Modelling to Quantify the Likelihood that Local Elimination of Transmission has Occurred Using Routine Gambiense Human African Trypanosomiasis Surveillance Data

Modelling timelines to elimination of sleeping sickness in the Democratic Republic of Congo, accounting for possible cryptic human and animal transmission

BACKGROUND

Sleeping sickness (gambiense human African trypanosomiasis, gHAT) is a vector-borne disease targeted for global elimination of transmission (EoT) by 2030. There are, however, unknowns that have the potential to hinder the achievement and measurement of this goal. These include asymptomatic gHAT infections (inclusive of the potential to self-cure or harbour skin-only infections) and whether gHAT infection in animals can contribute to the transmission cycle in humans.

METHODS

Using modelling, we explore how cryptic (undetected) transmission impacts the monitoring of progress towards and the achievement of the EoT goal. We have developed gHAT models that include either asymptomatic or animal transmission, and compare these to a baseline gHAT model without either of these transmission routes, to explore the potential role of cryptic infections on the EoT goal. Each model was independently calibrated to five different health zones in the Democratic Republic of the Congo (DRC) using available historical human case data for 2000-2020 (obtained from the World Health Organization's HAT Atlas). We applied a novel Bayesian sequential updating approach for the asymptomatic model to enable us to combine statistical information about this type of transmission from each health zone.

RESULTS

Our results suggest that, when matched to past case data, we estimated similar numbers of new human infections between model variants, although human infections were slightly higher in the models with cryptic infections. We simulated the continuation of screen-confirm-and-treat interventions, and found that forward projections from the animal and asymptomatic transmission models produced lower probabilities of EoT than the baseline model; however, cryptic infections did not prevent EoT from being achieved eventually under this approach.

CONCLUSIONS

This study is the first to simulate an (as-yet-to-be available) screen-and-treat strategy and found that removing a parasitological confirmation step was predicted to have a more noticeable benefit to transmission reduction under the asymptomatic model compared with the others. Our simulations suggest vector control could greatly impact all transmission routes in all models, although this resource-intensive intervention should be carefully prioritised.

Using Passive Surveillance to Maintain Elimination as a Public Health Problem for Neglected Tropical Diseases: A Model-Based Exploration

BACKGROUND

Great progress is being made toward the goal of elimination as a public health problem for neglected tropical diseases such as leprosy, human African trypanosomiasis, Buruli ulcer, and visceral leishmaniasis, which relies on intensified disease management and case finding. However, strategies for maintaining this goal are still under discussion. Passive surveillance is a core pillar of a long-term, sustainable surveillance program.

METHODS

We use a generic model of disease transmission with slow epidemic growth rates and cases detected through severe symptoms and passive detection to evaluate under what circumstances passive detection alone can keep transmission under control.

RESULTS

Reducing the period of infectiousness due to decreasing time to treatment has a small effect on reducing transmission. Therefore, to prevent resurgence, passive surveillance needs to be very efficient. For some diseases, the treatment time and level of passive detection needed to prevent resurgence is unlikely to be obtainable.

CONCLUSIONS

The success of a passive surveillance program crucially depends on what proportion of cases are detected, how much of their infectious period is reduced, and the underlying reproduction number of the disease. Modeling suggests that relying on passive detection alone is unlikely to be enough to maintain elimination goals.

Comparison of stochastic and deterministic models for gambiense sleeping sickness at different spatial scales: A health area analysis in the DRC

The intensification of intervention activities against the fatal vector-borne disease gambiense human African trypanosomiasis (gHAT, sleeping sickness) in the last two decades has led to a large decline in the number of annually reported cases. However, while we move closer to achieving the ambitious target of elimination of transmission (EoT) to humans, pockets of infection remain, and it becomes increasingly important to quantitatively assess if different regions are on track for elimination, and where intervention efforts should be focused. We present a previously developed stochastic mathematical model for gHAT in the Democratic Republic of Congo (DRC) and show that this same formulation is able to capture the dynamics of gHAT observed at the health area level (approximately 10,000 people). This analysis was the first time any stochastic gHAT model has been fitted directly to case data and allows us to better quantify the uncertainty in our results. The analysis focuses on utilising a particle filter Markov chain Monte Carlo (MCMC) methodology to fit the model to the data from 16 health areas of Mosango health zone in Kwilu province as a case study. The spatial heterogeneity in cases is reflected in modelling results, where we predict that under the current intervention strategies, the health area of Kinzamba II, which has approximately one third of the health zone's cases, will have the latest expected year for EoT. We find that fitting the analogous deterministic version of the gHAT model using MCMC has substantially faster computation times than fitting the stochastic model using pMCMC, but produces virtually indistinguishable posterior parameterisation. This suggests that expanding health area fitting, to cover more of the DRC, should be done with deterministic fits for efficiency, but with stochastic projections used to capture both the parameter and stochastic variation in case reporting and elimination year estimations.

Modelling to infer the role of animals in gambiense human African trypanosomiasis transmission and elimination in the DRC

Gambiense human African trypanosomiasis (gHAT) has been targeted for elimination of transmission (EoT) to humans by 2030. Whilst this ambitious goal is rapidly approaching, there remain fundamental questions about the presence of non-human animal transmission cycles and their potential role in slowing progress towards, or even preventing, EoT. In this study we focus on the country with the most gHAT disease burden, the Democratic Republic of Congo (DRC), and use mathematical modelling to assess whether animals may contribute to transmission in specific regions, and if so, how their presence could impact the likelihood and timing of EoT. By fitting two model variants-one with, and one without animal transmission-to the human case data from 2000-2016 we estimate model parameters for 158 endemic health zones of the DRC. We evaluate the statistical support for each model variant in each health zone and infer the contribution of animals to overall transmission and how this could impact predicted time to EoT. We conclude that there are 24/158 health zones where there is substantial to decisive statistical support for some animal transmission. However-even in these regions-we estimate that animals would be extremely unlikely to maintain transmission on their own. Animal transmission could hamper progress towards EoT in some settings, with projections under continuing interventions indicating that the number of health zones expected to achieve EoT by 2030 reduces from 68/158 to 61/158 if animal transmission is included in the model. With supplementary vector control (at a modest 60% tsetse reduction) added to medical screening and treatment interventions, the predicted number of health zones meeting the goal increases to 147/158 for the model including animal transmission. This is due to the impact of vector reduction on transmission to and from all hosts.

Identifying regions for enhanced control of gambiense sleeping sickness in the Democratic Republic of Congo

Gambiense human African trypanosomiasis (sleeping sickness, gHAT) is a disease targeted for elimination of transmission by 2030. While annual new cases are at a historical minimum, the likelihood of achieving the target is unknown. We utilised modelling to study the impacts of four strategies using currently available interventions, including active and passive screening and vector control, on disease burden and transmission across 168 endemic health zones in the Democratic Republic of the Congo. Median projected years of elimination of transmission show only 98 health zones are on track despite significant reduction in disease burden under medical-only strategies (64 health zones if > 90% certainty required). Blanket coverage with vector control is impractical, but is predicted to reach the target in all heath zones. Utilising projected disease burden under the uniform medical-only strategy, we provide a priority list of health zones for consideration for supplementary vector control alongside medical interventions.

Update of transmission modelling and projections of gambiense human African trypanosomiasis in the Mandoul focus, Chad

BACKGROUND

In recent years, a programme of vector control, screening and treatment of gambiense human African trypanosomiasis (gHAT) infections led to a rapid decline in cases in the Mandoul focus of Chad. To represent the biology of transmission between humans and tsetse, we previously developed a mechanistic transmission model, fitted to data between 2000 and 2013 which suggested that transmission was interrupted by 2015. The present study outlines refinements to the model to: (1) Assess whether elimination of transmission has already been achieved despite low-level case reporting; (2) quantify the role of intensified interventions in transmission reduction; and (3) predict the trajectory of gHAT in Mandoul for the next decade under different strategies.

METHOD

Our previous gHAT transmission model for Mandoul was updated using human case data (2000-2019) and a series of model refinements. These include how diagnostic specificity is incorporated into the model and improvements to the fitting method (increased variance in observed case reporting and how underreporting and improvements to passive screening are captured). A side-by-side comparison of fitting to case data was performed between the models.

RESULTS

We estimated that passive detection rates have increased due to improvements in diagnostic availability in fixed health facilities since 2015, by 2.1-fold for stage 1 detection, and 1.5-fold for stage 2. We find that whilst the diagnostic algorithm for active screening is estimated to be highly specific (95% credible interval (CI) 99.9-100%, Specificity = 99.9%), the high screening and low infection levels mean that some recently reported cases with no parasitological confirmation might be false positives. We also find that the focus-wide tsetse reduction estimated through model fitting (95% CI 96.1-99.6%, Reduction = 99.1%) is comparable to the reduction previously measured by the decline in tsetse catches from monitoring traps. In line with previous results, the model suggests that transmission was interrupted in 2015 due to intensified interventions.

CONCLUSIONS

We recommend that additional confirmatory testing is performed in Mandoul to ensure the endgame can be carefully monitored. More specific measurement of cases, would better inform when it is safe to stop active screening and vector control, provided there is a strong passive surveillance system in place.

Modelling gambiense human African trypanosomiasis infection in villages of the Democratic Republic of Congo using Kolmogorov forward equations

Stochastic methods for modelling disease dynamics enable the direct computation of the probability of elimination of transmission. For the low-prevalence disease of human African trypanosomiasis (gHAT), we develop a new mechanistic model for gHAT infection that determines the full probability distribution of the gHAT infection using Kolmogorov forward equations. The methodology allows the analytical investigation of the probabilities of gHAT elimination in the spatially connected villages of different prevalence health zones of the Democratic Republic of Congo, and captures the uncertainty using exact methods. Our method provides a more realistic approach to scaling the probability of elimination of infection between single villages and much larger regions, and provides results comparable to established models without the requirement of detailed infection structure. The novel flexibility allows the interventions in the model to be implemented specific to each village, and this introduces the framework to consider the possible future strategies of test-and-treat or direct treatment of individuals living in villages where cases have been found, using a new drug.

Modelling to explore the potential impact of asymptomatic human infections on transmission and dynamics of African sleeping sickness

Gambiense human African trypanosomiasis (gHAT, sleeping sickness) is one of several neglected tropical diseases (NTDs) where there is evidence of asymptomatic human infection but there is uncertainty of the role it plays in transmission and maintenance. To explore possible consequences of asymptomatic infections, particularly in the context of elimination of transmission—a goal set to be achieved by 2030—we propose a novel dynamic transmission model to account for the asymptomatic population. This extends an established framework, basing infection progression on a number of experimental and observation gHAT studies. Asymptomatic gHAT infections include those in people with blood-dwelling trypanosomes, but no discernible symptoms, or those with parasites only detectable in skin. Given current protocols, asymptomatic infection with blood parasites may be diagnosed and treated, based on observable parasitaemia, in contrast to many other diseases for which treatment (and/or diagnosis) may be based on symptomatic infection. We construct a model in which exposed people can either progress to either asymptomatic skin-only parasite infection, which would not be diagnosed through active screening algorithms, or blood-parasite infection, which is likely to be diagnosed if tested. We add extra parameters to the baseline model including different self-cure, recovery, transmission and detection rates for skin-only or blood infections. Performing sensitivity analysis suggests all the new parameters introduced in the asymptomatic model can impact the infection dynamics substantially. Among them, the proportion of exposures resulting in initial skin or blood infection appears the most influential parameter. For some plausible parameterisations, an initial fall in infection prevalence due to interventions could subsequently stagnate even under continued screening due to the formation of a new, lower endemic equilibrium. Excluding this scenario, our results still highlight the possibility for asymptomatic infection to slow down progress towards elimination of transmission. Location-specific model fitting will be needed to determine if and where this could pose a threat.

Gambiense African sleeping sickness is an infectious disease targeted for elimination of transmission by 2030. Despite this there is still some uncertainty how frequently some infected people who may not have symptoms could “self-cure” without ever having disease and whether some types of infections, such as infections only in the skin, but not the blood, could still contribute to transmission, yet go undiagnosed. To explore how problematic these asymptomatic infections could be in terms of the elimination goal, we use a mathematical model which quantitatively describes changes to infection and transmission over time and includes these different types of infection. We use results of published experimental or field studies as inputs for the model parameters governing asymptomatic infections. We examined the impact of asymptomatic infections when control interventions are put in place. Compared to a baseline model with no asymptomatics, including asymptomatic infection using plausible biological parameters can have a profound impact on transmission and slow progress towards elimination. In some instances it could be possible that even after initial decline in sleeping sickness cases, progress could stagnate without reaching the elimination goal at all, however location-specific modelling will be needed to determine if and where this could pose a threat.

What Can Modeling Tell Us About Sustainable End Points for Neglected Tropical Diseases?

As programs move closer toward the World Health Organization (WHO) goals of reduction in morbidity, elimination as a public health problem or elimination of transmission, countries will be faced with planning the next stages of surveillance and control in low prevalence settings. Mathematical models of neglected tropical diseases (NTDs) will need to go beyond predicting the effect of different treatment programs on these goals and on to predicting whether the gains can be sustained. One of the most important challenges will be identifying the policy goal and the right constraints on interventions and surveillance over the long term, as a single policy option will not achieve all aims—for example, minimizing morbidity and minimizing costs cannot both be achieved. As NTDs move toward 2030 and beyond, more nuanced intervention choices will be informed by quantitative analyses which are adapted to national context.