Summary

A medRxiv preprint uses an agent-based model to link mpox clade IIb’s persistence to asynchronous transmission across connected global sexual-contact networks. The model estimates that local extinction became much less likely when transmission continued in other regions and later re-entered Europe.

A preprint study suggests that mpox clade IIb persisted after its large 2022 outbreak not because transmission remained continuously intense in one region, but because connected populations experienced outbreaks at different times. Transmission continued in parts of Asia after cases had declined in Europe and the Americas, allowing immunity in Europe to wane before the virus was reintroduced.

The researchers used an agent-based model of mpox transmission on a sexual-contact network representing men who have sex with men (MSM) in Berlin. The model included viral shedding, vaccination, immunity acquired through infection, waning immunity and imported cases. It was then extended to represent connected European and global MSM populations.

Local extinction became less likely as the network expanded

For the Berlin model, the estimated probability that mpox would become extinct exceeded 80% in 2023. The authors attribute this high probability to acquired immunity fragmenting the transmission network: enough people who had been susceptible or highly exposed were no longer available to sustain continuous chains of infection.

That result changed when connections beyond Berlin were added. In a European MSM metapopulation, the estimated extinction probability fell below 60%. When global metapopulation dynamics were included, the probability became nearly zero in the model.

The difference reflects the importance of importation. People with mpox are infectious for approximately two to four weeks, so a continuing chain of transmission requires sufficient infections to occur within that period. Once local transmission declines, isolated outbreaks may be unable to maintain themselves. Repeated introductions from another population can restart transmission, particularly if immunity has partially declined or the local network again contains enough susceptible people.

The study calls this pattern epidemic asynchrony: outbreaks rise and fall at different times in different regions. In the model, transmission in Asia continued after transmission had fallen in Europe and the Americas. That interval was long enough for immune waning to restore some transmission potential in Europe, after which re-importation could support continued circulation.

Phylogenetic patterns support the model’s explanation

The authors compared their model-based interpretation with phylogenetic evidence from Germany. They report that all active German transmission clusters were linked either to re-importation of clade F from the Americas or to the emergence of clades C and E associated with Asia.

This combination of modelling and viral-lineage evidence points to a global rather than purely local explanation for persistence. A region can appear to have brought transmission under control while remaining connected to populations where infections are still occurring. Surveillance and prevention efforts that focus only on one country or region may therefore miss the conditions that enable later reintroduction.

The findings come from a computational epidemiology study, not a clinical trial or an intervention study. The researchers used publicly available, aggregated surveillance counts and recruited no new human participants. The extinction probabilities are outputs of a calibrated model and depend on how it represents sexual-contact networks, viral shedding, vaccination, infection-derived immunity, immune waning and case importations. The work is also presented as a medRxiv preprint, so its conclusions are available before formal peer review.

The central public-health implication is that coordinated international surveillance and control may be important even after local case numbers have fallen sharply. Tracking transmission across connected populations can help distinguish genuine interruption from a temporary decline followed by reintroduction.

Sources