Predictive Risk Mapping of Schistosomiasis in Madagascar Using Ecological Niche Modeling and Precision Mapping

Impact of environmental factors on Biomphalaria pfeifferi vector capacity leading to human infection by Schistosoma mansoni in two regions of western Côte d'Ivoire

ABSTRACT: BACKGROUND: Intestinal schistosomiasis remains a worrying health problem, particularly in western Côte d'Ivoire, despite control efforts. It is therefore necessary to understand all the factors involved in the development of the disease, including biotic and abiotic factors. The aim of this study was to examine the factors that could support the maintenance of the intermediate host and its vectorial capacity in western Côte d'Ivoire.

METHODS

Data on river physicochemical, microbiological, and climatic parameters, the presence or absence of snails with Schistosoma mansoni, and human infections were collected between January 2020 and February 2021. Spearman rank correlation tests, Mann-Whitney, analysis of variance (ANOVA), and an appropriate model selection procedure were used to analyze the data.

RESULTS

The overall prevalence of infected snails was 56.05%, with infection reaching 100% in some collection sites and localities. Of 26 sites examined, 25 contained thermophilic coliforms and 22 contained Escherichia coli. Biomphalaria pfeifferi was observed in environments with lower land surface temperature (LST) and higher relative air humidity (RAH), and B. pfeifferi infection predominated in more acidic environments. Thermal coliforms and E. coli preferred higher pH levels. Lower maximum LST (LST_Max) and higher RAH and minimum LST (LST_Min) were favorable to E. coli, and lower LST_Max favored coliforms. The presence of B. pfeifferi was positively influenced by water temperature (T °C), LST_Min, RAH, and precipitation (Pp) (P < 0.05) and negatively influenced by pH, total dissolved solids (TDS), electrical conductivity (EC), LST_Max, and mean land surface temperature (LST). The parameters pH, TDS, EC, LST_Min, LST, and Pp had a positive impact on snail infection, while LST_Max had a negative impact on infection. Only pH had a positive effect on coliform and E. coli abundance. Of the 701 people examined for human schistosomiasis, 73.13% were positive for the point-of-care circulating cathodic antigen (POC-CCA) test and 12.01% for the Kato-Katz (KK) test. A positive correlation was established between human infections and the abundance of Biomphalaria (r2 = 0.879, P = 0.04959).

CONCLUSIONS

The results obtained reflect the environmental conditions that are conducive to the maintenance of S. mansoni infection in this part of the country. To combat this infection as effectively as possible, it will be necessary not only to redouble efforts but also to prioritize control according to the level of endemicity at the village level.

A potential transition from a concentrated to a generalized HIV epidemic: the case of Madagascar

BACKGROUND

HIV expansion is controlled by a range of interrelated factors, including the natural history of HIV infection and socio-economical and structural factors. However, how they dynamically interact in particular contexts to drive a transition from concentrated HIV epidemics in vulnerable groups to generalized epidemics is poorly understood. We aim to explore these mechanisms, using Madagascar as a case-study.

METHODS

We developed a compartmental dynamic model using available data from Madagascar, a country with a contrasting concentrated epidemic, to explore the interaction between these factors with special consideration of commercial and transactional sex as HIV-infection drivers.

RESULTS

The model predicts sigmoidal-like prevalence curves with turning points within years 2020-2022, and prevalence reaching stabilization by 2033 within 9 to 24% in the studied (10 out of 11) cities, similar to high-prevalence regions in Southern Africa. The late/slow introduction of HIV and  circumcision, a widespread traditional practice in Madagascar, could have slowed down HIV propagation, but, given the key interplay between risky behaviors associated to young women and acute infections prevalence, mediated by transactional sex, the protective effect of circumcision is currently insufficient to contain the expansion of the disease in Madagascar.

CONCLUSIONS

These results suggest that Madagascar may be experiencing a silent transition from a concentrated to a generalized HIV epidemic. This case-study model could help to understand how this HIV epidemic transition occurs.

Schistosomiasis elimination in Madagascar: challenges and opportunities for implementing the new WHO guidelines

Madagascar is one of the countries with the highest burden of schistosomiasis worldwide. The release from the WHO of the new 2021-2030 neglected tropical disease (NTD) roadmap alongside with the schistosomiasis guidelines sets the ambitious goal of eliminating schistosomiasis as a public health problem worldwide. In Madagascar, implementation barriers exist. This paper has the objective of identifying strengths, weaknesses, opportunities and threats in order to build on their basis practices and policies that can help the country to align with the international global health agenda and reach the ambitious goal set by the WHO.

Open-source environmental data as an alternative to snail surveys to assess schistosomiasis risk in areas approaching elimination

BACKGROUND

Although the presence of intermediate snails is a necessary condition for local schistosomiasis transmission to occur, using them as surveillance targets in areas approaching elimination is challenging because the patchy and dynamic quality of snail host habitats makes collecting and testing snails labor-intensive. Meanwhile, geospatial analyses that rely on remotely sensed data are becoming popular tools for identifying environmental conditions that contribute to pathogen emergence and persistence.

METHODS

In this study, we assessed whether open-source environmental data can be used to predict the presence of human Schistosoma japonicum infections among households with a similar or improved degree of accuracy compared to prediction models developed using data from comprehensive snail surveys. To do this, we used infection data collected from rural communities in Southwestern China in 2016 to develop and compare the predictive performance of two Random Forest machine learning models: one built using snail survey data, and one using open-source environmental data.

RESULTS

The environmental data models outperformed the snail data models in predicting household S. japonicum infection with an estimated accuracy and Cohen's kappa value of 0.89 and 0.49, respectively, in the environmental model, compared to an accuracy and kappa of 0.86 and 0.37 for the snail model. The Normalized Difference in Water Index (an indicator of surface water presence) within half to one kilometer of the home and the distance from the home to the nearest road were among the top performing predictors in our final model. Homes were more likely to have infected residents if they were further from roads, or nearer to waterways.

CONCLUSION

Our results suggest that in low-transmission environments, leveraging open-source environmental data can yield more accurate identification of pockets of human infection than using snail surveys. Furthermore, the variable importance measures from our models point to aspects of the local environment that may indicate increased risk of schistosomiasis. For example, households were more likely to have infected residents if they were further from roads or were surrounded by more surface water, highlighting areas to target in future surveillance and control efforts.

Perspective on Schistosomiasis Drug Discovery: Highlights from a Schistosomiasis Drug Discovery Workshop at Wellcome Collection, London, September 2022

In September 2022, the Drug Discovery Unit at the University of Dundee, UK, organised an international meeting at the Wellcome Collection in London to explore the current clinical situation and challenges associated with treating schistosomiasis. The aim of this meeting was to discuss the need for new treatments in view of the clinical situation and to ascertain what the key requirements would be for any potential new anti-schistosomals. This information will be essential to inform ongoing drug discovery efforts for schistosomiasis. We also discussed the potential drug discovery pathway and associated criteria for progressing compounds to the clinic. To date, praziquantel (PZQ) is the only drug available to treat all species causing schistosomiasis, but it is often unable to completely clear parasites from an infected patient, partially due to its inactivity against juvenile worms. PZQ-mediated mass drug administration campaigns conducted in endemic areas (e.g., sub-Saharan Africa, where schistosomiasis is primarily prevalent) have contributed to reducing the burden of disease but will not eliminate the disease as a public health problem. The potential for Schistosoma to develop resistance towards PZQ, as the sole treatment available, could become a concern. Consequently, new anthelmintic medications are urgently needed, and this Perspective aims to capture some of the learnings from our discussions on the key criteria for new treatments.

Temporal modelling of Lymnaea natalensis (Krauss, 1848) in tropical aquatic habitats

Lymnaea natalensis is the only snail intermediate host of Fasciola gigantica, the causative agent of fascioliasis, in Nigeria. The species also serves as intermediate host for many other African trematode species of medical and veterinary importance, and it is found throughout the country. However, there is no detailed information on the factors that influence its distribution and seasonal abundance in the tropical aquatic habitats in Nigeria. This study used the geographic information system and remotely sensed data to develop models for predicting the distribution of L. natalensis in South-Western Nigeria. Both land surface temperature (LST) and normalised difference vegetation index (NDVI) were extracted from Landsat satellite imagery; other variables (slope and elevation) were extracted from a digital elevation model (DEM) while rainfall data were retrieved from the European Meteorology Research Programme (EMRP). These environmental variables were integrated into a geographic information system (GIS) to predict suitable habitats of L. natalensis using exploratory regression. A total of 1410 L. natalensis snails were collected vis-à-vis 22 sampling sites. Built-up areas recorded more L. natalensis compared with farmlands. There was no significant difference in the abundance of snails with season (p  0.05). The regression models showed that rainfall, NDVI, and slope were predictors of L. natalensis distribution. The habitats suitable for L. natalensis were central areas, while areas to the north and south were not suitable for L. natalensis.Contribution: The predictive risk models of L. natalensis in the study will be useful in mapping other areas where the snail sampling could not be conducted.

Prevalence and risk distribution of schistosomiasis among adults in Madagascar: a cross-sectional study

BACKGROUND

The goal to eliminate the parasitic disease of poverty schistosomiasis as a public health problem is aligned with the 2030 United Nations agenda for sustainable development goals, including universal health coverage (UHC). Current control strategies focus on school-aged children, systematically neglecting adults. We aimed at providing evidence for the need of shifting the paradigm of schistosomiasis control programs from targeted to generalized approaches as key element for both the elimination of schistosomiasis as a public health problem and the promotion of UHC.

METHODS

In a cross-sectional study performed between March 2020 and January 2021 at three primary health care centers in Andina, Tsiroanomandidy and Ankazomborona in Madagascar, we determined prevalence and risk factors for schistosomiasis by a semi-quantitative PCR assay from specimens collected from 1482 adult participants. Univariable and multivariable logistic regression were performed to evaluate odd ratios.

RESULTS

The highest prevalence of S. mansoni, S. haematobium and co-infection of both species was 59.5%, 61.3% and 3.3%, in Andina and Ankazomborona respectively. Higher prevalence was observed among males (52.4%) and main contributors to the family income (68.1%). Not working as a farmer and higher age were found to be protective factors for infection.

CONCLUSIONS

Our findings provide evidence that adults are a high-risk group for schistosomiasis. Our data suggests that, for ensuring basic health as a human right, current public health strategies for schistosomiasis prevention and control need to be re-addressed towards more context specific, holistic and integrated approaches.

Open-source environmental data as an alternative to snail surveys to assess schistosomiasis risk in areas approaching elimination

Background: Although the presence of intermediate snails is a necessary condition for local schistosomiasis transmission to occur, using them as surveillance targets in areas approaching elimination is challenging because the patchy and dynamic quality of snail host habitats makes collecting and testing snails labor-intensive. Meanwhile, geospatial analyses that rely on remotely sensed data are becoming popular tools for identifying environmental conditions that contribute to pathogen emergence and persistence. Methods: In this study, we assessed whether open-source environmental data can be used to predict the presence of human Schistosoma japonicum infections among households with a similar or improved degree of accuracy compared to prediction models developed using data from comprehensive snail surveys. To do this, we used infection data collected from rural communities in Southwestern China in 2016 to develop and compare the predictive performance of two Random Forest machine learning models: one built using snail survey data, and one using open-source environmental data. Results: The environmental data models outperformed the snail data models in predicting household S. japonicum infection with an estimated accuracy and Cohen’s kappa value of 0.89 and 0.49, respectively, in the environmental model, compared to an accuracy and kappa of 0.86 and 0.37 for the snail model. The Normalized Difference in Water Index (NDWI) within half to one kilometer of the home and the distance from the home to the nearest road were among the top performing predictors in our final model. Homes were more likely to have infected residents if they were further from roads, or nearer to waterways. Conclusion: Our results suggest that in low-transmission environments, investing in training geographic information systems professionals to leverage open-source environmental data could yield more accurate identification of pockets of human infection than using snail surveys. Furthermore, the variable importance measures from our models point to aspects of the local environment that may indicate increased risk of schistosomiasis. For example, households were more likely to have infected residents if they were further from roads or were surrounded by more surface water, highlighting areas to target in future surveillance and control efforts.

Ecological Niche Model of Bacillus cereus Group Isolates Containing a Homologue of the pXO1 Anthrax Toxin Genes Infecting Metalworkers in the United States

While Bacillus cereus typically causes opportunistic infections in humans, within the last three decades, severe and fatal infections caused by isolates of the B. cereus group harboring anthrax toxin genes have been reported in the United States. From 1994 to 2020, seven cases of anthrax-like illness resulting from these isolates have been identified. With one exception, the cases have occurred in the Gulf States region of the United States among metalworkers. We aimed to develop an ecological niche model (ENM) to estimate a spatial area conducive to the survival of these organisms based on the presence of known human infections and environmental variables. The estimated ecological niche for B. cereus was modeled with the maximum entropy algorithm (Maxent). Environmental variables contributing most to the model were soil characteristics (cation exchange capacity, carbon content, soil pH), temperature, enhanced vegetation index (EVI), and land surface temperature (LST). Much of the suitable environments were located throughout the Gulf Coast Plain, Texas Backland Prairies, East Central Texas Plains, Edwards Plateau, Cross Timbers, Mississippi Alluvial Plain, and Central Great Plains. These findings may provide additional guidance to narrow potential risk areas to efficiently communicate messages to metalworkers and potentially identify individuals who may benefit from the anthrax vaccine.