HIV co-infection is associated with reduced <i>Mycobacterium tuberculosis</i> transmissibility in sub-Saharan Africa

Abstract Persons living with HIV are known to be at increased risk of developing tuberculosis (TB) disease upon infection with Mycobacterium tuberculosis ( Mtb ). However, it has remained unclear how HIV co-infection affects subsequent Mtb transmission from these patients. Here, we customized a Bayesian phylodynamic framework to estimate the effects of HIV co-infection on the Mtb transmission dynamics from sequence data. We applied our model to four Mtb genomic datasets collected in sub-Saharan African countries with a generalized HIV epidemic. Our results confirm that HIV co-infection is a strong risk factor for developing active TB. Additionally, we demonstrate that HIV co-infection is associated with a reduced effective reproductive number for TB. Stratifying the population by CD4+ T-cell count yielded similar results, suggesting that, in this context, CD4+ T-cell count is not a better predictor of Mtb transmissibility than HIV infection status. Together, our genome-based analyses complement observational household studies, and firmly establish the negative association between HIV co-infection and Mtb transmissibility. Author summary Many sub-Saharan African countries have seen a considerable rise in TB incidence since the introduction of HIV, suggesting a strong interaction between HIV and TB epidemics. HIV infection is recognized as an important risk factor for developing TB, but the contribution of HIV-infected TB patients to further Mtb transmission is poorly understood. In this study, we analyzed four sets of Mtb genomic sequences collected in different countries, including sequences from HIV-negative and HIV-positive TB patients. We applied a phylodynamic model to these sequences, aimed at inferring transmission dynamics within and between different host populations. While our findings support that HIV is a strong risk factor for TB, we show that HIV-positive TB patients generate a significantly lower number of secondary TB cases than HIV-negative patients. This suggests that HIV-positive patients often act as sinks in Mtb transmission chains, while HIV-negative patients are a major source of transmission.