Rift Valley fever (RVF), a disease transmitted from livestock (cattle, sheep, goats, camels) to humans more commonly occurs in the East and Central Africa (ECA) regions where more than 15 major epidemics affecting more than one country have been reported over the past 50 years. Within the region, there are specific areas, referred to as hotspots, which support RVF virus maintenance via low-level virus circulation between animals, humans, and mosquitoes. Most outbreaks originate from these hotspots. Our goal is to conduct studies in RVF hotspots in four ECA countries, Kenya, Uganda, Tanzania, and Democratic Republic of Congo (DRC) to determine the burden of RVF disease among humans, wildlife and livestock during inter-epidemic periods (IEPs) and discover circulation of undetected infectious diseases. This information is important for use in developing an early warning system and possibly a vaccination strategy. The study will take place in Uganda, Kenya, Tanzania and Democratic Republic of Congo

STUDY BACKGROUND

The East and Central Africa (ECA) region represents a unique environment to investigate RVF virus maintenance and transmission as it has reported the largest number of RVF epidemics. Recent studies have identified hotspots characterized by continuous low levels of human and livestock virus infections during the inter-epidemic periods (IEPs).

Although sheep are at highest risk of severe disease, how virus-host interactions differ between animal species is an area for further research. Seroprevalence and incidence studies at the wildlife, livestock, human interface are also important in understanding the nature of cross-species transmission. Our study will investigate the public health threat posed by RVFV during IEPs by determining the burden of human and livestock diseases, and the degree of sustained viral transmission in animal species during cryptic cycles. This will be carried out through surveillance among humans, livestock and wildlife. Understanding virus maintenance especially within the environment, vectors and wildlife populations will aid in identifying potential risk factors for RVFV infection in order to prevent future outbreaks.

While disease occurrence among animals may frequently go unnoticed due to inadequate animal surveillance systems and low levels of animal abortion, human disease may be the first indication that RVF virus amplification is occurring at high levels in a particular region . Furthermore, understanding the social aspects of disease transmission through human studies exploring human behaviour and cultural practices that lead to direct exposure to infected animal blood, tissues, secretions and excretions could provide valuable information to disease spill over dynamics . Additional knowledge of how routes for human exposure e.g. through mosquito bites or contact with infected animal products affect the immune response and disease outcomes is required . This will enhance global efforts in preparing for, and preventing, the possible spread of RVF disease to new regions.

While RVF may be an under-diagnosed cause of febrile illness within the ECA region it is important to consider the occurrence of other unknown or undiagnosed infectious diseases. Serious clinical disease due to a variety of pathogens in humans or animals is most likely missed in settings where health care access is limited and diagnostics scarce. Notably, multiple studies from the region report low levels of hospital visits, particularly among marginalized communities, including those in remote rural arid lands that have few health facilities and poorly functioning transport infrastructure that discourages traveling long distances to access health care. Over a third of suspected infectious diseases among humans go undiagnosed even when considerable testing is carried out; thus, many retrospective studies of emerging pathogens uncover cases in the region that occurred long before the pathogen was identified and epidemics reported.

HYPOTHESIS AND OBJECTIVES

Hypothesis and Goal

Our working hypothesis is that RVF virus, maintained through cryptic vertebrate-mosquito cycling during inter-epidemic periods is a substantial public health burden. The overall goal is to determine the public health threat posed by RVF virus during IEPs in East and Central Africa region and identify potential opportunities for prevention and control strategies to reduce the likelihood of major outbreaks.

Primary objectives

(i) Determine whether low-level RVF virus transmission and disease among or between animals and humans occurs during the inter-epidemic periods (IEPs)

(ii) Conduct detailed niche modelling of RVF high risk and low risk ecologies to determine the important ecological risk factors associated with persistence of RVF virus and recurrence of outbreaks in animals and humans

(iii) Investigate impact of climate change on the RVF permissive ecologies and its effect on RVF virus transmission

Secondary objective

(iv) Conduct pathogen discovery on human, livestock, and wildlife samples collected from ECA countries to identify other circulating emerging infectious pathogens which are of epidemic or pandemic potential.

DESIGN AND METHODOLOGY

We will conduct the field studies in four countries as listed below, but with the understanding that other emerging RVF hotspots in the region could be added.

(i) Tana river, Marsabit, Isiolo and Murangá counties in Kenya (ii) Kabale, Rubanda and Isingiro districts in Uganda (iii) North and South Kivu provinces in DRC (iv) Any other site in the 4 countries emerging as a RVF hotspot or site of

STUDY PROCEDURES

1. In and outpatient health facility-based studies:

   We will conduct a 2-year hospital-based study at each site to determine the burden of RVF infections in humans, risk factors associated with infection, and severity of disease during IEPs. A sample size of 707-1,500 human participants will be recruited over the 2 years at each site, based on assumptions of RVFV seroprevalence. At each health facility, longitudinal acute febrile illness (AFI) surveillance will be established. Patients, both adults and children presenting with acute fever or reported fever in the last 4 weeks, will be included prospectively over a period of two years in order to capture seasonal variability. We will focus on cases of undifferentiated fever, hence those with a clearly defined clinical disease, for example, malaria, an acute upper respiratory tract infection or urinary tract infection will be excluded. However, twenty percent of samples from patients who test positive for malaria by rapid diagnostic test or blood smear will be included in the study as there are common risk factors for both malaria and RVF infection. A clinical history and physical examination by a clinical officer/nurse on-ground will provide information on clinical manifestation and severity of illness. Consenting patients will be enrolled, serum collected and tested for RVFV RNA, and IgM and IgG antibodies. Convalescent serum will be collected 4 weeks after the date of enrolment for RVFV antibodies. In addition, the individual may be contacted to provide blood specimen up to 24 months after enrolment for analysis of peripheral blood mononuclear cells (PBMCs). The specimens will also be tested for other etiology for acute febrile illness including brucellosis, q-fever, coronaviruses and potentially other novel and emerging pathogens by molecular methods. Using a standard questionnaire, we will collect information on contact with domestic animals including assisting with birth, slaughter, handling hides and skins, and consuming raw milk or blood. Data on contact with wildlife including eating bush meat will also be collected.

2. Community studies:

   To determine burden of disease in livestock, 2 - 6 cross-sectional surveys at each site will be conducted over 5 years, primarily during rainy seasons. The livestock sample size, assuming an average RVF seroprevalence of 4%, precision of 2%, confidence level of 95%, and power of at least 80% is 369 animals per survey. To account for herd clustering, a design effect of 1.5 is included for a total of 554 animals per site for each survey. The number of cattle, sheep, goats will be selected proportionate to animal species population size in the area. Where no animal census records are available in an area the number of animals per species will be divided equally (approx. 185 animals of each species). At the Marsabit site, camels will be included. At the start of the survey, we will develop a sampling frame of the number of households in the study site. To provide a geographically representative sample, households will be randomly selected from the study site. The breed of animal whether exotic or indigenous will be captured. Livestock blood samples will be collected and tested for RVFV RNA, and anti-viral IgM and IgG antibodies.

   The household members of the selected field sites will be included in a seroprevalence (and sero-conversion) study, to be able to link animal, vector and human infection. All consented household members will be included. Each participant will provide a sample for serological analysis and social behaviour will be assessed, using the standard questionnaire - to collect information on contact with domestic animals including assisting with birth, slaughter, handling hides and skins, and consuming raw milk or blood. Data on contact with wildlife including eating bush meat will also be collected.

3. Slaugther house studies:

   Sampling of cattle and goats will also be performed at slaughter houses in Goma. Most of these animals originate from neighbouring countries. Sheep that are rarely slaughtered at Goma will also be sampled when available. Slaughter house sampling will be conducted at the same occasion with livestock sampling. All the samples analysed using serological and molecular techniques for the detection of anti-RVFV antibodies and RVFV nucleic acid.

4. Wildlife sampling:

   Convenient wildlife sampling will be carried out in the field sites or nearby national conservation areas in Kenya, Uganda, Tanzania and DRC. Primarily blood/serum samples collected by wildlife officers during animal disease surveillance and/or translocation; and blood/serum and tissue samples from dead or killed wildlife will be collected. We will also obtain archived samples from these locations. The samples will be tested for RVF RNA and antibodies, and for pathogen discovery to detect other circulating emerging pathogens.

5. Vector studies:

   Our approach to investigating the role of vectors in cryptic cycles during IEPs will be guided by the fact that only a small proportion of mosquitoes are normally infected, even during epidemics. Therefore, to identify which areas to test for presence of infected mosquitoes we will interpret acute RVFV infections (presence of viral RNA) as a measure of presence of infected mosquitoes and mosquito-to-vertebrate transmission. In these areas with acute RVFV infection we will collect and test mosquito pools in the community grazing areas and households during livestock sampling, identify the species and test for RVFV RNA.

   Since it is difficult to establish colonies of the known RVF vectors, Aedes mcintoshi and Aedes ochraceus, for laboratory studies, we will carry out suitability studies in the field using two approaches. First, eggs and larvae of mosquitoes will be collected during the long (April - July) and short (October - November) rains from dambos located at low elevation in all five field sites in Kenya, Uganda, Tanzania, and DRC and geocodes recorded in areas around the study site that have solonetz, planosol, solonchak, or vertisol soil types. Larvae in 3rd, 4th stages and pupae will be collected using dippers into plastic whirl-pack bags for transportation (in cool boxes) to the laboratory. Once in the laboratory, the immatures will be reared to adults and then identified morphologically. Adult mosquitoes around dambo habitats will be sampled using odour-baited traps including collections of those resting on vegetation using backpack aspirators for eventual transport and sorting according to site and species. The physico-chemical properties of water from selected dambos will be determined and the soil types identified. Water from each of the dambos will be subject to headspace volatile trapping and chemical analyses to identify potential oviposition cues for eventual development of tools for monitoring gravid mosquito cohorts.

   For vector adaptation studies, A. mcintoshi and A. ochraceous mosquitoes collected from the five field sites will be subjected to the expected temperatures and rainfall/water conditions and their adaptability assessed. Assessment of adaptability will include observations of overall survival, egg survival, changes in biting habits, feeding preferences and resting behaviour within a controlled environment.

6. RVF modelling studies:

   Modelling studies support RVFV transmission during IEPs in areas associated with RVF epidemics. The studies predict that the continuous cycling of RVFV is powered by presence of mosquito vectors throughout the year (but varying densities), and rapid turnover of livestock populations, particularly sheep, goats, and cattle to sustain the high herd susceptibility required for new infections. We will use mathematical modelling to predict the timelines and locations of major RVF human and domestic animal epidemics; these data will in turn guide prevention and control efforts. We will investigate why there are areas with RVF circulation but no apparent human infections considering that some of these areas are adjacent to areas where epidemics occur. This is important since it is possible that RVFV circulating at low levels has the potential for causing human epidemics during El Nino events. Human, livestock, and vector data from all field sites conducting RVFV transmission studies will be fed into the model. Model parameters will be optimized by fitting simulations to field observations (e.g. the seroprevalence data in humans or livestock) using Bayesian inference techniques (e.g. MCMC and POMP models). Comparative analysis of all study sites will help elucidate the factors associated with RVFV transmission and emergence. The model will factor data from areas associated with RVFV epidemics (e.g. Tana river in Kenya and Kabale district in Uganda), and areas where virus circulation is evident but no epidemics have occurred (e.g. Udzungwa Mountains in Tanzania) in order to understand differences in transmission dynamics. These data may give new insights on the epidemic risk. Comparative analysis will look into the spatio-temporal heterogeneity of transmission in the four countries. Other important data to be collected from our field studies and published literature include vector species/vector density/vector RVFV positivity, animal presence/animal density/animal RVFV positivity, human-animal contact, population density/population positivity, spectrum of disease presentation and severity, co-circulation of other arboviruses, and genetic characterization of RVFV in various areas.

7. Testing of archived samples:

This approach will be utilised in DRC using samples collected from the following studies:

Study A: A cross-sectional study including outpatient acute febrile syndromes in both children and adults, between November 2015 and June 2016 in Kinshasa, DRC: 342 patients, aged 2 to 68 years, were included. The study was registered in a public repository (https://www.clinicaltrials.gov/ct2/show/NCT02656862); and funded under framework agreement between the Institute of Tropical Medicine and the Belgian development cooperation and Vlaamse Interuniversitaire Raad - Universitaire Ontwikkelingssamenwerking (VLIR-UOS, Grant reference ZRDC2014MP083). These sampes will be analysed using serological and molecular techniques for the detection of anti-RVF antibodies or RVFV nucleic acid.

Study B: Archived samples were collected from cattle and goats in Goma abattoirs, and in cattle farms in the North-Kivu province in 2017. Some of these animals originated from neighboring countries and provide useful materials to assess the risk of RVFV circulation in the region. Serological and molecular techniques will be used for the detection of anti-RVF antibodies and RVFV nucleic acid.

Study C: Archived samples collected from wild animals in Virunga Park and kept at the Veterinary Laboratory in Goma will be screened for the presence of anti-RVF antibodies and RVFV nucleic acid. Most of these samples originated from non-human primates raised in close contact with the local population. Such samples will provide useful information on the circulation of the RVF at the interface wild animal - human interface within the game park.

Inclusion Criteria for health facility based study:

Category A: A sample of patients presenting to the health facility meeting this inclusion criteria will be enrolled:

• Persons ≥ 10 years of age who are malaria negative AND have undifferentiated acute fever at the time of presentation (≥ 37.5°C) or reported fever in the past 4 weeks.

Category B: A sample of patients presenting to the health facility meeting this inclusion criteria will be enrolled:

• Persons ≥ 10 years of age who are malaria positive AND have undifferentiated acute fever at the time of presentation (≥ 37.5°C) or reported fever in the past 4 weeks.

Category C: All patients at the health facility meeting these criteria will be enrolled in the study

Persons ≥ 10 years of age with:

• Unexplained bleeding with or without fever manifesting as either: Blood in vomitus, Bleeding from the gums, Bleeding from the nose, Bleeding in the eyes (red eyes), Non-menstrual genital bleeding, Bleeding from any other body site OR

• Infectious disease illness of unknown etiology requiring hospitalization. The illness should not be responding to antimalarials and/or antibiotics following 7 days of treatment

Inclusion Criteria for community cross sectional study:

Member of household 2 or more years of age.

Exclusion Criteria:

Those who do not consent or refuse written consent. If a participant is severely ill and has no next of kin to provide consent +/- witness on their behalf.

If a participant is known to have blood clotting disorders, allergies arising from injections, or if a participant has ever fainted during previous injections or blood collection

For health facility based study: Patients attending enrolled health facilities in study areas i.e. North and South Kivu (DRC), Isingiro, Rubanda and Kabale districts (Uganda), Muranga, Isiolo, Tana River and Marsabit counties (Kenya) and Iringa district (Tanzania).

For community cross sectional survey: Residents of households in study areas i.e. North and South Kivu (DRC), Isingiro, Rubanda and Kabale districts (Uganda), Muranga, Isiolo, Tana River and Marsabit counties (Kenya) and Iringa district (Tanzania).

There are no plans as yet to make individual participant data available. However, researchers may contact the PI for individual requests.