A closeup of the malaria vaccine in a syringe.

Mosquirix is being introduced only in selected areas in Malawi, Kenya, and Ghana. A decision about a broader rollout is expected in 2021.

Thoko Chikondi

Read our COVID-19 research and news.

First malaria vaccine rolled out in Africa—despite limited efficacy and nagging safety concerns

MALAWI—In a small room at the Phalula Health Centre in southern Malawi's Balaka district, two young mothers are sitting on a wooden bench, each with a 5-month-old baby on their lap. Across from them, behind a desk, sits Alfred Kaponya, a community health worker. A colleague is busy preparing a vaccine, tapping the syringe to dislodge bubbles. Kaponya explains the procedure to the women, writes down the vaccines' serial numbers in the children's vaccination booklets, and copies them onto a spreadsheet in his binder.

Then, one of the mothers bares her son's thigh for the shot; he starts to cry, and she strokes his back. The procedure is repeated for the other baby, a girl.

It may sound routine, but it's not. These two children have just received the first malaria vaccine to move beyond the stage of clinical testing—a landmark event in the battle against a disease that each year takes more than 400,000 lives, most of them children in Africa. Thirty years in the making, RTS,S, also known by its brand name, Mosquirix, targets Plasmodium falciparum, the most common and most lethal of four malaria parasite species. It is an answer to a dire need. After decades of declining numbers of cases and deaths, the fight against malaria has stalled. Parasites resistant to the most widely used treatment, called artemisinin-based combination therapy, are spreading, while malaria mosquitoes are increasingly resistant to insecticides.

And yet the rollout, here and in two other African countries, isn't quite the breakthrough the field has been waiting for. Mosquirix's efficacy and durability are mediocre: Four doses offer only 30% protection against severe malaria, for no more than 4 years. Some experts question whether that is worth the cost and effort. 

The biggest concerns, however, are about the vaccine's safety. In the largest trial, children who received Mosquirix had a risk of meningitis 10 times higher than those who received a control vaccine. Mosquirix may not have triggered the meningitis cases—there are other possible explanations—but the possible risk worried the global health community so much that, rather than rolling out the vaccine across Africa, the World Health Organization (WHO) has decided to set up a pilot in Malawi, Ghana, and Kenya in which the vaccine will be given to hundreds of thousands of children.

Many ways to thwart a parasite

Malaria vaccines in development interfere with many different parts of the parasite’s complex life cycle. Mosquirix targets so-called sporozoites after they are injected by a mosquito and before they move to the liver.

Salivary gland Zygote Gamete Gametocytes Sporozoites In humans In mosquitoes Plasmodium falciparum Vaccine Transmission-blocking vaccines would be given to humans but are designed to stop the parasite from developing in the mosquito. Vaccine Blood stage vaccines target the parasite during replication in red blood cells. Vaccine Liver-stage vaccines would prevent malaria parasites from re-entering the bloodstream. Vaccine Mosquirix and some other candidate vaccines attack sporozoites. Merozoites Parasite Ookinete Male and female Male and female Parasites Mosquito injects Parasites Ookinete ’s life cycle 7 6 5 4 3 2 1 sporozoites into bloodstream. develops in mosquito’s midgut. replicate in liver, form merozoites. replicate inside red blood cells that burst. Symptoms first develop. travels to salivary glands, forming sporozoites. gametocytes merge, forming zygote. gametocytes form, ready to be picked up by another mosquito.
N. Desai/Science

The pilot is not a clinical trial, but a closely monitored vaccination program to collect more data to make sure Mosquirix is safe and effective before wider introduction. "I think the pilot is a scientific and pragmatic way to move forward," says Marcel Tanner, a former director of the Swiss Tropical and Public Health Institute in Basel who has been involved in several clinical studies with the vaccine. "It is a way to monitor all the aspects of the vaccine and watch if something happens."

But monitoring the vaccinated children is a daunting task in a country like Malawi, which lacks digital systems to record health and mortality statistics. When two WHO panels recommended setting up the pilot, "There was no detailed discussion of the mechanics of doing this," says epidemiologist Peter Smith of the London School of Hygiene & Tropical Medicine (LSHTM), a scientific adviser to the pilot. And critics are still skeptical the pilot will deliver a clear verdict about Mosquirix's potential in the real world.

THE MALARIA PARASITE is a challenging target for a vaccine. It has a complex life cycle that begins when an infected female mosquito bites a human and spits Plasmodium cells called sporozoites into the bloodstream. They multiply in the liver, emerge as another cell type named merozoites, invade red blood cells, and continue to multiply. The blood cells burst, causing fever, headache, chills, muscle aches, and often anemia. (They also flood the blood with gametocytes—the parasite's reproductive cells—ready to be picked up by the next mosquito.) Along the way, the parasite frequently changes its surface proteins. That makes it an elusive target for the immune system, and for a vaccine.

Mosquirix, developed in the 1980s by a team in Belgium at SmithKline-RIT, now part of GlaxoSmithKline (GSK), stimulates an immune response against a protein that occurs only on the sporozoites' surface. To bolster the response, the research team fused the vaccine protein with a hepatitis B surface protein and added an adjuvant. "Many people were very skeptical at the time, because there had been so many attempts done and so many failures," says molecular biologist Joe Cohen, who led the effort until he retired from GSK in 2012.

But results of the first large trial, among 2000 children aged 1 to 4 in Mozambique, were promising: Malaria infections during the first 6 months after vaccination were down 58%, Cohen and his colleagues reported in a 2004 paper in The Lancet. An even larger trial enrolled 15,000 children in seven African countries between 2009 and 2011 and had mixed results. Among babies who received their first shot between 6 and 12 weeks of age, protection against severe malaria was close to zero. That dashed hopes that the vaccine could protect the most vulnerable group and that it could be given together with other routine vaccines for young babies, Smith says. But among children first vaccinated between 17 months and 25 months of age, malaria infections overall were down by 40%, and severe infections by about 30%.

Those are paltry numbers compared with the measles vaccine, which is 97.5% protective. "But nobody really expected to get very high efficacy because malaria is such a complicated disease," Smith says. The trials also revealed that, even after three shots given 1 month apart, protection drops to near zero in about a year and a half. A booster shot given at that point bolsters the protective effect, but it tapers off again in another 18 months. Still, even a short-lived, partly effective vaccine "could make a huge difference," Smith says. Cohen notes that it offers some protection during a crucial period: "The real period during which children are at risk of severe disease and death is under 5 years old."

The European Medicines Agency (EMA) agreed. In July 2015, it declared the vaccine was safe and effective enough to be introduced in Africa, under a special procedure designed to help regulatory agencies in developing countries make decisions.

The long road to a lackluster vaccine

Development of Mosquirix began in the 1980s. A pilot rollout now underway will help determine whether it should be introduced widely in Africa in the next decade.

1984 1997 2005 2010 2015 2020 2025 2015 1984–1997 2004 2009–2014 2015 2016 2017 2019 2021 2021 2023 Three-country pilot completed, including data on feasibility and efficacy WHO expected to decide on wider introduction First data expected on mortality and safety, including meningitis risk Pilot begins in three countries. WHO recommends phased introduction in selected areas Expert meeting in Geneva, Switzerland, recommends pilot implementation projects Study reports 58% protection against severe malaria in young children European Medicines Agency issues positive opinion Early development at SmithKline- RIT (now GSK) in Belgium Large trial finds moderate protection at ages 5–17 months; none at 6–12 weeks Ghana, Kenya, and Malawi announced as pilot countries
(Scroll right to see more.)

That didn't end the doubts. Some scientists feared the vaccine would be less effective in the real world than in studies, says LSHTM malaria researcher Brian Greenwood, who led the trial, because people might count on the vaccine for protection and become less careful. "What if people no longer went to sleep under their mosquito nets or parents no longer brought their feverish child to a clinic for a malaria test? That would have to be monitored," he says. Others worried that the money spent on Mosquirix would come at the expense of other malaria prevention methods. A full series of four shots costs about $20, versus $5 for a mosquito net and $1.50 to give a child malaria drugs prophylactically, during the rainy season. "There are effective measures available that we can use even better," says Micaela Serafini of Doctors Without Borders in Geneva, Switzerland.

The safety data, meanwhile, were puzzling. About 20 of the 6000 children vaccinated against malaria in the seven-country trial contracted meningitis, compared with one of 3000 children in the control group, who received a rabies vaccine—a 10-fold increase in risk. One-third of the children with meningitis died. The research team and EMA both think this outcome was a fluke. The meningitis episodes occurred at random intervals after vaccination, most of them at two of the seven study sites, and there is no good explanation for why Mosquirix would lead to meningitis, the authors of the Lancet paper argue. Other scientists have suggested the difference may be due to the rabies vaccine that control children received. That group had a remarkably low rate of meningitis—suggesting the rabies vaccine may somehow prevent it.

THE DEBATES ABOUT MOSQUIRIX came to a head at a meeting at WHO headquarters in Geneva in October 2015. Proponents of rolling out the vaccine, including many African representatives, argued that an imperfect vaccine was better than none. Others, mainly vaccine experts, argued that Mosquirix just wasn't safe and effective enough for introduction.

A relative outsider at the meeting, Danish anthropologist and vaccine researcher Peter Aaby of the Bandim Health Project in Guinea-Bissau, offered another argument against introduction. After reanalyzing the data from the biggest trial, Aaby discovered that although the vaccinated children had malaria less often, they did not die less often. Among girls, overall mortality was almost doubled, Aaby told his colleagues at the meeting. "This vaccine is killing girls," he recalls saying. Whereas WHO expects the vaccine to save one life per 200 children vaccinated, Aaby believes one in 200 will die as a result of it; he predicts "a nightmare."

The community has welcomed the vaccine.

Alfred Kaponya, Phalula Health Centre

Aaby and Christine Stabell Benn, a global health professor at the University of Southern Denmark, have an explanation. The married couple has studied routine vaccinations in Africa for decades and believes vaccines can "train" the immune system in ways that don't affect just the target disease. Vaccines that contain a living, weakened pathogen—such as the vaccines against measles and tuberculosis—strengthen the immune system generally, Aaby and Stabell Benn say, making recipients better able to fight off other infections. But vaccines that contain a killed pathogen or only bits of it weaken the immune system, their theory goes—especially in girls, because their immune systems seem to respond more strongly to vaccines in general.

Few share Aaby's concerns about Mosquirix. WHO's Mary Hamel, who leads the vaccine pilot from Geneva, says the trial was not designed to study mortality. Just being enrolled in the trial meant children received better care, and their mortality was 70% lower than among children near the study sites who weren't enrolled. “The surprise,” Hamel says, was the low mortality in girls who received the rabies vaccine. Tanner says Aaby has "enriched our field" but "has turned a bit into a missionary with regard to the nonspecific immune effects. I am confident that the vaccine will have an overall positive effect on mortality." Yet Aaby's presentation at the meeting intensified the doubts. At one point, it appeared the vaccine might be abandoned altogether, Greenwood says.

To address the concerns, the attendees agreed to the three-country pilot. In each country, the WHO team and the national government would randomly select areas where about 120,000 children annually would receive the vaccine between 2019 and 2022. Researchers would monitor how well the rollout went and compare rates of malaria, meningitis, and other diseases, as well as mortality, in vaccinated and control areas.

MALAWI WOULD WELCOME an effective vaccine. It has seen progress in the fight against malaria—primarily thanks to mosquito nets, rapid diagnostics, and improved access to treatment. Ten years ago, about 60% of children were infected at any given moment, versus 17% today. But Don Mathanga, director of the Malaria Alert Centre at the University of Malawi in Blantyre, says the current infection rate is "not a figure that we are very content with." The death toll was about 7000 in 2017 and, as in many other countries, the decline has stalled. "So we were very happy to be selected as a pilot country," Mathanga says.

Emily Phadzula from Bereu, Malawi, holds her daughter Margaret, who has received three doses of Mosquirix.

Thoko Chikondi

So are many parents. At the Queen Elizabeth Central Hospital in Blantyre, Jaquiline Masomba, 32, is dabbing the head of her 3-year-old son Edson with a washcloth. He probably contracted malaria from a mosquito that found its way through a hole in her worn-out bed net, she says. "For our living I depend on alcohol distillation," Masomba says. "And my business has been on hold for 2 weeks" as she cares for Edson. Masomba had never heard about the vaccine, but now says: "Of course I would like to have it when it becomes available."

Violet Wilson, 26, in the southern village of Nkwazi, explains that she tries everything to keep her 5-month-old son from becoming infected. Cracks and holes in the walls of her brick house are smeared with loam to keep mosquitoes out. Wilson and her son sleep under a mosquito net every night. If her baby gets a fever, she'll rush him to a clinic, Wilson says. She was delighted to hear about the arrival of the vaccine, but her son can't get it because they live in a control area.

Emily Phadzula, 30, from the nearby town of Bereu, was luckier: In late April, her daughter Margaret, then 5 months old, was the first child in her village to be vaccinated. Sitting on the doorstep of her concrete house and tightly holding Margaret, who is wearing a pink sweater, Phadzula explains she heard about the vaccine during a community meeting. "I didn't have any doubts," she said. "I was very happy that after having to see my other children suffer so much, finally this child could be better protected." Margaret has had no side effects except a slight fever—and no malaria. Phadzula says she recommends the shots to other mothers.

By September, almost 35,000 children in Malawi had received at least one shot. Getting the vaccine to these children has been straightforward, but collecting follow-up information is a challenge. "In trials, you have a controlled environment. You know where the participants are, keep track of them, if there is an issue you take care of them," says Bernhards Ogutu, a pediatrician and malaria researcher at the University of Nairobi who is involved in Kenya's vaccine pilot program. "But in the pilot, once you've given the child the vaccine, they go home."

Health worker Dennis Nkuma opens a vaccine registry book in a health center in southern Malawi.

Thoko Chikondi

As a first step to monitoring them, researchers in Malawi have set up a surveillance system in four "sentinel hospitals" that treat both vaccinated and unvaccinated children. Comparing data about illness and mortality should show the vaccine's impact, says pediatrician Tisu Mvalo of the University of North Carolina Project–Malawi in Lilongwe. "We had to train hospital staff to collect these data. We hope that the numbers will be big enough to detect potential differences."

Keeping track of children who die outside the hospitals is even harder. "Unfortunately, we still have no death registration in this country," Mathanga says. Instead, the team built a new system based on a cultural practice: In rural Malawi, town chiefs must allocate a place for the dead to be buried. The researchers figured they would know which children have died. Many chiefs are illiterate, however, so they had to be assigned an assistant. Pilot evaluation staff use motorbikes to visit the chiefs regularly and collect the paper files; they also interview relatives of deceased children to identify the most likely cause of death. But these "verbal autopsies" aren't always correct.

MEANWHILE, the pressure to deliver results is rising. In April, WHO's advisory committees said they would assess any impact of Mosquirix on meningitis, severe malaria, and mortality—and determine whether the vaccine can be introduced in the rest of Africa—after just 2 years, instead of the four originally planned. "The evaluation will continue, but we don't want to keep the rest of Africa waiting any longer," says David Schellenberg, who left LSHTM in 2016 to join the team at WHO. Another reason to hasten the decision is that GSK needs to know whether it can continue production, Schellenberg says.

Whether the researchers will have enough data to make a decision remains to be seen, however. In Kenya, the pilot was delayed by 4 months, apparently because the government was overwhelmed by the implementation of universal health care. In Ghana, an advocacy group called the Coalition for Ghana's Independence Now called for a boycott of the vaccine in May, asking for an injury compensation program similar to the one in the United States, to help those who might be harmed by Mosquirix. That, too, might cause delays. What's more, a declining number of meningitis cases and deaths in the three countries—although good news—may mean researchers need to continue the pilot longer than expected to detect statistically significant differences.

Aaby says showing whether the vaccine increases mortality among girls might also take longer than planned. He worries that even without a clear safety verdict, the global health community will exert enormous pressure to get the vaccine approved around Africa. But the WHO team in Geneva remains confident it will know whether the benefits are worth any risks. "We have a lot of checks on the evaluation. So, this setup should do it," Hamel says.

Many people in Malawi have no doubts. "The community has welcomed the vaccine," Kaponya says, after he closes the door behind the two women and their newly vaccinated babies. "The only challenge we are having here, is that people from the nonvaccine areas come and ask for it."

With reporting by Saulos Jali in Malawi. This story was supported by the European Journalism Fund.

*Correction, 4 December, 2:40 p.m.: Quotes by Peter Smith and Mary Hamel have been corrected in this story, along with a few details about the main trial of Mosqurix.