key: cord-0724317-43ivpt0e
authors: Flamand, C.; Enfissi, A.; Bailly, S.; ALVES SARMENTO, C.; Beillard, E.; Gaillet, M.; Michaud, C.; Servas, V.; Clement, N.; Perilhou, A.; Carage, T.; Musso, D.; Carod, J.-F.; Eustache, S.; Tourbillon, C.; Boizon, E.; James, S.; Djossou, F.; Salje, H.; Cauchemez, S.; Rousset, D.
title: Seroprevalence of anti-SARS-CoV-2 IgG at the epidemic peak in French Guiana
date: 2020-09-28
journal: nan
DOI: 10.1101/2020.09.27.20202465
sha: 6d63587f06763423e7929a627b25d4eb00ecbe76
doc_id: 724317
cord_uid: 43ivpt0e

Background SARS-CoV-2 seroprevalence studies are crucial for clarifying dynamics in affected countries and determining the route that has already been achieved towards herd immunity. While Latin America has been heavily affected by the pandemic, only a few seroprevalence studies have been conducted there. Methods A cross-sectional survey was performed between 15 July 2020 and 23 July 2020 in 4 medical biology laboratories and 5 health centers of French Guiana, representing a period shortly after the epidemic peak. Samples were screened for the presence of anti-SARS-CoV-2 IgG directed against domain S1 of the SARS-CoV-2 spike protein using the anti-SARS-CoV-2 enzyme-linked immunosorbent assay (ELISA) from Euroimmun. Results The overall seroprevalence was 15.4% [9.3%-24.4%] among 480 participants, ranging from 4.0% to 25.5% across the different municipalities. The seroprevalence did not differ according to gender (p=0.19) or age (p=0.51). Among SARS-CoV-2 positive individuals, we found that 24.6% [11.5%-45.2%] reported symptoms consistent with COVID-19. Conclusions Our findings revealed high levels of infection across the territory but a low number of resulting deaths, which can be explained by the young population structure.

The world's attention remains focused on the spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), that causes coronavirus disease 2019 , and the implementation of drastic control measures to limit its expansion. By the end of July 2020, more than 17 million confirmed cases and approximately 650,000 deaths have been reported worldwide [1] . With more than 4,500,000 cases and 190,000 deaths, Latin America has been particularly affected by the crisis [1].

A thorough evaluation of the proportion that has already been infected by SARS-CoV-2 and is likely immunized is important to estimate the level of herd immunity of the population [2] and to inform public health decision making. Data on laboratory molecular -confirmed cases do not capture the full extent of viral circulation because a majority of infected individuals have asymptomatic or mild infections and may therefore not seek care [3, 4] . In contrast, population immunity is typically estimated through cross-sectional surveys of representative samples using serological tests. Numerous serological surveys conducted in affected countries at the beginning of the COVID-19 epidemic indicate that nationwide antibody prevalence varies between 1-10%, with peaks around 10-15% in heavily affected urban areas [5] . Most of the serological studies already available have been carried out in continental Europe [6] [7] [8] [9] [10] [11] [12] and in the United States [13] [14] [15] [16] . Although Latin America has been heavily affected by the pandemic, only a few seroprevalence studies have been conducted across the continent, meaning the underlying level of infection remains largely unknown [5, 17, 18] . French Guiana, a French overseas department with 290,000 inhabitants [19] , located in Latin America in the Amazonian forest complex experienced a large SARS-CoV-2 epidemic wave. A territorywide lockdown was set up from March 17 th 2020 concomitantly with the rest of French territories, at a time when five imported cases and one secondary case were being confirmed on the territory [20] . The lockdown resulted in limited viral transmission until it was ended on May 11th 2020. In the middle of June there was a rapid intensification of viral circulation over a large part of the territory with 917 confirmed cases of COVID-19 detected from March 4 th 2020 to June 11 th 2020 [21]. This was followed by the implementation of strict mitigation measures such as curfews and local lockdowns in the course of June and July. The epidemic peaked at the beginning of July with 4,440 cumulative confirmed cases [22] 

French Guiana is composed of two main inhabited geographical regions: a central, urbanized area including a coastal strip along the Atlantic Ocean, where 90% of the population lives, and a more remote area along the Surinamese and Brazilian borders (Figure 1 ). This territory has the highest crude birth rate in the Americas (25.6 per 1,000 people) and 32% of the population is under the age of 15 [19] . Although living conditions are substantially more precarious than those of mainland France, French and European status gives French Guiana higher healthcare and diagnosis capacities than most South American countries. The healthcare system includes 8 medical biology laboratories and 3 hospital centers located in the . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . https://doi.org/10.1101/2020.09.27.20202465 doi: medRxiv preprint main municipalities of the coastal area, as well as 17 healthcare centers located in more isolated areas.

A cross-sectional survey was performed between 15 July 2020 and 23 July 2020 in 4 medical biology laboratories located in the coastal urban area of French Guiana and in 5 health centers located in isolated areas along the Surinamese and Brazilian borders ( Figure 1 ).

All individuals visiting the recruitment centers during the study period were invited to participate to the study, expected those admitted for SARS-CoV-2 RT-PCR testing.

Publicity and information about the survey was provided at the reception desk of recruitment centers. Fieldworker investigators were trained to explain the project, collect informed consent and carry out the interviews. All individuals visiting the recruitment centers were invited to take part in the study during a preliminary face-to-face interview. For participants under 18 years-old, age-appropriate information was given and the consent of legally responsible person was collected. A specific educational-style comic poster was designed to explain, in an understandable way, the nature and objectives of the survey and inform them about the voluntary nature of the participation of the study.

Demographic data, including age, gender, residential region and occupation of each participant, were collected through a standardized questionnaire. Participants were asked to report the occurrence of a presumptive COVID-19 infection, to list the infection associated symptoms and to specify if they had consulted a doctor or obtained a biological confirmation of their infection. Thereafter, a venous blood sample of 3.5mL was collected from each participant, in accordance with current biosafety standards.

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . https://doi.org/10.1101/2020.09.27.20202465 doi: medRxiv preprint

The study was recorded on Clinicaltrials.gov (ID: NCT04490850) and approved by the "CPP EST-III" Ethical Research Committee (No.CPP 20.07.04-8827; N°ID-RCB 2020-A01826-33). Personal data processing for this study comply with the requirements of the "reference methodology MR-001" established by the French Data Protection Authority (CNIL) regarding data processing in health research.

Blood samples were collected into 3.5 ml gold BD Vacutainer SST II advance tubes with gel for serum separation (Becton-Dickinson, USA). Immediately after puncture, samples were stored at 4°C-8°C until centrifugation within 12 hours. Sera were then frozen and stored at -20°C until used at the National Reference Center for respiratory viruses in Institut Pasteur in French Guiana.

Collected samples were screened for the presence of anti-SARS-CoV-2 IgG directed against domain S1 of the SARS-CoV-2 spike protein using the anti-SARS-CoV-2 enzyme-linked immunosorbent assay (ELISA) from Euroimmun (Lübeck, Germany). Semiquantitative results were calculated as the ratio of the extinction of samples over the extinction of a calibrator. According to the distributor, the specificity of the test was 99.6%. We internally validated the specificity of the serological assay with serum samples collected from 186 individuals of a cross-sectional serosurvey [25, 26] in healthy individuals taken prior to the SARS-CoV-2 outbreak (June to October 2017). Of these, 40.8% were male and the median age was 33.3 years. In this validation subset, the serological test showed a specificity of 97.9% consistent of a recent assessment of this assay [27] . According to the distributor, the . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . sensitivity of the test is 75.0% if the test is performed 10 to 20 days after infection and 93.8%

if it is performed more than 20 days after infection.

In order to obtain population representative estimates of overall seropositivity in the territory we weight each sample by the population size within each municipality, age and sex group. We employ the following notation to describe the study design (Table 1) 

We enrolled 480 participants between July 15 and July 23, 2020, in 16 municipalities (Table   1 ). The mean age of participants was 38.3 ranging from 0.2 to 87 years old. Comparison of the socio-demographic characteristics of the study sample to the census data demonstrated an over-representation of women (68.1% vs 50% in the general population of French Guiana) and adults over 25 years (79% vs 53% in French Guiana). These differences were corrected in the analyses of seroprevalence and risk factors by allocating a post-stratification weight to each participant.

Between July 15 and July 23 2020, the crude proportion seropositive was 13 Since the study was implemented two weeks after the epidemic peak and the sensitivity of the test is limited in the three weeks that follow infection, this likely represents a lower bound for the proportion of individuals infected by the time the epidemic peaked.

The seroprevalence did not differ according to gender (p=0.19) or age (p=0.51) ( Table 2 ).

Serological results in the different geographical areas (Figure 2) showed that SARS-CoV-2 circulated in most of French Guiana (Table 1) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. Table 3 ).

We report the first population-based serosurvey for the detection of SARS-CoV-2 antibodies in French Guiana. We found that 15.4% [9.3%-24.4%] of the population was seropositive two weeks after the peak of the epidemic. Assuming a two or three-week delay for seroconversion, our estimation reflects the level of infection of the population at the end of June or beginning of July, which roughly corresponds to the epidemic peak. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted September 28, 2020. . https://doi.org/10.1101/2020.09.27.20202465 doi: medRxiv preprint confirmed cases reported by public health surveillance system by the first week of July [22] .

In Brazil, one of the bordering countries in French Guiana, seroprevalence estimates varied markedly across the country's cities and regions, from below 1.0% in most cities in the south and center-west regions to up to 25.0% in the city of Breves in the Amazon (North) region [17] . Nevertheless, overall seroprevalence was estimated at 1.4% (95% CI 1.3-1.6 ). In contrast, our findings highlighted high but also relatively homogeneous levels of infection in most municipalities, ranging from 10% to 20%.

The case fatality rate of COVID-19 was low during the outbreak as there were 65 COVID-19 related deaths from the beginning of the outbreak up to September 17 [23] across the territory while about 45,000 people have been infected at the beginning of July. These was probably due in part to the young age of the population of French Guiana.

Younger populations are likely to have more social ties than older populations, and therefore physical distancing may be more difficult to implement than in countries with ageing populations. Furthermore, since young people are less exposed to disease severity, they may be less likely to adopt physical distancing measures when they are infected and potentially contagious in a context of high level of transmission.

Our study has several limitations inherent in the study design. Our approach made it possible to obtain rapid estimates of the impact of the epidemic. However, convenience sampling may result in a lack of population representation if part of the general population has lower access to the laboratories and health centers participating in the study. In our study, we observed a significant under-representation of men and children under 15 years of age, and therefore performed a post-stratification adjustment. However, this may have led to large confidence intervals for some of parameter estimates. In addition, sample size calculation was determined to obtain a sufficient point estimate of territory-wide prevalence estimates but not to study risk factors of infection. A few municipalities with no laboratory or health centers were not . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. .

represented. However, the municipalities represented by the laboratories and health centers involved in the study are home to 80% of the population, so that our estimates are likely a good reflection of the situation across the territory. We may underestimate infection levels if precarious populations are at higher risk of infection and have limited access to health facilities.

Although specificity of the test used in our study is satisfactory according to our internal validation, we cannot rule out the possibility that a proportion of people infected at the beginning of the epidemic i.e. more than 3 months before our survey may have become seronegative [30] . In the other hand, it is possible that infected people did not develop specific SARS-CoV-2 antibodies or that these antibodies were not detected by our assay. Since the study was performed shortly after the peak of the epidemic, a proportion of individuals In conclusion, we found that at least 15% of the population in French Guyana was infected by SARS-CoV-2 by the time the epidemic peaked in July. This corresponds to an elevated infection burden given the relatively limited mortality, which can be explained by French Guyana's young population structure. epidemio-regional-hebdomadaire-special-covid-19

Santé Publique France. COVID-19 : point épidémiologique en Guyane du 17 septembre2020 [Internet] . [cited 2020 Sep 14] . Available from:

https://www.guyane.ars.sante.fr/point-epidemio-regional-hebdomadaire-special-covid-19

24. Ludvigsson JF. Systematic review of COVID-19 in children shows milder cases and a . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . https://doi.org/10.1101/2020.09.27.20202465 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . https://doi.org/10.1101/2020.09.27.20202465 doi: medRxiv preprint 

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted September 28, 2020. . https://doi.org/10.1101/2020.09.27.20202465 doi: medRxiv preprint

Coronavirus Disease (COVID-19) Situation Reports -193

COVID-19 herd immunity: where are we?

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention

SARS-CoV-2 seroprevalence in COVID-19 hotspots. The Lancet

Estimating the seroprevalence of SARS

CoV-2 infections: systematic review. medRxiv

Seroprevalence of anti-SARS-CoV-2 IgG antibodies in Geneva, Switzerland (SEROCoV-POP): a population-based study

Prevalence of SARS-CoV-2 in the Luxembourgish population: the CON-VINCE study. medRxiv

SARS-CoV-2 antibody seroprevalence in industry workers in Split-Dalmatia and Sibenik-Knin County, Croatia. medRxiv

Epidemiological study to detect active SARS

CoV-2 infections and seropositive persons in a selected cohort of employees in the Frankfurt am Main metropolitan area. medRxiv

COVID-19 experience: first Italian survey on healthcare staff members from a

Mother-Child Research hospital using combined molecular and rapid immunoassays test | medRxiv

SARS-CoV-2 seroprevalence and neutralizing activity in donor and patient blood from the San Francisco Bay Area. medRxiv

Seroprevalence of SARS-CoV-2-Specific

Remarkable variability in SARS-CoV-2 antibodies across Brazilian regions: nationwide serological household survey in 27 states. medRxiv

Repeated population-based surveys of antibodies against SARS-CoV-2 in Southern Brazil. medRxiv

Communiqué de presse : Coronavirus Guyane : 1 cas secondaire confirmé à Saint-Laurent du Maroni

COVID-19 : point épidémiologique en Guyane du 11 juin

COVID-19 : point épidémiologique en Guyane du 2 Juillet