Impact of Heterogeneity in Sexual Behavior on Effectiveness in Reducing HIV Transmission with Test-and-Treat Strategy

Model-based evaluation of the impact of a potential HIV cure on HIV transmission dynamics

The development of an HIV cure is a global health priority, with the target product profile (TPP) for an HIV cure guiding research efforts. Using a mathematical model calibrated to data from men who have sex with men (MSM) in the Netherlands, we assessed whether an effective cure could help end the HIV epidemic. Following the TPP, we evaluated two scenarios: (i) HIV remission, where the virus is suppressed in an individual without ongoing antiretroviral therapy (ART) but may rebound, and (ii) HIV eradication, which aims to completely remove the virus from the individual. Here, we show that sustained HIV remission (without rebound) or HIV eradication could reduce new HIV infections compared to a scenario without a cure. In contrast, transient HIV remission with a risk of rebound could increase new infections if rebounds are not closely monitored, potentially undermining HIV control efforts. Our findings emphasize the critical role of cure characteristics in maximizing cure benefits for public health and highlight the need to align HIV cure research with public health objectives to end the HIV epidemic.

Transitions in sexual behaviour among gay, bisexual, and other men who have sex with men in England: Data from a prospective study

The effectiveness of population-level intervention for HIV elimination is influenced by individual-level variation in sexual behaviour. We assess within-person changes in the frequency of condomless anal sex with two or more partners (CLS2+), estimate the transition probabilities and examine the predictors of transitions among a prospective cohort of HIV-negative gay, bisexual, and other men who have sex with men (GBMSM). Participants were recruited through one of three sexual health clinics in London and Brighton (July 2013 to April 2016) and self-completed a baseline paper questionnaire in the clinic. During follow-up, they were invited to complete four-monthly questionnaires twice a year and subsequent annual online questionnaires once a year (March 2015 to March 2018). We used Markov chain models to estimate transition probabilities from 'higher-risk' (CLS2+) to 'lower-risk' (no CLS2+) and vice versa, and to assess factors associated with transitions between different sexual risk levels. Among 1,162 men enrolled in the study, 622 (53.5%) completed at least one online questionnaire. Higher-risk behaviour was reported in 376/622 (60.4%) men during online follow-up. Overall, 1,665/3,277 (37.5%) baseline and follow-up questionnaires reported higher-risk behaviour. More than 60% of men (376/622) reported higher-risk behaviour at least one period during the follow-up, while 39.5% of men (246/622) never reported CLS2+ during the follow-up. In the next four months, the estimated probability of continuing higher-risk behaviour among men who reported higher-risk behaviour was 78%. Calendar time, recent HIV tests, PrEP and PEP use were the predictors of staying in higher-risk behaviour, while less stable housing status was associated with switching to lower-risk behaviour. Among men who reported lower-risk behaviour, the probability of engaging in the same behaviour was 88%. Recent HIV tests, PrEP and PEP use, recreational drugs, chemsex-associated drug and injection drugs, and bacterial STIs diagnosis were the predictors of switching to higher-risk behaviour. Our results indicate that at any one point in time, the majority of GBMSM are at low risk for HIV acquisition, although many experience short periods in which they are at higher risk. Markers of transitions can be utilized to identify which GBMSM are likely to increase or decrease their risk, thus helping the timing of HIV prevention interventions.

Impact of increased diagnosis of early HIV infection and immediate antiretroviral treatment initiation on HIV transmission among men who have sex with men in the Netherlands

The number of new HIV infections among men who have sex with men (MSM) in the Netherlands has been decreasing, but additional efforts are required to bring it further down. This study aims to assess the impact of increased diagnosis of early HIV infection combined with immediate antiretroviral treatment (ART) initiation on reducing HIV transmission among MSM. We developed an agent-based model calibrated to HIV surveillance and sexual behavior data for MSM in the Netherlands in 2017-2022. Starting in 2023, we simulated a 10-year intervention that accelerates HIV diagnosis during the first 3 or 6 months after HIV acquisition across five levels of increased diagnosis rates (2, 4, 8, 16, and 32-fold), followed by immediate ART initiation. The upper limit of the intervention's impact over 10 years is projected to result in the cumulative 298 (95-th QI: 162-451) HIV infections averted. A 32-fold increase in the diagnosis rate within 3 months after HIV acquisition (corresponding to 100% of all new HIV infections diagnosed within 3 months of acquisition) results in 269 (95-th QI: 147-400) infections averted, approaching closely maximum impact. By extending the scope of the intervention to individuals who acquired HIV infection within the previous 6 months, a smaller 8-fold increase in the diagnosis rate (corresponding to 97% of new HIV infections diagnosed within 6 months of acquisition) approaches closely the maximum impact of the intervention by averting 256 (95-th QI: 122-411) HIV infections. Our sensitivity analyses showed that, in an epidemiological context similar to the modern-day the Netherlands, immediate initiation of ART accompanying accelerated diagnosis of individuals with early HIV infection does not significantly affect HIV transmission dynamics. Accelerating early HIV diagnosis through increased awareness, screening, and testing can further reduce transmission among MSM. Meeting this goal necessitates a stakeholder needs assessment.

Risk heterogeneity in compartmental HIV transmission models of ART as prevention in Sub-Saharan Africa: A scoping review

BACKGROUND

Transmission models provide complementary evidence to clinical trials about the potential population-level incidence reduction attributable to ART (ART prevention impact). Different modelling assumptions about risk heterogeneity may influence projected ART prevention impacts. We sought to review representations of risk heterogeneity in compartmental HIV transmission models applied to project ART prevention impacts in Sub-Saharan Africa.

METHODS

We systematically reviewed studies published before January 2020 that used non-linear compartmental models of sexual HIV transmission to simulate ART prevention impacts in Sub-Saharan Africa. We summarized data on model structure/assumptions (factors) related to risk and intervention heterogeneity, and explored multivariate ecological associations of ART prevention impacts with modelled factors.

RESULTS

Of 1384 search hits, 94 studies were included. 64 studies considered sexual activity stratification and 39 modelled at least one key population. 21 studies modelled faster/slower ART cascade transitions (HIV diagnosis, ART initiation, or cessation) by risk group, including 8 with faster and 4 with slower cascade transitions among key populations versus the wider population. In ecological analysis of 125 scenarios from 40 studies (subset without combination intervention), scenarios with risk heterogeneity that included turnover of higher risk groups were associated with smaller ART prevention benefits. Modelled differences in ART cascade across risk groups also influenced the projected ART benefits, including: ART prioritized to key populations was associated with larger ART prevention benefits. Of note, zero of these 125 scenarios considered lower ART coverage among key populations.

CONCLUSION

Among compartmental transmission models applied to project ART prevention impacts in Sub-Saharan Africa, representations of risk heterogeneity and projected impacts varied considerably. Inclusion/exclusion of risk heterogeneity with turnover, and intervention heterogeneity across risk groups could influence the projected impacts of ART scale-up. These findings highlight a need to capture risk heterogeneity with turnover and cascade heterogeneity when projecting ART prevention impacts.

Challenges for modelling interventions for future pandemics

Mathematical modelling and statistical inference provide a framework to evaluate different non-pharmaceutical and pharmaceutical interventions for the control of epidemics that has been widely used during the COVID-19 pandemic. In this paper, lessons learned from this and previous epidemics are used to highlight the challenges for future pandemic control. We consider the availability and use of data, as well as the need for correct parameterisation and calibration for different model frameworks. We discuss challenges that arise in describing and distinguishing between different interventions, within different modelling structures, and allowing both within and between host dynamics. We also highlight challenges in modelling the health economic and political aspects of interventions. Given the diversity of these challenges, a broad variety of interdisciplinary expertise is needed to address them, combining mathematical knowledge with biological and social insights, and including health economics and communication skills. Addressing these challenges for the future requires strong cross-disciplinary collaboration together with close communication between scientists and policy makers.

Temporal Variation in One-Time Partnership Rates Among Young Men Who Have Sex With Men and Transgender Women

BACKGROUND

Volatility in sexual contact rates has been recognized as an important factor influencing HIV transmission dynamics. One-time partnerships may be particularly important given the potential to quickly accumulate large number of contacts. Yet, empirical data documenting individual variation in contact rates remain rare. This study provides much needed data on temporal variation in one-time partners to better understand behavioral dynamics and improve the accuracy of transmission models.

METHODS

Data for this study were obtained from a longitudinal cohort study of young men who have sex with men and transgender women in Chicago. Participants provided sexual network data every 6 months for 2 years. A series of random effects models examined variation in one-time partnership rates and disaggregated within and between associations of exposure variables. Exposure variables included prior number of one-time partners, number of casual partners, and having a main partner.

RESULTS

Results indicated substantial between-person and within-person variation in one-time partners. Casual partnerships were positively associated and main partnerships negatively associated with one-time partnership rates. There remained a small positive association between prior one-time partnerships and the current number of one-time partnerships.

CONCLUSIONS

Despite the preponderance of a low number of one-time partners, substantial variation in one-time partnership rates exists among young men who have sex with men and transgender women. Accordingly, focusing on high contact rate individuals alone may be insufficient to identify periods of highest risk. Future studies should use these estimates to more accurately model how volatility impacts HIV transmission and better understand how this variation influences intervention effectiveness.

Influence of sexual risk behaviour and STI co-infection dynamics on the evolution of HIV set point viral load in MSM

HIV viral load (VL) is an important predictor of HIV progression and transmission. Anti-retroviral therapy (ART) has been reported to reduce HIV transmission by lowering VL. However, apart from this beneficial effect, increased levels of population mean set-point viral load (SPVL), an estimator for HIV virulence, have been observed in men who have sex with men (MSM) in the decade following the introduction of ART in The Netherlands. Several studies have been devoted to explain these counter-intuitive trends in SPVL. However, to our knowledge, none of these studies has investigated an explanation in which it arises as the result of a sexually transmitted infection (STI) co-factor in detail. In this study, we adapted an event-based, individual-based model to investigate how STI co-infection and sexual risk behaviour affect the evolution of HIV SPVL in MSM before and after the introduction of ART. The results suggest that sexual risk behaviour has an effect on SPVL and indicate that more data are needed to test the effect of STI co-factors on SPVL. Furthermore, the observed trends in SPVL cannot be explained by sexual risk behaviour and STI co-factors only. We recommend to develop mathematical models including also factors related to viral evolution as reported earlier in the literature. However, this requires more complex models, and the collection of more data for parameter estimation than what is currently available.

Model-based evaluation of school- and non-school-related measures to control the COVID-19 pandemic

The role of school-based contacts in the epidemiology of SARS-CoV-2 is incompletely understood. We use an age-structured transmission model fitted to age-specific seroprevalence and hospital admission data to assess the effects of school-based measures at different time points during the COVID-19 pandemic in the Netherlands. Our analyses suggest that the impact of measures reducing school-based contacts depends on the remaining opportunities to reduce non-school-based contacts. If opportunities to reduce the effective reproduction number (Re) with non-school-based measures are exhausted or undesired and Re is still close to 1, the additional benefit of school-based measures may be considerable, particularly among older school children. As two examples, we demonstrate that keeping schools closed after the summer holidays in 2020, in the absence of other measures, would not have prevented the second pandemic wave in autumn 2020 but closing schools in November 2020 could have reduced Re below 1, with unchanged non-school-based contacts.

The dynamics of a simple, risk-structured HIV model

Many diseases, such as HIV, are heterogeneous for risk. In this paper, we study an infectious-disease model for a population with demography, mass-action incidence, an arbitrary number of risk classes, and separable mixing. We complement our general analyses with two specific examples. In the first example, the mean of the components of the transmission coefficients decreases as we add more risk classes. In the second example, the mean stays constant but the variance decreases. For each example, we determine the disease-free equilibrium, the basic reproduction number, and the endemic equilibrium. We also characterize the spectrum of eigenvalues that determine the stability of the endemic equilibrium. For both examples, the basic reproduction number decreases as we add more risk classes. The endemic equilibrium, when present, is asymptotically stable. Our analyses suggest that risk structure must be modeled correctly, since different risk structures, with similar mean properties, can produce different dynamics.

Diverging trends in incidence of HIV versus other sexually transmitted infections in HIV-negative MSM in Amsterdam

OBJECTIVES

We investigated changes in incidence rates of HIV and sexually transmitted infections (STIs) and trends in sexual behavior in MSM from 2009 to 2017.

DESIGN

Open prospective cohort study.

METHODS

HIV-negative MSM enrolled in the Amsterdam Cohort Studies were included. Participants semiannually completed a questionnaire on sexual behavior and were tested for HIV-1, syphilis, and urethral, anal and pharyngeal chlamydia and gonorrhea. Time trends in incidence rates were analyzed using exponential survival models.

RESULTS

During follow-up, 42 of 905 MSM acquired HIV. The HIV incidence rate was 1.9/100 person-years [95% confidence interval (CI) 1.0-3.7] in 2009 and decreased to 0.5/100 person-years (95% CI 0.2-1.4) in 2017 (P = 0.03). The largest decrease was observed in participants aged at least 35 years (P = 0.005), while the trend remained stable in 18-34 year olds (P = 0.4). The incidence rate for any bacterial STI was 16.8/100 person-years (95% CI 13.4-21.0) in 2010, and increased to 33.1/100 person-years (95% CI 29.0-37.9) in 2017 (P < 0.001). Between 2009 and 2017, the percentage reporting condomless anal sex with casual partners increased from 26.9 to 39.4% (P < 0.001), and the mean number of casual partners from eight (95% CI 8-8) to 11 (95% CI 10-11) (P = 0.05). Condomless anal sex with steady partner(s) remained stable over time (P = 0.5).

CONCLUSION

Among MSM in Amsterdam, incidence rates of HIV versus other STI show diverging trends. The increase in STI incidence coincides with a decrease in condom use with casual partners. The decrease in HIV incidence, despite increased sexual risk behavior, suggests that other HIV prevention methods have been successful in reducing HIV transmission among MSM.

Impact of sexual trajectories of men who have sex with men on the reduction in HIV transmission by pre-exposure prophylaxis

Changes in sexual risk behavior over the life course in men who have sex with men (MSM) can influence population-level intervention efficacy. Our objective was to investigate the impact of incorporating sexual trajectories describing long-term changes in risk levels on the reduction in HIV prevalence by pre-exposure prophylaxis (PrEP) among MSM. Based on the Amsterdam Cohort Study data, we developed two models of HIV transmission in a population stratified by sexual behavior. In the first model, individuals were stratified into low, medium and high risk levels and did not change their risk levels. The second model had the same stratification but incorporated additionally three types of sexual behavior trajectories. The models assumed universal antiretroviral treatment of HIV⁺ MSM, and PrEP use by high risk HIV^- MSM. We computed the relative reduction in HIV prevalence in both models for annual PrEP uptakes of 10% to 80% at different time points after PrEP introduction. We then investigated the impact of sexual trajectories on the effectiveness of PrEP intervention. The impact of sexual trajectories on the overall prevalence and prevalence in individuals at low, medium and high risk levels varied with PrEP uptake and time after PrEP introduction. Compared to the model without sexual trajectories, the model with trajectories predicted a higher impact of PrEP on the overall prevalence, and on the prevalence among the medium and high risk individuals. In low risk individuals, there was more reduction in prevalence during the first 15 years of PrEP intervention if sexual trajectories were not incorporated in the model. After that point, at low risk level there was more reduction in the model with trajectories. In conclusion, our study predicts that sexual trajectories increase the estimated impact of PrEP on reducing HIV prevalence when compared to a population where risk levels do not change.

Elimination prospects of the Dutch HIV epidemic among men who have sex with men in the era of preexposure prophylaxis

Objective:

Preexposure prophylaxis (PrEP) is a promising intervention to help end the HIV epidemic among men who have sex with men (MSM) in the Netherlands. We aimed to assess the impact of PrEP on HIV prevalence in this population and to determine the levels of PrEP coverage necessary for HIV elimination.

Design and methods:

We developed a mathematical model of HIV transmission in a population stratified by sexual risk behavior with universal antiretroviral treatment (ART) and daily PrEP use depending on an individual's risk behavior. We computed the effective reproduction number, HIV prevalence, ART and PrEP coverage for increasing ART and PrEP uptake levels, and examined how these were affected by PrEP effectiveness and duration of PrEP use.

Results:

At current levels of ART coverage of 80%, PrEP effectiveness of 86% and PrEP duration of 5 years, HIV elimination required 82% PrEP coverage in the highest risk group (12 000 MSM with more than 18 partners per year). If ART coverage increased by 9%, the elimination threshold was at 70% PrEP coverage. For shorter PrEP duration and lower effectiveness elimination prospects were less favorable. For the same number of PrEP users distributed among two groups with highest risk behavior, prevalence dropped from the current 8 to 4.6%.

Conclusion:

PrEP for HIV prevention among MSM could, in principle, eliminate HIV from this population in the Netherlands. The highest impact of PrEP on prevalence was predicted when ART and PrEP coverage increased simultaneously and PrEP was used by the highest risk individuals.

The effect of assortative mixing on stability of low helminth transmission levels and on the impact of mass drug administration: Model explorations for onchocerciasis

Background

Stable low pre-control prevalences of helminth infection are not uncommon in field settings, yet it is poorly understood how such low levels can be sustained, thereby challenging efforts to model them. Disentangling possible facilitating mechanisms is important, since these may differently affect intervention impact. Here we explore the role of assortative (i.e. non-homogenous) mixing and exposure heterogeneity in helminth transmission, using onchocerciasis as an example.

Methodology/Principal findings

We extended the established individual-based model ONCHOSIM to allow for assortative mixing, assuming that individuals who are relatively more exposed to fly bites are more connected to each other than other individuals in the population as a result of differential exposure to a sub-population of blackflies. We used the model to investigate how transmission stability, equilibrium microfilarial (mf) prevalence and intensity, and impact of mass drug administration depend on the assumed degree of assortative mixing and exposure heterogeneity, for a typical rural population of about 400 individuals. The model clearly demonstrated that with homogeneous mixing and moderate levels of exposure heterogeneity, onchocerciasis could not be sustained below 35% mf prevalence. In contrast, assortative mixing stabilised onchocerciasis prevalence at levels as low as 8% mf prevalence. Increasing levels of assortative mixing significantly reduced the probability of interrupting transmission, given the same duration and coverage of mass drug administration.

Conclusions/Significance

Assortative mixing patterns are an important factor to explain stable low prevalence situations and are highly relevant for prospects of elimination. Their effect on the pre-control distribution of mf intensities in human populations is only detectable in settings with mf prevalences <30%, where high skin mf density in mf-positive people may be an indication of assortative mixing. Local spatial variation in larval infection intensity in the blackfly intermediate host may also be an indicator of assortative mixing.

Most mathematical models for parasitic worm infections predict that at low prevalences transmission will fade out spontaneously because of the low mating probability of male and female worms. However, sustained low prevalence situations do exist in reality. Low prevalence areas have become of particular interest now that several worm infections are being targeted for elimination and the question arises whether transmission in such areas is driven locally and should be targeted with interventions. We hypothesise that an explanation for the existence of low prevalence areas is assortative mixing, which is the preferential mixing of high-risk groups among themselves and which has been shown to play an important role in transmission of other infectious diseases. For onchocerciasis, assortative mixing would mean that transmission is sustained by a sub-group of people and a connected sub-population of the blackfly intermediate host that mix preferentially with each other. Using a mathematical model, we study how assortative mixing allows for sustained low prevalences and show that it decreases the probability of interrupting transmission by means of mass drug administration. We further identify data sources that may be used to quantify the degree of assortative mixing in field settings.

Modelling the dynamics of population viral load measures under HIV treatment as prevention

In 2011 the Centers for Disease Control and Prevention (CDC) published guidelines for the use of population viral load (PVL), community viral load (CVL) and monitored viral load (MVL), defined as the average viral load (VL) of all HIV infected individuals in a population, of all diagnosed individuals, and of all individuals on antiretroviral treatment (ART), respectively. Since then, CVL has been used to assess the effectiveness of ART on HIV transmission and as a proxy for HIV incidence. The first objective of this study was to investigate how aggregate VL measures change with the HIV epidemic phase and the drivers behind these changes using a mathematical transmission model. Secondly, we aimed to give some insight into how well CVL correlates with HIV incidence during the course of the epidemic and roll out of ART. We developed a compartmental model for disease progression and HIV transmission with disease stages that differ in viral loads for epidemiological scenarios relevant to a concentrated epidemic in a population of men who have sex with men (MSM) in Western Europe (WE) and to a generalized epidemic in a heterosexual population in Sub-Saharan Africa (SSA). The model predicts that PVL and CVL change with the epidemic phase, while MVL stays constant. These dynamics are linked to the dynamics of infected subgroups (undiagnosed, diagnosed untreated and treated) in different disease stages (primary, chronic and AIDS). In particular, CVL decreases through all epidemic stages: before ART, since chronic population builds up faster than AIDS population and after ART, due to the build-up of treated population with low VL. The trends in CVL and incidence can be both opposing and coinciding depending on the epidemic phase. Before ART is scaled up to sufficiently high levels, incidence increases while CVL decreases. After this point, CVL is a useful indicator of changes in HIV incidence. The model predicts that during the ART scale-up HIV transmission is driven by undiagnosed and diagnosed untreated individuals, and that new infections decline due to the increase in the number of treated. Although CVL is not able to capture the contribution of undiagnosed population to HIV transmission, it declines due to the increase of people on ART too. In the scenarios described by our model, the present epidemic phase corresponds to declining trends in CVL and incidence.

Does antiretroviral therapy change partnership dynamics and HIV risk behaviours among HIV-infected adults

Objective:

We explore the impact of antiretroviral therapy (ART) on partnership acquisition and dissolution rates and changes in sexual behaviours among HIV-infected adults.

Design:

Using detailed longitudinal data from a prospective cohort of HIV-infected adults with CD4⁺ cell count below 200 cells/μl (ART-eligible) or CD4⁺ cell count above 500 cells/μl (pre-ART) conducted in rural KwaZulu-Natal, South Africa, from 2009 to 2012.

Methods:

Partnership acquisition and dissolution are explored through survival analysis methods, whereas generalized linear models were fitted for the sexual behaviour outcomes with interaction terms to allow the association with ART to vary over time. Throughout, the primary comparison of interest for each outcome is differences between the two ART groups.

Results:

ART is not associated with partner acquisition or relationship dissolution. During follow-up, the two ART groups do not differ in the odds of being sexually active nor the number of sex acts, whereas the odds of unprotected sex are significantly lower for partnerships of ART-eligible participants (adjusted odds ratio 0.26, 95% confidence interval 0.15, 0.43). Relationship-level characteristics including cohabitation status and wanting more children with that partner are associated with higher odds and increased frequency of sexual activity, and increased odds of unprotected sex, whereas living with partner, higher relationship quality and longer relationship duration are associated with lower risk of partnership dissolution.

Conclusion:

Being on ART was not associated with increased sexual risk behaviours, a reassuring finding given the WHO recommends ART initiation upon HIV diagnosis. The importance of relationship-level characteristics provides evidence that HIV care services should offer routine support for HIV disclosure and sexual risk reduction, and promotion of couples-testing and positive couple relationships.