Abstract and Introduction
Abstract
Objectives: Understanding the drivers of HIV-1 transmission is of importance for curbing the ongoing epidemic. Phylogenetic methods based on single viral sequences allow us to assess whether two individuals are part of the same viral outbreak, but cannot on their own assess who potentially transmitted the virus. We developed and assessed a molecular epidemiology method with the main aim to screen cohort studies for and to characterize individuals who are ‘potential HIV-1 transmitters’, in order to understand the drivers of HIV-1 transmission.
Methods: We developed and validated a molecular epidemiology approach using longitudinally sampled viral Sanger sequences to characterize potential HIV-1 transmitters in the Swiss HIV Cohort Study.
Results: Our method was able to identify 279 potential HIV-1 transmitters and allowed us to determine the main epidemiological and virological drivers of transmission. We found that the directionality of transmission was consistent with infection times for 72.9% of 85 potential HIV-1 transmissions with accurate infection date estimates. Being a potential HIV-1 transmitter was associated with risk factors including viral load [adjusted odds ratiomultivariable (95% confidence interval): 1.86 (1.49–2.32)], syphilis coinfection [1.52 (1.06–2.19)], and recreational drug use [1.45 (1.06–1.98)]. By contrast for the potential HIV-1 recipients, this association was weaker or even absent [1.18 (0.82–1.72), 0.89 (0.52–1.55) and 1.53 (0.98–2.39), respectively], indicating that inferred directionality of transmission is useful at the population level.
Conclusions: Our results indicate that longitudinally sampled Sanger sequences do not provide sufficient information to identify transmitters with high certainty at the individual level, but that they allow the drivers of transmission at the population level to be characterized.
Introduction
HIV-1 infection remains a global health problem, with over 1.8 million new infections annually and 37.9 million individuals infected worldwide in 2018.[1] The Swiss HIV-1 epidemic started in the early 1980s, initially driven by men who have sex with men (MSM) and intravenous drug users (IDUs). Several events had a favourable impact on the Swiss HIV-1 epidemic: the open drug policy with easy access to safe injection material since the late 1980s, the combination antiretroviral therapy (cART) introduced in 1996, and the Swiss statement in 2008, which led to earlier treatment of people living with HIV (PLWH).[2–4] Nowadays, the HIV-1 epidemic among Swiss IDUs is well controlled. Moreover, the basic reproductive number (Ro) is < 1 for the heterosexual route, indicating a decreasing viral transmission in Switzerland for this group.[5,6] In addition, access to cART prevents HIV-1 transmission and confers a better life expectancy on PLWH, approaching that of HIV-uninfected people.[7–10] Nevertheless, despite the country-wide availability of treatment and prevention, HIV-1 transmission is still ongoing and in Switzerland primarily happens among MSM with high-risk behaviour as in many other resource-rich settings.[5,11–14]
Findings of Marzel et al.[14] in 2016 showed that HIV-1 transmission disproportionally often occurs during acute HIV-1 infection and treatment interruption and failures. Transmitted drug resistance mutations further threaten the success of first-line treatments.[15] This emphasizes the importance of early diagnosis as well as systematic testing and monitoring of drug resistance mutations in the effort to stop HIV-1 transmission.[4] As a complement to this, finding new medical solutions and understanding the pattern of transmission and intra-host diversity are essential to reduce the number of new HIV-1 infections.
Phylogenetic methods were used heavily in HIV research to identify the origin of the pandemic, understand transmission patterns, and calculate R o and the heritability of different viral traits.[5,16–19] Further, combining phylogenetic methods with epidemiological and clinical data can be used to characterize HIV-1 transmission events.[14,20–22]
So far, there is no consensus on methods to define HIV-1 transmission clusters or potential HIV-1 transmitters. The traditional molecular epidemiology approaches are cross-sectional, using only one viral sequence per individual, and they focus on identifying individuals who belong to the same transmission network. They further verify the clusters identified using patristic distance, bootstrapping or cluster density criteria. However, the directionality of infection remains difficult to assess with these methods without additional data such as infection times and also because of sparse or incomplete sampling.[5,11,14,18,21,23–25] For these reasons, understanding the drivers of the HIV-1 epidemic in Switzerland remains challenging.
In this study, we developed and tested a molecular epidemiology method based on phylogeny reconstruction and cluster analysis to characterize potential HIV-1 transmitters and understand the drivers of the HIV-1 epidemic in Switzerland. Our method was developed and validated using longitudinally sampled HIV-1 partial polymerase (pol) nucleotide sequences, routinely and retrospectively obtained, in the drug resistance database (DRDB) of the Swiss HIV Cohort Study (SHCS).