New research conducted by British and African scientists suggests that mass azithromycin distribution is linked to persistent, elevated macrolide resistance in one of the communities where it's been tested.
In a study published yesterday in The Lancet Infectious Diseases, a team led by researchers with the Malawi Liverpool Wellcome Research Programme and University College London reported that macrolide resistance in Streptococcus pneumoniae remained elevated in children in a community in Malawi more than 3 years after they received twice-yearly doses of azithromycin—a macrolide antibiotic—as part of a randomized controlled trial. Furthermore, macrolide-resistant S pneumoniae was also elevated in the children from the village who received placebo and in those born after the trial was conducted.
That trial, dubbed MORDOR (Macrolides Oraux pour Réduire les Décès avec un Oeil sur la Résistance), was conducted in rural villages in Malawi, Niger, and Tanzania to assess the impact of mass drug administration (MDA) with azithromycin on childhood mortality. The results showed a 13.5% reduction in all-cause mortality among children aged 1 to 59 months who received the antibiotic compared with those who received placebo, with the largest impact seen in younger children.
Based on those results and subsequent trials, the World Health Organization (WHO) in 2020 conditionally recommended MDA for children ages 1 to 11 months in high-mortality regions. Since then, MDA programs have expanded in sub-Saharan Africa.
But there has long been concern about the potential for promoting macrolide resistance, particularly in S pneumoniae, which is a leading cause of pneumonia, sepsis, and meningitis in children in the region. The authors of the current study say their findings highlight the need for ongoing antimicrobial resistance (AMR) surveillance in MDA programs.
"Without timely detection and intervention, these trends could become difficult to reverse," they warn.
Elevated resistance in azithromycin- and placebo-treated kids
For the follow-up study, researchers collected and performed whole-genome sequencing on nasal swabs from children ages 1 to 9 years old in Mangochi, Malawi, in 2021—3.5 years after the MORDOR trial.
The participants included children ages 4 through 9 who lived in azithromycin- or placebo-treated clusters in the community, and children ages 1 through 3 who lived in one of the clusters but were born after the MORDOR trial. (The authors note that, 6 months after the final post-MDA survey of the MORDOR trial, all eligible children in the placebo-treated clusters received one round of azithromycin.)
The primary outcome was the prevalence of S pneumoniae macrolide resistance in azithromycin- and placebo-treated clusters at baseline, 6 months post-MDA, and 3.5 years post-MDA.
Of the 924 children screened for the study, researchers analyzed isolates from 905, including 452 from azithromycin-treated clusters and 453 from placebo-treated clusters. Of the children, 451 were living in a cluster during the MORDOR trial (MORDOR-exposed) and 454 were born after the end of the trial (MORDOR-unexposed). For comparison, the researchers also analyzed sequenced pneumococcal genomes from a different community in Malawi that had not been part of the trial (non-MDA).
In the children in azithromycin-treated clusters, genotypic macrolide resistance in S pneumoniae increased from 21.7% at baseline to 31.9% 6 months post-MDA and 32.5% 3.5 years post-MDA. In the placebo-treated clusters, macrolide resistance showed similar trends, rising from 21.0% at baseline to 25.0% 6 months post-MDA and to 30.9% 3.5 years post-MDA. No significant differences were observed in odds ratios between the treatment groups across the timepoints: 0.97 (95% confidence interval [CI], 0.36 to 2. 55) at baseline, 1.46 (95% CI, 0.67 to 3.17) at 6 months post-MDA, and 1.12 (95% CI, 0.66 to 1.91) at 3.5 years post-MDA.
Without timely detection and intervention, these trends could become difficult to reverse.
"Overall, macrolide resistance did not return to baseline levels in the azithromycin-treated clusters, spilled over into the placebo-treated clusters, and further increased after a round of MDA in the placebo clusters after the MORDOR trial," the authors wrote.
Macrolide resistance in both treatment clusters in Mangochi was notably higher than at the non-MDA site, where it remained stable (16.9% at baseline, 16.5% at 6 months post-MDA, and 16.5% at 2.5 years post-MDA). And, among 114 children born into azithromycin-treated clusters after MDA, macrolide resistance was 36%.
The researchers also found that multidrug resistance (defined as resistance to at least three antibiotic classes) in S pneumoniae was significantly higher in azithromycin-treated (41.1%) and placebo-treated (43.2%) clusters at 3.5 years post-MDA compared with non-MDA sites (22.8%) at 2.5 years. Further genomic analysis suggests the macrolide-resistant and multidrug-resistant S pneumoniae strains are associated with transmissible genetic elements and likely emerged and expanded after the MORDOR trial.
"Together, these findings suggest that AMR resulting from MDA could persist and spread through a population, facilitated by emergent AMR pneumococcal lineages and transmissible elements," they wrote.
Complex relationship between exposure, resistance
The authors say the observed persistence of macrolide resistance following MDA contradicts previous research on MDA programs for trachoma that suggested resistance might dissipate once antibiotic pressure has been removed, indicating a "complex relationship between antimicrobial exposure and resistance." And while they didn't find that resistant S pneumoniae strains had spread to non-MDA sites, they warn that these strains could pose a regional and global health threat if left unmonitored.
The authors add that the data underscore the potential cumulative effect of MDA exposure in regions like sub-Saharan Africa, where AMR is already rising.
"Antimicrobial stewardship, particularly in the context of MDA programmes, is essential to balance the life-saving but potentially short-term benefits of MDA against the long-term risks of AMR," they wrote. "Macrolide resistance has not reverted to pre-exposure levels, emphasising the need for ongoing surveillance to establish whether this resistance has become permanently fixed in the population."
