfizer’s Investigator Initiated Research program. They estimated the study would need 1 million doses, total. That amounted to 412,632 bottles, a volume of liquid that would fill 260 hotel-sized bathtubs.
“MORDOR was a huge study,” says Julie Jenson, Pfizer’s director of corporate responsibility. “On an annual basis, we were donating around 100 million doses a year of Zithromax to fight trachoma. So, we had a system in place to manufacture and ship the MORDOR team’s order.”
Delivering the Zithromax was achievable, but Pfizer hadn’t manufactured a placebo for the drug since the 1990s, when Zithromax itself was undergoing clinical trial. Pfizer contracted one of its leading pharmaceutical manufacturers in Italy to prepare an exact flavor match to the syrup used in Zithromax, minus azithromycin.
“We were trying to work quickly,” Jenson says, “because we knew the study had the potential of saving children's lives.”
The last hurdle was to design identical packaging for the placebo that made it indistinguishable to the researchers, health workers, and patients.
“Our Zithromax bottles used in the trachoma donation program have bright pink labels,” says Jenson. “But when we put on the placebo labels, the pink showed through and you could tell the difference between the two.”
To solve the problem, Pfizer printed pink labels to put on the bare bottles beneath the placebo labels to ensure the product would be indistinguishable. The drug and placebo were randomized and then shipped to Pfizer’s distribution center in Belgium where it was shipped to Tanzania, Malawi, and Niger for the studies. After 12 months, the MORDOR order was ready.
The clinical trial ran from 2015 to 2017, documenting 4,000 infant deaths – most of them in Niger.
“In these settings death is still, unfortunately, quite common,” Izadnegahdar says.
Breaking down the study’s results, the three nations had varying rates of improvement. In Tanzania, the childhood mortality rate dropped to 5.4 deaths per 1,000 person-years, in Malawi to 9.1 deaths, and in Niger to 22.5 deaths.
MORDOR has been met with measured excitement in the global health community. The results were discussed at the World Health Assembly at a meeting of the West African Health Organization, and presented twice to the cabinet in Mali.
“It’s complicated taking something from the realm of a controlled study into a public health program,” Izadnegahdar says. “We’re now in the phase of asking: Where do we go next? What is the foundation’s role?”
The answer: understanding how the MORDOR effect works, and what the larger impacts might be. A leading concern was that the mass administration of azithromycin might create antimicrobial resistance in children.
“It goes against dogma at the moment because everyone else is trying to reduce antibiotic use,” Per Ashorn told Nature. Ashorn is a specialist in pediatric infectious diseases at the World Health Organization (WHO).
When the MORDOR study concluded, the Gates Foundation had a provisional grant ready to fund an immediate follow-up trial in Niger, the country where azithromycin had the largest impact. It continued twice-a-year azithromycin doses for children in the study area, including those from the original placebo group, creating an extended four-year study. This study was also extended into neighboring Burkina Faso, where childhood mortality rates were similarly high.
Preliminary data from MORDOR trial sites indicates low rates of resistance, which aligns with data showing lower overall resistance rates in sub-Saharan Africa.
“Resistance to an antibiotic may be driven up according to the size of dose given at the community level,” Izadnegahdar says. “By treating just the children between the ages of one and 59 months, the community was given a much smaller dose than community-wide trachoma programs which treat all community members above the age of six months. This level of exposure in a limited window of time may not be enough to drive resistance.”
Which brings up the central paradox of the study. Did MORDOR work because the background use of antibiotics was so low? Regular doses of any antibiotic would, as Lietman puts it, “pick the lowest-hanging fruit of infectious diseases.” MORDOR was, in effect, a mass drug administration. It is logical that childhood survival rates would improve. But at what cost?
“Azithromycin is incredible,” says Dr. Chris Damman, a senior program officer and lead on gut health at the Gates Foundation. “In a risk-benefit analysis, the risks are theoretical and the benefit is real. But we have to remain agnostic to the mechanism of azithromycin. We don’t know whether it works by decreasing pathogens or by increasing the body’s own barrier against infection.”
Simultaneous to MORDOR, the Gates Foundation has been supporting research projects around the world focused on understanding the impacts of azithromycin use.
“What’s the optimal package of interventions?” says Izadnegahdar. “Maybe it’s not azithromycin alone, but also some modulation of the gut microbiome. Once we understand the mechanism of action, we can develop a next-generation azithromycin without the blunt effects of using an antibiotic on the microbiome or causing antimicrobial resistance.”
This much is clear: the MORDOR effect is a powerful tool for combating infant mortality.
“We’re working closely with the WHO to have a policy and recommendation in place by the end of the year,” says Izadnegahdar. “This could be an important intervention with significant impact in high mortality setting.”
Read next
Looking back on vaccines in 2018
