Journal of Public Health in Africa 2017; volume 8:732

Seroprevalence of Ebola virus December 2015. Retrospective serologic infection in Bombali District, data3 suggest that EBOV might have been in Correspondence: Carla Montesano, Tor circulation in Sierra Leone since 2006. Vergata University of Rome, via della Ricerca Sierra Leone The role of antibody responses in viral Scientifica 1, 00133 Rome, Italy. clearance and protection against Ebola Tel.: +39.06.72594237 - Fax: +39.06.72494224. Nadege Goumkwa Mafopa,1-3 Virus (EBOV) in humans is not fully under- E-mail: [email protected] Gianluca Russo,3,4 stood. Among EVD survivors, antibodies Key words: Ebola virus infection; Sierra 1-3,5 appear as early as day-5 after symptom Raoul Emeric Guetiya Wadoum, Leone; seroprevalence. Emmanuel Iwerima,6 onset, peak 2 weeks after recovery, then 4-7 Vincent Batwala,6,7 Marta Giovanetti,3,5 decline slowly over several years. In Acknowledgments: the data presented here Antonella Minutolo,3,5 Patrick Turay,3 patients with fatal outcome, antibody titers were obtained with the financial support of the are low or absent.8 The presence of Thomas B. Turay,8 Brima Kargbo,9 Italian Episcopal Conference (Camillian Task detectable anti-EBOV antibodies in asymp- Massimo Amicosante,5 Force), the Italian Cooperation (Dokita ONG) tomatic individuals suggests exposure and a and the Italian Geographical Society (Centro Maurizio Mattei,5 Carla Montesano2,3,5 putative role of antibody response in the Relazioni con l’Africa). Authors wish to thank Ebola Study Group control of EVD and provides information C.D. Pauza (Institute of Human Virology, 1University of Dschang, Dschang, on EBOV seroprevalence in at risk popula- University of Maryland School of Medicine, Cameroon; 2University of Makeni, tions.9 Monitoring anti-EBOV antibodies in Baltimore, USA) for helpful discussions and language revision. Makeni, Sierra Leone; 3Holy Spirit EVD-community contacts (CCs) of index cases is crucial for assessing immune pro- Hospital, Makeni, Sierra Leone; Contributions: NGM, GR and REGW con- 4 La Sapienza University of Rome, Rome, tection against EBOV. Due to the large tributed equally to this article as first authors. Italy; 5University of Rome Tor Vergata, number of infected HCWs, evaluation of CM and MM contributed equally to this article Rome, Italy; 6African Union Support to anti-Ebola antibodies in this group is a use- as last authors. CM, GR had the original idea Ebola Outbreak in West Africa; ful tool for measuring the risk of occupa- of the study; CM, GR, VB, MM designed the tional exposure to EBOV. epidemiological survey; CM, GR, NGM 7Directorate of Research and Graduate only The aim of the study was to assess the wrote the manuscript; NGM, REGW, EI, MG, Training, Mbarara University of Science PT have participated in the fieldwork for col- 8 extent of asymptomatic or mild cases of and Technology, Uganda; Partner EBOV infection among the CCs of labora- lecting samples in the villages, treatment of Relief Development, Freetown, Sierra tory confirmed EVD cases (survivors or samples and ELISA test; CM, MA, AM partic- 9 Leone; Ministry of Health and deceased) and in HCWs from health facili-useipated in the assessment of assay, in samples classification and performed statistical analy- Sanitation, Freetown, Sierra Leone ties not involved in the management of sis; TBT, BK have organized the samples col- EVD cases. lection, supervising logistic.

We dedicate this work to the memory of Abstract Massimo Amicosante, colleague and friend, to A serosurvey of anti-Ebola Zaire virus Materials and Methods honour his enthusiasm and his great scientific nucleoprotein IgG prevalence was carried skill. out among Ebola virus disease survivors Study populations and their Community Contacts in Bombali The study was conducted between Received for publication: 30 June 2017. District, Sierra Leone. Our data suggest that February and March 2015 in the rural area Accepted for publication: 1 July 2017. the specie of Ebola virus (Zaire) responsible of Sanda Loko Chiefdom, in the villages of commercial This work is licensed under a Creative of the 2013-2016 epidemic in West Africa Kamalo, Maron and Makasa (Figure 1) in Commons Attribution NonCommercial 4.0 may cause mild or asymptomatic infection the context of Ebola survivor Mobile Health License (CC BY-NC 4.0). in a proportion of cases, possibly due to an Clinic program aimed to provide integrated efficient immune response. primary healthcare services to address the ©Copyright N.G. Mafopa et al., 2017 Nonmedical and psychosocial needs of Ebola Licensee PAGEPress, Italy survivors living in areas with low medical Journal of Public Health in Africa 2017; 8:732 doi:10.4081/jphia.2017.732 coverage.10 Introduction A total of 6 EVD-deaths and 10 EVD- Ebola Virus Disease (EVD) spreads in survivors were reported (Table 1). EVD communities through person-to-person cases were reported from Kamalo (4 deaths Makeni municipality (headquarter of transmission, with infection resulting from and 10 survivors) and Maron (2 deaths) vil- Bombali District) that were not involved direct contact with blood, secretions, other lages, whereas no cases were reported from in the management of EVD cases (Table 1). body fluids or organs, and indirect contact Makasa village (Table 2). EVD survivors The definition of HCWs included physi- with contaminated environmental materials were defined as individuals who showed cians, nurses, laboratory technicians, or infected animals. In Sierra Leone, a total EVD clinical signs or symptoms with administrative personnell and cleaners of 8704 confirmed cases and 3589 deaths EBOV infection confirmed by RT-PCR (Table 1). were reported by the end of December assay. Considering the organization of the The Ethical Review Board of the 2015,1 of which 307 (221 deaths) were local rural society, CCs were individuals Ministry of Health and Sanitation of Sierra among health care workers (HCWs).2 In from communities placed under quarantine. Leone approved this study, and participants Bombali District, located in the northern In addition, seventy-nine HCWs were gave their approval by signing or finger- region of the country, a total of 1050 con- included in this study from Holy Spirit printing a written consent form. firmed EBV cases have been cumulatively Hospital (n=59) and Loreto Clinic (n=20), Demographic data and information about reported since May 2013 to the end of two religious health care facilities in clinical history were collected using pre-

[page 132] [Journal of Public Health in Africa 2017; 8:732] Article tested questionnaires before obtaining 4 ml sons respectively. cases (P<0.05, Table 2) suggesting that of peripheral blood by venipuncture. Although the knowledge of Ebola EBOV circulated in those populations with- dynamics, pathogenesis and clinical course out causing disease. Anti Ebola virus antibody level is improving, some aspects remain unclear, In this paper we report a high sero- Anti-Zaire Ebola Nucleoprotein including the role for naturally-acquired prevalence in asymptomatic CCs of Sanda (ZEBOV-NP) IgG antibodies were quanti- humoral immunity. Loko Chiefdom although we have recently fied using Enzyme-Linked Immunosorbent The influence of humoral immunity on shown that only 2.6% (10 of 388) of asymp- 8 assay (ELISA) by a commercial kit EVD outcome was observed in Gabon and tomatic members of Ebola-affected house- 7 (AE320620-1, Alpha Diagnostic confirmed by a serosurvey in Uganda. A holds had evidence of Ebola virus infection 9 11 International [ADI], Texas, USA), accord- larger EBOV serosurvey in Gabon showed in Sierra Leone. The extend of asymp- ing to the manufacturer’s instructions. significant seroprevalence across the coun- tomatic EBOV infection is still unclear and Positive control and calibrators provided by try even in areas considered Ebola-free, and several studies reported a wide variability 9,12-14 the kit were used in each test run. Optimal raised important questions about EBOV of seroprevalence (from 1% to 46%), sample dilution was previously determined spread, virulence and the existence of natu- among the household and CCs. This high at 1:500 by testing 68 positive control plas- ral protective immunization. variability can be ascribed to the different ma samples from EVD-survivors and 10 The overall seroprevalence rate among antigens and assays used and also the defi- negative control plasma samples from asymptomatic CCs in the area of Sanda nition of contact varied in the studies. expatriates not exposed to Ebola infection. Loko Chiefdom was 11.4% with significant The seroprevalence rate among all Threshold index to discriminate posi- differences observed among the three vil- HCWs was 3.8% even if the health facilities tive and negative antibody reactivity to lages. Anti-EBOV-positive subjects were where they worked were not included in the ZEBOV NP was calculated following man- found in all villages with or without con- network for managing EVD cases. We can- ufacturer’s instructions. firmed EVD cases. Interestingly, the highest not rule out that they may have been seroprevalence rates was observed in vil- exposed to EVD outside the work place. Statistical analysis lages with the lowest incidence of EVD The observed moderate grade of EBOV Data were collected and entered in a only customized template in EpiData software version 3.1. The prevalence was calculated Table 1. Prevalence of ZEBOV-seropositivity in EVD survivors, community contacts and health care workers. by dividing the number of seropositive cases by the number of examined individu- N Testeduse (n) Seropositives (n) Seroprevalence (%) als. Comparison among frequency of Sanda Loko EVD-survivors 10* 4 4 100 seropositive CCs from different villages Sanda Loko EVD deaths 6 - - - was performed by one-tail Fisher’s exact test assuming the hypothesis of seropreva- Sanda Loko CCs NA 105 12 11.4 lence association. Total HCWs 79 79 3 3.8 Holy Spirit Hospital 59 59 3 5.8 Loreto Clinic 20 20 0 0.0 *4 survivors were enrolled and tested in this study. NA, not available. Results and Discussion Among Sanda Loko EVD survivors, all Table 2. EVD deaths and survivors, community contacts and seroprevalence in Sanda those enrolled from Kamalo were ZEBOV- Loko Chiefdom villages. NP-IgG positive (n=4/4). Anti-ZEBOV-NP- commercialVillage Deaths Survivors CCs Anti-ZEBOV Seroprevalence IgG were detected in 12 out of the 105 test- (n) (n) (n*) positive CCs (n) (%) among CCs ed CCs (11.4% seroprevalence rate) (Table 1). As described in Table 2, the seropreva- Kamalo 4 10° 51 3 5.9# lence for ZEBOV-NP-IgG in CCs from dif- Maron 2 0 35 7 20.0 ferent villages was 5.9% (n=3/51)Non in Makasa 0 0 19 2 10.5 Kamalo, 20.0% (n=7/35) in Maron, and *CC (n) refers to the number of Community Contacts included in this study; °four survivors were enrolled and tested in this study; #P<0.05. 10.5% (n=2/19) in Makasa. Among HCWs (Table 1), no ZEBOV- seropositive cases were found in the Loreto Clinic group (n=0/20), whereas 3 seroposi- tive cases (1 nurse and 2 cleaners) out of 59 (5.8%) were found at the Holy Spirit Hospital. Overall, the seroprevalence among HCWs was 3.8% (n=3/79). ZEBOV-seropositive CCs did not report fever or other symptoms/signs suggestive of EVD during the previous 8 months. Three of them reported mild, non-specific symp- toms, including headache (n=1), joint pain (n=1), or both (n=1). Risk behaviours, such as hunting activity in the forest, or partici- Figure 1. Maps of Sierra Leone, Bombali District and Sanda Loko Chiefdom. Kamalo pating in a burial service during the past 8 and Makasa villages are reported in local language as Kamalu and Makassa respectively. months, were reported by one and two per-

[Journal of Public Health in Africa 2017; 8:732] [page 133] Article transmission in HCWs needs further analy- Cameroon), Marina Potestà (University of 2013;208: 299-309. sis of larger groups, along with assessing Rome “Tor Vergata”, Rome, Italy), 8. Baize S, Leroy EM, Georges-Courbot the adherence to use of personal protection Massimo Ciccozzi (National Institute of MC, et al. Defective humoral responses (gowns, gloves, masks). This data is consis- Health – ISS, Rome, Italy), Giovanni Rezza and extensive intravascular apoptosis tent with previous serosurvey performed in (National Institute of Health – ISS, Rome, are associated with fatal outcome in 9 Gabon and led to hypothesize that the host- Italy), Jeffrey R. Dorfman (International Ebola virus-infected patients. Nat Med EBOV interaction might determine the Center for Genetic Engineering and 1999;5:423-26. development of specific antibodies capable Biotechnology, Cape Town, South Africa; 9. Becquart P, Wauquier N, Mahlakõiv T, to control the infection, leading to mild or University of Cape Town, Cape Town, South et al. High prevalence of both humoral asymptomatic EVD. Africa). and cellular immunity to Zaire Our findings that anti-ZEBOV-IgG was ebolavirus among rural populations in detected in both EVD-survivors and CCs, Gabon. PLoS One 2010;5:e9126. although at different prevalence rates, sug- 10. Guetiya Wadoum RE, Samin A., gests that the immunity, previously reported References Goumkwa Mafopa N, et al. Mobile as persistent among survivors,7 may be 1. Ministry of Health and Sanitation, Health Clinic for Medical Management important for protection that allowed sub- Sierra Leone. Ebola virus disease - clinical disease among CCs. Situation report; 24 December 2015. of Clinical Sequelae Experienced by Available from: http://nerc.sl/ sites/ Survivors of the 2013-2016 Ebola Virus default/files/Ebola%20Situation%20Re Disease Outbreak in Sierra Leone, West port_Vol%20575.pdf Africa. Eur J Clin Microbiol Infect Dis Conclusions 2. World Health Organization (WHO). 2017; 36:2193-200. Our study provides significant new Ebola Situation Report; 30 December 11. Glynn JR, Bower H, Johnson S, et al. insight into population exposure to EBOV 2015. Available from: http://apps.who. Asymptomatic infection and unrecog- and outcomes of virus infection. This caus- int/ebola/current-situation/ebola-situa- nised Ebola virus disease in Ebola- es us to question the assumption that EBOV tion-report-30-december-2015 affectedonly households in Sierra Leone: a is highly fatal. Instead, EBOV, similar to 3. Schoepp RJ, Rossi CA, Khan SH, et al. cross-sectional study using a new non- Lassa Fever virus15,16 and many other viral Undiagnosed acute viral febrile illness- invasive assay for antibodies to Ebola pathogens, may have high fatality as a pro- es, Sierra Leone. Emerg Infect Dis virus. Lancet Infect Dis 2017;17:645- portion of clinical cases but relatively high 2014;20:1176-82. use 53. seroprevalence among persons without evi- 4. Rowe AK, Bertolli J, Khan AS, et al. 12. Clark DV, Kibuuka H, Millard M, et al. dent disease arguing the fatality rate is actu- Clinical, virologic, and immunologic Long-term sequelae after Ebola virus ally lower than has been predicted. These follow-up of convalescent Ebola hem- disease in Bundibugyo, Uganda: a ret- findings raise important questions about orrhagic fever patients and their house- rospective cohort study. Lancet Infect EBOV pathogenicity and argue that some hold contacts, Kikwit, Democratic Dis 2015;15:905-12. individuals have achieved long lasting pro- Republic of the Congo. J Infect Dis 13. Richardson ET, Kelly JD, Barrie MB, et tection following attenuated infection7 that 1999;179:S28-35. al. Minimally symptomatic infection in should be taken in account for vaccine pro- 5. Ksiazek TG, Rollin PE, Williams AJ, et an Ebola 'hotspot': a cross-sectional grams. al. Clinical virology of Ebola hemor- serosurvey. PLoS Negl Trop Dis 2016; There is a strong rationale for expand- rhagic fever (EHF): virus, virus antigen, 10:e0005087. ing the serological screening of exposed and IgG and IgM antibody findings 14. Leroy EM, Baize S, Volchkov VE, et al. populations and HCWs in countries affect- commercialamong EHF patients in Kikwit, ed by EVD to understand their risk of infec- Democratic Republic of the Congo, Human asymptomatic Ebola infection tion and to guide the implementation of 1995. J Infect Dis 1999;179:S177-87. and strong inflammatory response. quarantine when needed, and vaccine pro- 6. Wauquier N, Becquart P, Gasquet C, Lancet 2000;355:2210-15. grams when available. Leroy EM. Immunoglobulin G in Ebola 15. Fhogartaigh CN, Aarons E. Viral haem- Non outbreak survivors, Gabon [Letter]. orrhagic fever. Clin Med 2015;15:61-6. Ebola Study Group Emerg Infect Dis 2009. 16. Sanchez A, Geisbert TW, Feldmann H. Joseph Turay (University of Makeni, 7. Sobarzo A, Groseth A, Dolnik O, et al. Filoviridae: Marburg and Ebola viruses. Makeni, Sierra Leone); Mahmoud B. Profile and persistence of the virus-spe- In: Knipe DM, Howley PM, eds. Fields Kargbo (Partner Relief Development, cific neutralizing humoral immune virology. Philadelphia: Lippincott Freetown, Sierra Leone), Martin Sanou response in human survivors of Sudan Williams and Williams; 2007. pp 1409- Sobze (University of Dschang, Dschang, ebolavirus (Gulu). J Infect Dis 1448.

[page 134] [Journal of Public Health in Africa 2017; 8:732]