Background. The 2018-2020 Ebola virus disease outbreak in the Democratic Republic of the Congo (DRC) was the country's largest, and the second largest globally, amid armed conflict and community mistrust. Transmission heterogeneity (superspreading) is recognised in Ebola epidemics, but empirical estimates of its extent and determinants remain scarce for DRC outbreaks. We quantified transmission heterogeneity and its determinants during this outbreak. Methods. In this retrospective observational study, we reconstructed transmission chains for confirmed and probable cases (Aug 1, 2018, to June 25, 2020) using routinely collected Ministry of Health and Medecins Sans Frontieres surveillance data. We modelled the offspring distribution with a Bayesian negative binomial framework, correcting for incomplete contact tracing, to estimate the effective reproduction number (Reff), dispersion parameter (k), and proportion of cases responsible for 80% of transmission (prop80), overall, by subgroup, and over time. Individual-level determinants were assessed with a regression extension, adjusting for covariates. Findings. Among 3481 cases, 2008 transmission events linked 2402 (69%) individuals into 415 chains (median size 3, range 2-102). Overall Reff was 1.00 (95% CI 0.92-1.08) with k 0.29 (0.26-0.32); 17.8% of cases generated 80% of transmission. Overdispersion stayed stable despite fluctuating Reff. Non-isolation (IRR 1.79), death outside a treatment centre (IRR 4.34), and unfollowed contact status (IRR up to 4.48) predicted more secondary cases; vaccination cut transmission by about 60%. Interpretation. Epidemiological investigations linked 67.7% (2356/3481) of cases into 415 transmission chains (median size 3, range 2-102); linkage to a known infector fell to 10% during the November 2018-February 2019 period of peak insecurity. Transmission was heterogeneous overall, with dispersion parameter k of 0.29 (0.26-0.32), such that 17.8% (16.8-18.9) of cases generated 80% of onward transmission confirming superspreading as a stable, structural feature of Ebola dynamics. Critically, k remained stable throughout the outbreak, including during periods of elevated Reff, indicating that transmission surges reflected intensification of the same underlying process rather than new superspreading contexts, and that Reff alone is an insufficient summary of epidemic potential. Regression analyses identified predominantly modifiable determinants: cases not isolated in an Ebola treatment centre (IRR 1.79 [1.54-2.06]) or who died outside one (IRR 4.34 [3.47-5.31]) generated substantially more secondary cases, as did those registered as contacts but not followed up (IRR 3.01 [2.39-3.70]) or unregistered altogether (IRR 4.48 [3.67-5.38]) relative to actively followed-up contacts. Vaccination reduced onward transmission by 60-64% (IRR 0.36-0.40). These findings indicate that transmission was shaped less by gaps in epidemiological knowledge than by the operational reach of contact tracing, isolation, and vaccination delivery, particularly during periods of insecurity.

Abstract Background Seasonal malaria chemoprevention (SMC) is a malaria intervention in which antimalarial drugs are administered monthly to children under 5 years of age during the high-transmission season. In the district of Moissala in southern Chad, SMC has been implemented since 2013, with an interruption in 2019, resumption in 2020, and expansion to five rounds of treatment in 2021. Recent World Health Organization (WHO) guidelines allow countries to adapt the timing and number of SMC rounds to local transmission patterns, creating a need to identify optimal strategies for each setting. In this study, we used mathematical modeling for three primary purposes: 1) to estimate the effectiveness of SMC in Moissala from 2018 to 2023, 2) to assess the impact of changes to SMC strategies since 2018, and 3) to determine the optimal SMC strategy in Moissala. Methods and findings We adapted a compartmental, climate-informed malaria transmission model to represent malaria dynamics in the presence of SMC. The model incorporates temperature and rainfall data to capture how climate variability influences malaria transmission over time. It was calibrated to routine surveillance data on malaria cases in children under five years old from 2018 to 2023. Using the calibrated model, we simulated malaria cases under alternative scenarios, including the absence of SMC and variations in the number and timing of SMC rounds. These simulations were then used to estimate the overall effectiveness of SMC, assess the impact of past changes in SMC strategies, and identify the optimal strategy in Moissala. Between 2018 and 2023, SMC reduced malaria cases in children under five by 26% (95% credible interval: 21%, 31%) relative to a scenario without SMC, corresponding to an average of approximately 14400 cases averted each year. The interruption of SMC in 2019 led to an estimated increase of 13600 cases (95% credible interval: 11200, 15800), representing a 31% rise during the high-transmission season. Expanding from four to five SMC rounds in 2021 reduced cases by 7% relative to a four-round schedule, while starting the five-round schedule earlier in June rather than July led to an additional 5% reduction. Overall, the most effective strategy from 2018 to 2023 was a five-round schedule beginning in mid-June. Conclusions Seasonal malaria chemoprevention has substantially reduced malaria incidence among children under five in Moissala. The currently implemented strategy of five rounds of SMC starting in June was estimated to achieve the greatest reduction in cases over the study period. Climate-informed modelling and open-source software can support timely decision-making across settings under changing climate and transmission conditions.

The Coalition for Epidemic Preparedness Innovations’ (CEPI) 100 Days Mission, and broader global pandemic preparedness efforts, require overcoming challenges in evaluating vaccine efficacy during emerging epidemics and outbreaks. Key challenges, addressed in a workshop jointly hosted by CEPI and the University of Oxford’s Pandemic Sciences Institute, arise from the sporadic and heterogeneous nature of outbreaks, high fatality rates that may preclude traditional placebo-controlled trials and biases from small or real-world studies. Here, we review recent developments, potential solutions and areas for innovation emphasized in the workshop to address these challenges. These opportunities for research and methodological development encompass four main areas: serology, exposure measurements, statistical modelling and trial design.

BACKGROUND:Lassa fever is an emerging zoonotic disease endemic to west Africa. Several vaccines aimed at preventing Lassa fever are currently under development, creating a need to assess how best to administer them once licensed for human use. We aimed to project the health-economic burden of Lassa fever from 2025 to 2037 across age and sex groups in subnational administrative divisions of west Africa with endemic Lassa mammarenavirus transmission and to estimate the cost-effectiveness of targeting Lassa vaccination to different risk groups. METHODS:In this vaccine-impact modelling study, we developed a mathematical model using a zoonosis risk map and epidemiological data from recent and ongoing cohort studies to predict the health-economic burden of Lassa fever across age and sex groups in endemic regions. We simulated vaccination campaigns targeting different risk groups to estimate the cost-effectiveness of various strategies for Lassa vaccine administration. Threshold vaccine costs (TVCs), which represent the break-even price per dose of vaccine administered, were estimated in international dollars (INT$ 2023), accounting for health-care costs, productivity losses, and monetised disability-adjusted life-years (DALYs) averted by vaccination. FINDINGS:Lassa fever was estimated to cause 6·23 (95% uncertainty interval (UI) 4·21-8·42) hospitalisations, 0·75 (0·48-1·10) deaths and 31·1 (17·7-52·2) DALYs per 100 000 person-years. Vaccine strategies targeting adolescents-adults aged 15-49, older adults aged 50 years and older, and women of childbearing age (WCBA) aged 15-49 years prevented, respectively, the most hospitalisations, deaths, and DALYs per 100 000 vaccine doses. Under base case assumptions, the most cost-effective strategy (greatest net monetary benefit) was untargeted vaccination for a vaccine costing INT$2 per dose, and targeting adolescents-adults at $5 per dose. At $10 per dose or more, none of the considered strategies were cost-effective. The highest TVC for a single-dose vaccine was estimated at $7·39 (95% UI 4·33-11·60) when targeting adolescents-adults, followed by $6·69 (4·17-9·85) when targeting older adults, $6·10 (3·56-9·74) when targeting WCBA, and $1·94 (1·10-3·10) when targeting children. INTERPRETATION:Targeting of adolescents-adults appears to generate the greatest health-economic value per vaccine dose. However, the most cost-effective vaccination strategy will depend on vaccine price. FUNDING:Coalition for Epidemic Preparedness Innovations.

ABSTRACT: Introduction:Lassa fever (LF), a viral haemorrhagic disease, poses a significant public health challenge in West Africa. Lassa virus infection frequently causes mild malaria-like symptoms, potentially leading to misdiagnosis and an underestimated burden. Severe LF can lead to multi-organ failure, and survivors may experience sensorineural hearing loss (SNHL). Building on the contributions of the Enable Lassa Research Programme (ENABLE 1.0), which ran in West Africa from 2020 to 2024, ENABLE 1.5 aims to further address gaps in understanding LF disease burden to inform future late-stage vaccine trials. The study will assess the incidence of symptomatic reverse transcription (RT)-PCR-confirmed LF disease, including malaria coinfection. Methods and analysis:The ENABLE 1.5 prospective cohort study will be conducted across five study sites: one in Liberia, three in Nigeria and one in Sierra Leone. Stratified cluster sampling will identify eligible individuals at the household level from communities either involved in ENABLE 1.0 or identified through recent LF surveillance as hotspots. A total of 5000 participants will be recruited, 1000 per study site (minimum) and equally stratified in the following ages: 0-5, 6-10, 11-17, 18-50 and >50 years. All participants will be followed up for 12 months. Baseline data collection will gather key variables and blood specimens from all participants, with baseline SNHL prevalence assessed at three study sites. Active follow-up of all participants will involve symptom assessments every 2 weeks and blood draws every 3 months for serological testing (IgG). Suspected LF cases will undergo thorough evaluations, including malaria rapid diagnostic testing, clinical assessments and laboratory testing, including RT-PCR and malaria blood smear microscopy.

Clinical trials in settings with intermittent or non-existent internet and power connectivity, for example during humanitarian emergencies, present challenges in the synchronisation of data across different sites, in addition to accessing a centralised database in real-time. To overcome these, we designed a novel hybrid analogue/digital data management system which was deployed during the rapid implementation of a Phase III evaluation of a two-dose preventative vaccine for Ebola virus disease in Goma, Democratic Republic of the Congo, from 2019 to 2022. We provided study participants with an Enhanced Participant Record Card (EPRC) that served as eligibility for, and confirmation of, vaccination and was used in combination with Open Data Kit (ODK) electronic case report forms to create an off-grid study participant management system. To understand the utility of the EPRC, we analysed data from 15,327 study participants who received both vaccines and various types of prompts or reminders to return for dose 2, including home visits, telephone calls, or short messaging service (SMS). A total of 53% participants referred to the date on the EPRC as a prompt to return for dose 2 and 36.1% mentioned this as the only prompt. A multivariable generalised linear mixed-effects model showed that those who were not working, those aged 45-64 years or who had a chronic medical condition identified prior to receiving dose 2 were more likely to use the date on the EPRC as a prompt. Our findings demonstrate the utility of this system in the facilitation of decentralised data collection in off-grid locations that may be useful for future trials in complex humanitarian settings. Clinical Trials Registration Number: ClinicalTrials.gov NCT01128790.

BACKGROUND:Partway into the 2018-20 Ebola outbreak in the Democratic Republic of the Congo (DR Congo), a new strategy of decentralised care was initiated to address delays in care seeking, improve community acceptance, and reduce the risk of Ebola virus disease (EVD) transmission through early case isolation. Unlike centralised EVD facilities (transit and treatment centres), which operated in parallel to the existing health-care system and focused exclusively on EVD, decentralised facilities were integrated into existing health-care structures with which communities were already familiar, and designed to continue providing health care for patients with other non-EVD illnesses. Here we aim to assess the strategy of decentralised care by comparing admission delays and patient outcomes among the three types of EVD facilities (decentralised, transit, and treatment). METHODS:We performed a retrospective analysis of routinely collected data from all individuals admitted to EVD facilities (12 treatment, nine transit, and 21 decentralised facilities) at any point during the Ebola outbreak from July 27, 2018, to June 24, 2020 in DR Congo. We used multivariate mixed-effect regression to model admission delays (the number of days between symptom onset and admission to an EVD facility) and patient outcomes (survived or died), as functions of facility type at first admission and date of admission, while controlling for a variety of other covariates. FINDINGS:Over the course of the outbreak 60 465 patients were admitted to EVD facilities, of which 2289 (3·8%) were confirmed to be EVD positive. Covariate-adjusted admission delays were somewhat higher among patients presenting to transit facilities (adjusted rate ratio 1·14 [95% CI 0·95-1·32]) or treatment facilities (1·18 [1·00-1·36]) compared with decentralised facilities. Similarly, compared with decentralised facilities, adjusted case-fatality risks were slightly higher among patients presenting to transit facilities (adjusted risk ratio 1·04 [0·82-1·26]) or treatment facilities (1·03 [0·82-1·24]). INTERPRETATION:As was observed during the 2013-16 west Africa outbreak and the 2020 outbreak in the Equateur province of DR Congo, patients suspected of EVD that presented to decentralised facilities had modestly shorter admission delays than patients presenting to centralised facility types. Case-fatality risks were slightly lower among patients presenting to decentralised facilities; however, this finding was not statistically significant and so it is difficult to assess the generalisability. FUNDING:Médecins Sans Frontières. TRANSLATION:For the French translation of the abstract see Supplementary Materials section.

Introduction Lassa fever (LF), a viral zoonotic disease endemic to West Africa, often causes no or mild, non-specific symptoms, but severe cases can result in haemorrhage, multi-organ failure, and death. Its burden remains poorly defined, yet, is essential for guiding vaccine trials. Here, we report on symptomatic LF incidence in a Nigerian site of the Enable Lassa research programme. Methods A prospective community-based longitudinal cohort enrolled participants ≥2 years old in Edo State, Nigeria, from 2020. Participants were followed up for 30 months and monitored for acute febrile illness through active/passive surveillance. Participants meeting the acute febrile case definition were tested for LF by RT-PCR using acute blood samples. LF positive cases were hospitalised with hearing assessed at discharge and four months post-discharge to determine sensorineural hearing loss (SNHL). Results Of 5,025 participants recruited, 3,543 suspected LF cases were assessed, with 23 confirmed by RT-PCR, yielding an overall incidence of 1.90 (95%CI 1.20–2.85) per 1,000 person-years. Symptoms included headache (91%), abdominal pain (83%), muscle/joint pain (57%) and vomiting (35%). Two cases were fatal (CFR 9%). Nine of 21 (43%) LF cases with hearing tests performed had SNHL at discharge, increasing to 13 at four months follow up. Conclusion The LF incidence and symptoms data from the cohort reveal critical insights into disease burden, location, and long-term impact.

Abstract Background Zaire Ebolavirus disease (EVD) outbreaks can be controlled using rVSV-ZEBOV vaccination and other public health measures. People in high-risk areas may have pre-existing antibodies from asymptomatic Ebolavirus exposure that might affect response to rVSV-ZEBOV. Therefore, we assessed the impact pre-existing immunity had on post-vaccination IgG titre, virus neutralisation, and reactogenicity following vaccination. Methods In this prospective cohort study, 2115 consenting close contacts (“proches”) of EVD survivors were recruited. Proches were vaccinated with rVSV-ZEBOV and followed up for 28 days for safety and immunogenicity. Anti-GP IgG titre at baseline and day 28 was assessed by ELISA. Samples from a representative subset were evaluated using live virus neutralisation. Results Ten percent were seropositive at baseline. At day 28, IgG in baseline seronegative (GMT 0.106 IU/ml, 95% CI: 0.100 to 0.113) and seropositive (GMT 0.237 IU/ml, 0.210 to 0.267) participants significantly increased from baseline (both p < 0.0001). There was strong correlation between antibody titres and virus neutralisation in day 28 samples (Spearman’s rho 0.75). Vaccinees with baseline IgG antibodies against Zaire Ebolavirus had similar safety profiles to those without detectable antibodies (63.6% vs 66.1% adults experienced any adverse event; 49.1% vs 60.9% in children), with almost all adverse events graded as mild. No serious adverse events were attributed to vaccination. No EVD survivors tested positive for Ebolavirus by RT-PCR. Conclusions These data add further evidence of rVSV-ZEBOV safety and immunogenicity, including in people with pre-existing antibodies from suspected natural ZEBOV infection whose state does not blunt rVSV-ZEBOV immune response. Pre-vaccination serological screening is not required.

BACKGROUND:The recombinant vesicular stomatitis virus-Zaire Ebola virus (rVSV-ZEBOV) vaccine is the only WHO prequalified vaccine recommended for use to respond to outbreaks of Ebola virus (species Zaire ebolavirus) by WHO's Strategic Advisory Group of Experts on Immunization. Despite the vaccine's widespread use during several outbreaks, no real-world effectiveness estimates are currently available in the literature. METHODS:We conducted a retrospective test-negative analysis to estimate effectiveness of rVSV-ZEBOV vaccination against Ebola virus disease during the 2018-20 epidemic in the Democratic Republic of the Congo, using data on suspected Ebola virus disease cases collected from Ebola treatment centres. Those eligible for inclusion had an available Ebola virus RT-PCR result, available key data, were eligible for vaccination during the outbreak, and had symptom onset aligning with the period in which a ring-vaccination protocol was in use. After imputing missing data, each individual confirmed by RT-PCR to be Ebola virus disease-positive (defined as a case) was matched to one individual negative for Ebola virus disease (control) by sex, age, health zone, and month of symptom onset. Effectiveness was estimated from the odds ratio of being vaccinated (≥10 days before symptom onset) versus being unvaccinated among cases and controls, after adjusting for the matching factors. The imputation, matching and effectiveness estimation, was repeated 500 times. FINDINGS:1273 (4·8%) of 26 438 eligible individuals were positive for Ebola virus disease (cases) and 25 165 (95·2%) were negative (controls). 40 (3·1%) cases and 1271 (5·1%) controls were reported as being vaccinated at least 10 days before symptom onset. After selecting individuals who reported exposure to an individual with Ebola virus disease within the 21 days before symptom onset and matching, the analysis datasets comprised a median of 309 cases and 309 controls. 10 days or more after vaccination, the effectiveness of rVSV-ZEBOV against Ebola virus disease was estimated to be 84% (95% credible interval 70-92). INTERPRETATION:This analysis is the first to provide estimates of the real-world effectiveness of the rVSV-ZEBOV vaccine against Ebola virus disease, amid the widespread use of the vaccine during a large Ebola virus disease outbreak. Our findings confirm that rVSV-ZEBOV is highly protective against Ebola virus disease and support its use during outbreaks, even in challenging contexts such as in the eastern Democratic Republic of the Congo. FUNDING:Médecins Sans Frontières. TRANSLATION:For the French translation of the abstract see Supplementary Materials section.

Outbreaks such as mpox and COVID-19 underscore the need for swift, evidence-based responses to limit their spread and save lives. Such evidence should be generated in real time. Generating these insights requires synthesis of many different sources of information and a combination of multiple distinct analytical techniques, from basic descriptive analysis to complex dynamic models.1Polonsky JA Baidjoe A Kamvar ZN et al.Outbreak analytics: a developing data science for informing the response to emerging pathogens.Philos Trans R Soc Lond B Biol Sci. 2019; 37420180276Crossref PubMed Scopus (99) Google Scholar Timely generation of policy-relevant evidence relies on skilled staff and prebuilt outbreak analytics pipelines, alongside rapid literature reviews and ad-hoc data curation processes. Although these workflows are constantly improving, they continue to involve several time-intensive manual tasks such as abstract selection and data extraction, cleaning, and merging. Performing subsequent research and writing up results are also time intensive. At some instances, specific skills such as coding are required. In particular, multiple governments or organisations require regular situational updates against a background of changing data and epidemic dynamics, creating major bottlenecks in the ability to deliver analysis at scale. To improve and enhance the efficiency of real-time outbreak analytics and to reduce the number of potential bottlenecks, considerable opportunity to make use of newly emerging large language models (LLMs) exists. Recently, several advanced LLMs have become available, including GPT-4 from OpenAI and Gemini from Google. These LLMs are equipped with capabilities such as generating and interpreting text, writing and executing code, performing analytical tasks, and processing images. Integration of LLMs within outbreak analytic workflows can be achieved both directly and indirectly. The direct approach would be for analysts to interact with LLMs directly to assist in writing codes for analysis or to provide tailored feedback on written text, similar to the common tools currently available. The indirect approach would involve having the analyst interact with specialised LLM-based agents. LLM-based agents are computer programs that can deploy multiple LLMs with specific roles and specialisations to perform dedicated tasks, similar to how human teams divide projects on the basis of the experience and knowledge of the team members.2Xi Z Chen W Guo X et al.The rise and potential of large language model based agents: a survey.arXiv. 2023; (preprint)https://arxiv.org/abs/2309.07864Google Scholar Such agents can perform tasks to achieve specific goals, either on their own or collaboratively with other agents, with the outputs of one agent's task becoming an input for the next agent or in dialogue with the analyst. We observe several key advantages in utilising specialised LLM-based agents. A primary advantage of using LLM-based agents, compared with using LLMs directly through prompting, lies in their ability to formalise LLM team interactions, enabling optimisation of workflows and quality control to enhance consistency and reproducibility. Furthermore, using multiple agents allows for more review steps before presenting the answer to the analyst, which can yield a higher-quality output than that obtained using direct prompting. Moreover, simultaneous deployment of LLM-based agents allows multiple tasks to be performed simultaneously and at scale. Through collaboration, multiple agents could complete more complex tasks. Finally, LLM agents, equipped with memory capabilities, enhance their performance over time by learning from their previous experience to accomplish tasks more effectively. Given these characteristics and the anticipated increasing capabilities of LLMs, developing several specifically defined LLM-based agents as part of the outbreak analytics pipeline has considerable potential for improving its efficiency. Once developed and tested, these workflows can be scaled and replicated globally, providing round-the-clock operation and support and creating a more equitable distribution of analytical knowledge and expertise. However, as with all new technologies, the integration of these new LLMs should be approached with caution. Pilot studies, real-world testing, and validation against established benchmarks are necessary to understand the limitations of the models; learn about their stability and costs; acquire experience with optimal use; learn all ethical implications; and develop consistency, trust, and acceptance. This learning will also allow for implementation of necessary governance structures and oversight to prevent misuse or unintended consequences, as misinterpretation of data or failure to capture nuances could lead to erroneous conclusions or recommendations, having serious consequences. Furthermore, integrating the use of LLMs will require training of staff and adaptation of hardware and software infrastructures to support these pipelines. Although LLMs in outbreak responses can lead to important efficiency gains in time and quality, this article is not a call for all tasks to be replaced by LLM agents. A sufficiently nuanced interpretation of outbreak data is difficult, with profound consequences, for which human skill, responsibility, and input are key. Therefore, creating an environment in which skilled staff can focus more of their time on these problems is crucial. Cautious use of this approach might also empower less technically trained staff to solve these problems. We declare no competing interests. During the preparation of this manuscript, the author used OpenAI GPT-4 using GPT "Academic Assistant Pro" to generate an initial sketch draft based on an extended outline with relevant topics in bullet points produced by the main author. After using this tool/service, the authors substantially reworked and edited the content and then performed an AI check for grammar and spelling of a near final version. The authors take full responsibility for the content of the publication.

During the 2018–2020 Ebola virus disease outbreak in Democratic Republic of the Congo, a phase 3 trial of the Ad26.ZEBOV, MVA-BN-Filo Ebola vaccine (DRC-EB-001) commenced in Goma, with participants being offered the two-dose regimen given 56 days apart. Suspension of trial activities in 2020 due to the COVID-19 pandemic led to some participants receiving a late dose 2 outside the planned interval. Blood samples were collected from adults, adolescents, and children prior to their delayed dose 2 vaccination and 21 days after, and tested for IgG binding antibodies against Ebola virus glycoprotein using the Filovirus Animal Nonclinical Group (FANG) ELISA. Results from 133 participants showed a median two-dose interval of 9.3 months. The pre-dose 2 antibody geometric mean concentration (GMC) was 217 ELISA Units (EU)/mL (95% CI 157; 301) in adults, 378 EU/mL (281; 510) in adolescents, and 558 EU/mL (471; 661) in children. At 21 days post-dose 2, the GMC increased to 22,194 EU/mL (16,726; 29,449) in adults, 37,896 EU/mL (29,985; 47,893) in adolescents, and 34,652 EU/mL (27,906; 43,028) in children. Participants receiving a delayed dose 2 had a higher GMC at 21 days post-dose 2 than those who received a standard 56-day regimen in other African trials, but similar to those who received the regimen with an extended interval.

During the 2018–2020 Ebola virus disease (EVD) outbreak, residents in Goma, Democratic Republic of the Congo, were offered a two-dose prophylactic EVD vaccine. This was the first study to evaluate the safety of this vaccine in pregnant women. Adults, including pregnant women, and children aged ≥1 year old were offered the Ad26.ZEBOV (day 0; dose 1), MVA-BN-Filo (day 56; dose 2) EVD vaccine through an open-label clinical trial. In total, 20,408 participants, including 6635 (32.5%) children, received dose 1. Fewer than 1% of non-pregnant participants experienced a serious adverse event (SAE) following dose 1; one SAE was possibly related to the Ad26.ZEBOV vaccine. Of the 1221 pregnant women, 371 (30.4%) experienced an SAE, with caesarean section being the most common event. No SAEs in pregnant women were considered related to vaccination. Of 1169 pregnancies with a known outcome, 55 (4.7%) ended in a miscarriage, and 30 (2.6%) in a stillbirth. Eleven (1.0%) live births ended in early neonatal death, and five (0.4%) had a congenital abnormality. Overall, 188/891 (21.1%) were preterm births and 79/1032 (7.6%) had low birth weight. The uptake of the two-dose regimen was high: 15,328/20,408 (75.1%). The vaccine regimen was well-tolerated among the study participants, including pregnant women, although further data, ideally from controlled trials, are needed in this crucial group.

Background Lassa fever (LF), a haemorrhagic illness caused by the Lassa fever virus (LASV), is endemic in West Africa and causes 5000 fatalities every year. The true prevalence and incidence rates of LF are unknown as infections are often asymptomatic, clinical presentations are varied, and surveillance systems are not robust. The aim of the Enable Lassa research programme is to estimate the incidences of LASV infection and LF disease in five West African countries. The core protocol described here harmonises key study components, such as eligibility criteria, case definitions, outcome measures, and laboratory tests, which will maximise the comparability of data for between-country analyses. Method We are conducting a prospective cohort study in Benin, Guinea, Liberia, Nigeria (three sites), and Sierra Leone from 2020 to 2023, with 24 months of follow-up. Each site will assess the incidence of LASV infection, LF disease, or both. When both incidences are assessed the LASV cohort (nmin = 1000 per site) will be drawn from the LF cohort (nmin = 5000 per site). During recruitment participants will complete questionnaires on household composition, socioeconomic status, demographic characteristics, and LF history, and blood samples will be collected to determine IgG LASV serostatus. LF disease cohort participants will be contacted biweekly to identify acute febrile cases, from whom blood samples will be drawn to test for active LASV infection using RT-PCR. Symptom and treatment data will be abstracted from medical records of LF cases. LF survivors will be followed up after four months to assess sequelae, specifically sensorineural hearing loss. LASV infection cohort participants will be asked for a blood sample every six months to assess LASV serostatus (IgG and IgM). Discussion Data on LASV infection and LF disease incidence in West Africa from this research programme will determine the feasibility of future Phase IIb or III clinical trials for LF vaccine candidates.

INTRODUCTION:Ebola virus disease (EVD) continues to be a significant public health problem in sub-Saharan Africa, especially in the Democratic Republic of the Congo (DRC). Large-scale vaccination during outbreaks may reduce virus transmission. We established a large population-based clinical trial of a heterologous, two-dose prophylactic vaccine during an outbreak in eastern DRC to determine vaccine effectiveness. METHODS AND ANALYSIS:This open-label, non-randomised, population-based trial enrolled eligible adults and children aged 1 year and above. Participants were offered the two-dose candidate EVD vaccine regimen VAC52150 (Ad26.ZEBOV, Modified Vaccinia Ankara (MVA)-BN-Filo), with the doses being given 56 days apart. After vaccination, serious adverse events (SAEs) were passively recorded until 1 month post dose 2. 1000 safety subset participants were telephoned at 1 month post dose 2 to collect SAEs. 500 pregnancy subset participants were contacted to collect SAEs at D7 and D21 post dose 1 and at D7, 1 month, 3 months and 6 months post dose 2, unless delivery was before these time points. The first 100 infants born to these women were given a clinical examination 3 months post delivery. Due to COVID-19 and temporary suspension of dose 2 vaccinations, at least 50 paediatric and 50 adult participants were enrolled into an immunogenicity subset to examine immune responses following a delayed second dose. Samples collected predose 2 and at 21 days post dose 2 will be tested using the Ebola viruses glycoprotein Filovirus Animal Non-Clinical Group ELISA. For qualitative research, in-depth interviews and focus group discussions were being conducted with participants or parents/care providers of paediatric participants. ETHICS AND DISSEMINATION:Approved by Comité National d'Ethique et de la Santé du Ministère de la santé de RDC, Comité d'Ethique de l'Ecole de Santé Publique de l'Université de Kinshasa, the LSHTM Ethics Committee and the MSF Ethics Review Board. Findings will be presented to stakeholders and conferences. Study data will be made available for open access. TRIAL REGISTRATION NUMBER:NCT04152486.

Rift Valley fever (RVF) is an emerging, zoonotic, arboviral hemorrhagic fever threatening livestock and humans mainly in Africa. RVF is of global concern, having expanded its geographical range over the last decades. The impact of control measures on epidemic dynamics using empirical data has not been assessed. Here, we fitted a mathematical model to seroprevalence livestock and human RVF case data from the 2018-2019 epidemic in Mayotte to estimate viral transmission among livestock, and spillover from livestock to humans through both direct contact and vector-mediated routes. Model simulations were used to assess the impact of vaccination on reducing the epidemic size. The rate of spillover by direct contact was about twice as high as vector transmission. Assuming 30% of the population were farmers, each transmission route contributed to 45% and 55% of the number of human infections, respectively. Reactive vaccination immunizing 20% of the livestock population reduced the number of human cases by 30%. Vaccinating 1 mo later required using 50% more vaccine doses for a similar reduction. Vaccinating only farmers required 10 times as more vaccine doses for a similar reduction in human cases. Finally, with 52.0% (95% credible interval [CrI] [42.9-59.4]) of livestock immune at the end of the epidemic wave, viral reemergence in the next rainy season (2019-2020) is unlikely. Coordinated human and animal health surveillance, and timely livestock vaccination appear to be key to controlling RVF in this setting. We furthermore demonstrate the value of a One Health quantitative approach to surveillance and control of zoonotic infectious diseases.

Abstract Transmission trees can be established through detailed contact histories, statistical inference, phylogenetic inference, or a combination of methods. Each method has its limitations, and the extent to which they succeed in revealing a ‘true’ transmission history remains unclear. Moreover, the net value of pathogen sequencing in transmission tree reconstruction is yet to be assessed. We explored the accuracy and sensitivity to biases of a range of methods for transmission chain inference. We studied eight transmission chains determined by contact tracing, each one having more than a third of its cases sequenced (87 samples over 199 cases in total). We compared three inference methods on the selected transmission chains: (i) phylogenetic inference: the Ebola virus (EBOV) sequences derived from patients were mapped onto a dated EBOV phylogeny tree including 398 EBOV sequences sampled in Guinea between March 2014 and October 2015; (ii) statistical inference: we used the maximum likelihood framework developed by Wallinga and Teunis to infer the most likely transmitter-recipient relationships from the onset dates; (iii) combined method: we inferred probabilistic transmission events using both pathogen sequences and collection dates with the R package Outbreaker2. The cases coming from each transmission chain were mostly clustered together in the phylogenetic tree. The few misclassified cases were most likely allocated to the wrong chains of transmission because of the timing of their symptom onsets. Probabilistic transmission tree using only onset dates broadly matched the contact tracing data, but multiple potential infectors were identified for each case. The combined method showed that an a priori knowledge of the number of independent imports had an important impact on the outcome. Although cases were allocated to the correct transmission chains, discrepancies were found in identifying direct case linkage and transmission generations within a chain. Phylogenetic, epidemiological, and combined approaches for transmission chain reconstructions globally concurred in their output. Sequence data proved useful (if not necessary) to place the sampled cases in a wider context, identify transmission clusters, and misclassified cases when epidemiological chains are inferred from date of symptom onset only, and to identify links between supposedly independent chains of transmission.

This chapter focuses primarily on the choice of the approximate Bayesian computation (ABC) tolerances that are used in the ABC kernel to penalise the dissimilarity between the simulated and observed summary statistics. It introduces lower and upper tolerances, and aims to specify non-symmetric values that offset any bias in the vanilla ABC sampler. The chapter makes the acceptance region wide enough to obtain a pre-specified degree of computational efficiency. It suggests that the quality of the ABC approximation can be sufficiently improved through the calibrations. The chapter describes the ABC accept/reject step as an equivalence test rather than the standard point null hypothesis tests. It expands on the advantages of interpreting the ABC accept/reject step as the outcome of an equivalence test. The chapter focuses on the case where the data and simulations are just real values from a normal distribution. It discusses how one-sample equivalence hypothesis tests can be used within ABC.