Roles of community and sexual contacts as drivers of clade I mpox outbreaks

Abstract

Initial investigation into the emerging mpox outbreak of novel clade Ib in eastern Democratic Republic of the Congo has identified signs of sustained human-to-human transmission and epidemiological links to sexual contacts involving female sex workers (FSWs)¹, which have not been observed in previous clade Ia outbreaks. Using mathematical models incorporating age-dependent contact patterns, we quantified the role of frequent sexual interactions as opposed to community contacts in clade Ib's dynamics and found that this additional mode of transmission could explain its increased outbreak potential compared with clade Ia. As with the globally-circulating clade IIb, transmitted predominantly among men who have sex with men², our findings reinforce the importance of protecting key population groups, specifically FSWs for clade Ib, in controlling ongoing mpox outbreaks.

Figure 1.. Observed and modelled age distributions of mpox cases.

(A) Clade Ia cases with suspected human-to-human exposure from the Tshuapa province, DRC, 2011–2015. Those aged 34 and older were assumed to be smallpox-immunised. (B) Cases from clade Ia-endemic provinces in DRC, January–mid August 2024. Those aged 45 and older were assumed to be immunised. (C)–(F) Clade Ib cases from multiple settings in DRC, January–mid August 2024 and Burundi, mid September–October 2024. Those aged 45 and older were assumed to be immunised. Half-transparent lines represent the model fit without assuming transmission over sexual contact. Dots and whiskers represent the observed age distributions among cases and their 95% confidence intervals.

Figure 2.. Transmission dynamics of MPXV clade I outbreaks.

(A) Age-specific transmission patterns inferred for clade Ia in endemic provinces of DRC and clade Ib in North and South Kivu, and Burundi. Colours represent daily transmission rates between infector-infectee pairs in given age groups according to the estimated next generation matrices. (B) Projected effective reproduction number $\left(R_y\right)$ for clades Ia and Ib in DRC (blue lines) and Burundi (red lines), 2010–2030. Lines and shaded areas represent median estimates and 95% credible intervals of $R_y$. The estimated reproduction number of 0.82 from clade Ia case data between 2013–2017 in Tshuapa province, DRC was used as a reference value for 2015 (the midpoint of the study period). The empirical estimates of the effective reproduction number from the DRC incidence data in the pre-clade Ib period in 2023 (pre-Ib) and from the South Kivu incidence data (SK) in 2024, as proxy data for clade Ia and Ib, respectively, are displayed as dots for comparison. Paired dots (jittered horizontally for visual aid) denote different serial interval distributions used, which were estimated elsewhere from two separate sets of infector-infectee pairs (distribution 1: mean 17.5 days; distribution 2: mean 11.4 days). (C) Contour plots for estimated impact of different combinations of FSW-focused and mass vaccination strategies on the effective reproduction number in the Kivus.