HIV Co-Infection Can Keep Mpox Alive Even Below the Outbreak Threshold, Model Shows

When mpox, formerly known as monkeypox, spread across the globe in 2022, public health authorities were confronted with an epidemiological pattern that none of the existing single-disease models had anticipated. The virus, long considered endemic in parts of Central and West Africa with occasional spill-over events, suddenly established sustained human-to-human transmission, with sexual contact emerging as the dominant route of spread. The outbreak disproportionately affected men who have sex with men, a community in which HIV prevalence is substantially elevated compared with the general population. That overlap was not merely a demographic coincidence; it raised a fundamental question about how two pathogens interacting within the same host, and within the same transmission network, might reshape the dynamics of an epidemic. A new mathematical study published in PLOS Complex Systems offers the first mechanistic answer, and its conclusions carry uncomfortable implications for how outbreaks are managed in HIV-endemic settings.

The research team, led by Hailu Tkue Welu and colleagues, constructed a co-infection model that captures the bidirectional interplay between HIV and mpox. Unlike conventional models that treat each pathogen in isolation, this framework incorporates three innovations that the authors argue are essential for reproducing what actually happened in 2022. First, it accounts for the way HIV-induced immunological suppression alters the progression of mpox within a co-infected individual. Second, it allows transmission rates to depend on antiretroviral therapy status, recognizing that people whose HIV is well suppressed on treatment are biologically and behaviorally different from those who are untreated. Third, it stratifies vaccination by HIV status, acknowledging that the immune response to the smallpox-based vaccines deployed against mpox may differ dramatically depending on the state of a person’s immune system.

The most striking finding to emerge from the analysis is the presence of what epidemiologists call a backward bifurcation, a phenomenon entirely absent from single-pathogen models of mpox. In classical epidemic theory, a disease dies out when its basic reproduction number, R0, falls below one; each infected person transmits to fewer than one other person on average, and the outbreak chain inevitably breaks. The co-infection model shows that when HIV is circulating in the population, this tidy rule breaks down. The researchers identified a critical threshold, denoted Rc, equal to 0.83, below which mpox would normally be expected to fade. Yet because of the backward bifurcation, mpox can persist endemically even when the reproduction number lies between 0.83 and one. In other words, reducing transmission below the classical threshold is no longer sufficient to guarantee elimination when HIV co-infection is present at meaningful levels.

The mechanism behind this counterintuitive behavior lies in the synergy between the two viruses. According to the model, untreated HIV increases an individual’s susceptibility to mpox by a factor of 2.3, with a 95 percent confidence interval spanning 2.1 to 2.6. Immunologically, this makes sense: HIV progressively depletes CD4-positive T cells, the very immune cells that coordinate the response to poxviruses, leaving co-infected individuals both more vulnerable to infection and likely to carry higher viral loads for longer durations. Each person with untreated HIV who acquires mpox therefore contributes disproportionately to onward transmission, creating a reservoir of infection that sustains the epidemic even under conditions that would otherwise drive it to extinction. This feedback loop between immunosuppression and transmission is what generates the backward bifurcation and the associated region of bistability, where both a disease-free state and an endemic state can coexist for the same parameter values.

Vaccination, the cornerstone of the 2022 response, also behaves differently in this co-infection landscape. The model estimates that current vaccination strategies show a 38 percent reduction in efficacy among people with advanced HIV disease, defined as those with CD4 cell counts below 200 cells per cubic millimeter of blood. This finding echoes clinical observations from the outbreak, in which severe and prolonged mpox infections were documented in people with untreated advanced HIV. A vaccine that depends on a competent cellular immune response to protect against a poxvirus will inevitably perform less well in individuals whose cellular immunity has been eroded. The practical consequence is that a uniform vaccination campaign, applied without regard to HIV status, will systematically under-protect precisely the group most likely to sustain transmission, quietly undermining the herd immunity calculation that campaign planners rely upon.

Perhaps the most actionable result of the study is the identification of an optimal intervention window in which antiretroviral therapy expansion and targeted vaccination act synergistically rather than additively. The researchers found that combining 60 percent ART coverage with vaccination targeted according to HIV status reduces co-infection prevalence by a factor of 5.7, with a 95 percent confidence interval of 5.2 to 6.3, compared with either approach deployed in isolation. The synergy arises because the two interventions attack different links in the transmission chain. ART suppresses HIV, restoring immune function and thereby reducing the excess susceptibility to mpox that untreated infection confers, while vaccination directly blocks mpox acquisition in those who remain at risk. When both are applied together, the model shows that the effective reproduction number can be pushed not merely below one but below the critical threshold of 0.83, collapsing the bistable region and eliminating the possibility of endemic persistence.

The authors frame their results as resolving three specific gaps that hampered the 2022 outbreak response. The first was the absence of co-infection-specific transmission metrics; public health agencies tracked mpox cases without systematically accounting for how HIV status modified transmissibility and duration of infection. The second was the lack of quantified synergies between ART scale-up and vaccination, which meant that resources allocated to the two programs were planned independently even though they target overlapping populations. The third was the absence of HIV-stratified vaccine efficacy estimates, which left modelers unable to predict how coverage targets would translate into population-level protection in communities with high HIV prevalence. By supplying parameterized answers to all three questions within a single framework, the study converts what were qualitative concerns during the outbreak into quantitative planning tools.

Beyond the specific numbers, the study carries a broader message about the design of epidemic control programs. The researchers report that coordinated testing and prevention campaigns outperform sequential interventions by 21 to 34 percent across the epidemiological scenarios they examined. Sequential strategies, in which one intervention is rolled out first and the second is added later, allow the synergistic feedback between the two pathogens to continue operating during the gap, wasting transmission-reduction opportunities and potentially allowing the epidemic to settle into the endemic state created by the backward bifurcation. Integrated campaigns, by contrast, compress the intervention window and simultaneously weaken both the HIV-driven amplification of mpox and the mpox-driven burden on an already vulnerable population. In resource-constrained settings, this result suggests that the sequencing of interventions is not a logistical detail but a determinant of whether elimination is achievable at all.

The implications extend beyond mpox itself. The 2022 outbreak demonstrated that sexually transmitted infections can no longer be modeled as independent epidemics occurring in parallel within overlapping networks; they are coupled systems whose dynamics emerge from immunological interaction, behavioral correlation, and intervention overlap. The framework developed by Welu and colleagues provides a template for capturing such coupling, and its central lesson, that co-infection can shift the very thresholds that define epidemic control, is likely to apply wherever immunosuppressive pathogens and acute infections circulate in the same communities. As global health authorities refine preparedness plans for future mpox resurgences and other emerging sexually transmitted threats, the study argues that HIV status must move from the margins of outbreak modeling to its center, and that the era of single-pathogen thinking, however useful it has been, must give way to genuinely integrated approaches.

Subject of Research: Mathematical modeling of HIV-mpox co-infection dynamics and vaccination-dependent epidemic thresholds

Article Title: Vaccination-dependent bifurcations in HIV-mpox co-infection dynamics: Synergistic effects and endemic stability

Article References: Welu, H. T., Asgedom, A. A., Kefela, Y. Y., Atsbaha, H. A., & Berhe, H. W. (2026). Vaccination-dependent bifurcations in HIV-mpox co-infection dynamics: Synergistic effects and endemic stability. PLOS Complex Systems, 3(6), e0000098. https://doi.org/10.1371/journal.pcsy.0000098

Image Credits: AI Generated

DOI: 10.1371/journal.pcsy.0000098

Keywords: mpox, HIV, co-infection, backward bifurcation, vaccination, antiretroviral therapy, epidemiological modeling, endemic stability, men who have sex with men, public health, PLOS Complex Systems, transmission dynamics