Artigos

Introduction
Chikungunya (CHIK) is an arthropod-borne viral disease transmitted by infected Aedes mosquitoes and caused by the chikungunya virus (CHIKV), which belongs to the genus Alphavirus of the family Togaviridae. It is endemic in tropical and subtropical areas of the African, Indian and Asian continents1. Recently, CHIKV has expanded its area of distribution to Europe and the Americas causing large outbreaks in naive populations, and imposing a serious health burden2,3. Clinical manifestations of CHIKV infections are characterized mainly by sudden onset of fever, myalgia, headache, and severe or debilitating arthritis/arthralgia that can last after the acute phase in some patients from months to years4,5. Although mortality rate for CHIK alone is considered low, there is growing evidence that pre-existing chronic diseases may suffer decompensation during infection and, therefore, certain group of patients are at increased risk for severe disease due to exacerbation of pre-existing chronic-based conditions6,7. However, the mechanisms by which this occurs are unclear and need further investigation.
In 2014, Brazil confirmed the first report of local transmission of CHIKV in the cities of Oiapoque (Amapá State) and Feira de Santana (Bahia State)8. Since its spread in Brazil, the majority of CHIK confirmed cases have been reported in the Northeast region9. It is noteworthy that the city of Fortaleza (Ceará State) registered a total of 101,170 autochthonous cases from December 2015 to December 2022, 2017 being the epidemic year with the highest number of notifications, accounting for 61.1% (61,828/101,170) of all cases10. In 2022, the Brazilian Ministry of Health reported 172,082 suspected cases through epidemiological week 50, of which over 11.9% (20,555/172,082) were from the city of Fortaleza, an incident rate of 760.3 cases/100,000 inhabitants11.
In general, the dynamic of an infectious disease within a population is impacted, along with environmental and sociodemographic factors, by two hidden variables (susceptibility and immunity)12. The balance between these variables could affect the outcome of pathogen transmission and thus may contribute to shape the immunity landscape within the population as a consequence of host resistance or susceptibility at individual level, which can be assessed by serosurveys13,14. In fact, serology provides direct evidence of population immunity using detection of pathogen-specific immunoglobulin IgM and IgG antibodies. Several studies both in animal models and humans indicate that humoral immunity plays a key role against CHIKV infection15-18. However, it is important assessing not only antibody profiles in a given population from an endemic area, but also to understand how antibody levels wane over time, that could provide valuable information for surveillance programs. For instance, which immunity factors may contribute to a shift from subclinical infections to clinical manifestations of disease remain poorly understood19. Also in this context, how simultaneous circulation of other arbovirus, such as dengue (DENV) and zika virus (ZIKV), affect the dynamics of CHIKV epidemics is not known.
In this study, we examined the presence of total anti-CHIKV antibodies IgM and IgG for seroprevalence estimation, as well as specifically evaluated the kinetics of IgG levels in paired sera collected between February 2018 and August 2019 following the 2017 outbreak in the city of Fortaleza, Ceará, Northeast Region of Brazil. Participants were women aged 15 to 39 and were stratified by age, prior history of CHIK diagnosis and the collection intervals between baseline and first follow-up for further analysis.

Materials and Methods
Study design and sample selection
We analyzed samples of a subsample of a cohort study entitled ‘Zika in Fortaleza (ZIF): response of a cohort of women aged 15-39 years’20. On-demand blood sampling was conducted from February 2018 to August 2019, which included both baseline and first follow-up samples, and started nearly 1-2.5 years after the 2017 major CHIK outbreak in the city of Fortaleza in areas known to have had the highest numbers of CHIKV confirmed cases20. The participants were derived from four independent Primary Health Care Units (PHCU) located in the neighborhoods of Barra do Ceará, Rodolfo Teófilo and Conjunto Esperança districts. A convenience sampling strategy was adopted in the original cohort and participants of this study were randomly drawn among women seeking health care. The primary reasons of the study participants for seeking health care were not addressed directly but it could be a routine or sickness appointment for the participant (e.g. vaccine, dental care, prenatal appointment) or someone she was accompanying to the PHCU. As we aimed to determine the longitudinal seroprevalence of specific antibodies against CHIKV, along with the kinetics of IgG levels, and since this study was a subsample of a cohort, the target size of the participants was defined based on the followed inclusion criteria: (1) consent to participate in the study; (2) two paired blood samples collected, namely baseline and first follow-up; and (3) valid and conclusive serological test - negative or positive - for CHIKV. The collection intervals between baseline and first follow-up differed among participants, ranging from 2 months to up to 15 months. In order to evaluate the kinetics of anti-CHIKV IgG levels for those who were IgG positive both at baseline and first follow-up, we further stratified the collection intervals into two periods: short-term period was defined as from the day of enrollment (day 0; D0) to the day 60-210 (n=300; D60-210; or 2-7 months post-baseline sampling) and mid-term period as from the day 0 to the day 240-450 (n=259; D240-450; 8-15 months post-baseline sampling). Noteworthy, participants of the short-term period were not the same participants of the mid-term period and there is no gap between the two periods. In fact, samples included in D240 mid-term starting point encompassed those collected from day 211 to day 240. Of the 1,466 participants initially enrolled with conclusive serological results at baseline, 293 did not return for a second blood collection and 423, even though returning to the follow-up, did not have a second blood collection due to unforeseen circumstances. Consequently, only 750 attended the first follow-up for antibody testing, 191 displayed inconclusive results and thus 559 subjects met the inclusion criteria.

Data collection and blood sampling
At the day of enrollment, using the software Survey Monkey (SurveyMonkey Inc., San Mateo, California, USA), a presential interview was conducted using a pre-defined questionnaire to obtain information regarding sociodemographic and general health status of participants. Baseline blood samples were also collected. In particular, from the same venipuncture, two vacutainer vials containing 4 mL of peripheral venous blood were drawn from each participant (n=559). All study participants were invited a second time to provide a first follow-up blood sample for anti-CHIKV antibody testing in addition to the questionnaire. All samples were processed on the day of collection. Serum was aliquoted following centrifugation, and a backup sample was stored at -70oC, while the samples used immediately after for the enzyme-linked immunosorbent assay (ELISA) were kept at -20oC.

Laboratory assays
Serological tests were performed using an ELISA-based semi-quantitative assay from Euroimmun (Anti-CHIKV IgM Code: EI 293a-9601M; Anti-CHIKV IgG Code: EI 293a-9601G; Germany) for the detection of human CHIKV-specific IgM/IgG antibodies. Following manufacturer's instructions, optical densities (ODs) were evaluated in a spectrophotometer reader at 450 nm with 620-650 nm as reference wavelength. Samples were considered positive when ratio values (extinction of the control or participant sample / extinction of calibrator) ? 1.1 were returned; borderline with ratio values ranging from ? 0.8 and < 1.1, and negative with ratio values < 0.8. Non-conclusive (borderline) results were repeated. Samples that remained inconclusive were further excluded from the analyses. For the purpose of this study, we also used ratio values to temporally evaluate anti-CHIKV IgG levels in a post-epidemic scenario.

Study definitions
All participants were asked both at baseline and at first follow-up questionnaires whether they have had symptoms compatible with arbovirus infection. None reported as having symptoms on the day of blood collection and during the interval between baseline and first follow-up. Therefore, a case of CHIKV infection was defined by ELISA in those who had a positive result for IgM and/or IgG. In addition, participants who declared receiving a CHIK diagnosis by a physician prior to enrollment in the study were classified as having had symptomatic infection. Those who did not self-report illness or did not receive a clinical diagnosis of CHIK, but had positive results for IgM, IgG, or both in our study, we considered to have had asymptomatic infection.

Statistical analysis
The data were collected and entered into STATA v. 16.0 (StataCorp 2017, Stata Statistical Software: Release 15; StataCorp LLC, College Station, TX, USA) for descriptive statistics. Demographic and baseline characteristics were reported as number of participants (n) and percentage (%) for categorical variables and median with interquartile-range (IQR) for continuous variables. The CHIKV seroprevalence is expressed as percentages with 95% confidence intervals. In addition, we used non-parametric tests for inferential statistical analysis in a convenience sample, randomly drawn, of women aged 15-39 years' seeking health care from four independent areas with the highest numbers of CHIKV confirmed cases, regardless of the primary reasons for coming to the PHCU. Furthermore, we presumed that the likelihood of exposure to mosquito infected bites was similar among participants from areas with the highest numbers of CHIKV confirmed cases. Plasma paired samples were therefore compared using a two-tailed Wilcoxon matched-pairs signed-rank test for comparisons of anti-CHIKV IgG levels according to the age group, or prior clinical diagnosis of CHIK, and interval of collection, respectively. The Mann-Whitney U test was used for single comparisons between two independent groups as follows: median (IQR) IgG levels of 15-24 years age group versus 25-39 years age group, and Prior CHIK diagnosis versus No Prior CHIK diagnosis according to the interval of collection. A two-sided p-value less than 0.05 was considered to be statistically significant. Inferential statistics were analyzed with GraphPad Prism 8.4 (GraphPad Software, San Diego, CA, USA).

Results
We assessed 1,466 participants for eligibility, of whom 559 met the inclusion criteria of this study and were enrolled. Demographic and baseline characteristics of these women are shown in Table 1. Participants included 18.4% (n=103) pregnant women, 14.5% (n=79) with self-reported comorbidities, and 10.1% (n=56) with self-reported medical history of CHIKV infection prior to enrollment. The median (IQR) age of the participants was 25 (20-31) years, and 46.5% (n=260) were aged 15 to 24 years, while 53.5% (n=299) were aged 25 to 39 years.
Blood samples were collected from all 559 participants and serum were further separated to perform ELISA assays for detection of anti-CHIKV IgM/IgG antibodies. At baseline or day 0 (D0), the overall seroprevalence of CHIKV antibodies at baseline (IgM+ and/or IgG+) was 34% (190/559; 95% CI, 30.2 - 38.0) (Table 2). Furthermore, we found that seroprevalence ranged from 35.4% (92/260) to 32.8% (98/299) in the age groups 15-24 years and 25-39 years, respectively (Table 2). Among 559 paired samples, 27.9% (156/559; 95% CI, 24.3 - 31.8) were seropositive for IgG (regardless of positivity for IgM) at D0 and 28.3% (158/559; 95% CI, 24.7 - 32.1) were seropositive for IgG at the first follow-up (Table 3; Figure 1A). Moreover, we observed seroconversion (IgG- to IgG+) in 1.5% (6/403; 95% CI, 1.0 - 6.6) of paired samples and seroreversion (IgG+ to IgG-) in 2.6% (4/156; 95% CI, 0.7 - 3.3) paired samples (Table 3).
In this study, we also sought to evaluate changes in IgG levels among individuals who were seropositive for IgG at baseline and follow-up. Interestingly, although the overall percentage of IgG seropositivity did not differ significantly between D0 and the first follow-up (27.9% versus 28.3%, Figure 1A), there was a trending decrease of IgG levels over time (p < 0.0001, Figure 1B). In contrast, only 20.4% (31/152, Figure 1B) of participants displayed increased IgG levels compared to baseline, considering those who were initially IgG+ at D0. In order to investigate possible factors that might be contributing to these trends, subgroup analyses were performed based on age group and clinical history of CHIKV infection stratified by the collection intervals of paired serum samples. Accordingly, median (IQR) IgG levels among women seropositive at D0 (2.58 [2.03-3.55]), and with self-reported prior history of CHIK, trending to decrease over the short-term period of D60-210 after baseline (2.39 [1.90-3.51]) (Figure 2A). Among women with no self-reported clinical history of CHIK, median (IQR) IgG levels decreased significantly at D60-210 follow-up (2.74 [2.15-3.32] versus 2.17 [1.43-3.13], p<0.0001; Figure 2A). At the mid-term period of D240-450 after baseline, we also observed a decrease of median (IQR) IgG levels independently of prior history of CHIK diagnosis (prior diagnosis, 2.67 [2.17-3.68] versus 2.39 [1.74-2.99], p=0.0039; no prior diagnosis, 2.63 [1.99-3.29] versus 2.37 [1.57-3.23], p=0.0004; Figure 2C).
In addition, across age groups, we found that median (IQR) IgG levels decreased significantly in the 15-24 years age group both at D60-210 and D240-450 follow-up (2.78 [2.15-3.17] versus 2.18 [1.73-2.81], p<0.0001, Figure 2B; 2.58 [2.04-3.13] versus 2.35 [1.57-3.08], p=0.0056, Figure 2D). Also, we found similar patterns in the 25-39 years age group (2.65 [2.03-3.55] versus 2.27 [1.43-3.51], p=0.0002, Figure 2B; 2.68 [1.99-3.68] versus 2.39 [1.58-3.23], p=0.0003, Figure 2D). Altogether, despite these initial differences at baseline, we observed similar trends in the decay of IgG levels between the subgroups analyzed that were not apparently affected by prior history of CHIK or age at the endpoint of this study (Figure 2).

Discussion
The city of Fortaleza registered the first wave of CHIK in 2016 with 17,791 cases followed by a second wave in 2017 with 61,828 cases reported10. In this study conducted post-outbreak, we demonstrated a high baseline prevalence of IgM/IgG antibodies to CHIKV of 34% (95% CI, 30.2 - 38.0) in a subsample of a cohort of women in Fortaleza. Seroprevalence slightly decreased to 31.3% (175/559; 95% CI, 27.6 - 35.3) over the first follow-up period. Compared to some studies conducted in urban areas in Northeast Brazil, the seroprevalence of CHIKV observed in our study was lower than those found in Feira de Santana (57.1%; n=385) and Riachão do Jacuípe (45.7%; n=446) in Bahia State, 201521. Furthermore, a serological survey in 2016/2017 performed on 1,981 sera from the city of Feira de Santana found a seroprevalence of 22.1%, a percentage much lower than that observed previously22. Also, a cross-sectional, community-based study conducted in an urban slum area in Salvador (Bahia State) found a CHIKV seroprevalence of 11.8% (n=1,776)23. There were no significant differences in the distribution of CHIKV seropositivity by age group in our study. If we consider cumulative seroprevalence in the age group 15-39 years, we found it was 2.9 times higher (34%) than the 11.9% observed in a study conducted in Salvador for the same age group and, even though the latter included both male and female in the analysis, they found no differences in the prevalence of previous CHIKV infection by gender23. Altogether, these variations might be partially explained by differences in the intensity of CHIKV transmission and the intervening interval between the serosurvey and the outbreak in a naive host population, highlighting the importance of serological surveillance to estimate the intensity of infection spread in order to better understand the potential for future epidemics.
In addition to seroprevalence data on prior CHIKV infection, our report aimed to provide information regarding CHIKV IgG levels kinetics over time in paired samples from a female cohort study. To the best of our knowledge, our study was the first that explored the short- and mid-term kinetics of antibodies responses against CHIKV following the major CHIKV outbreak in the city of Fortaleza in 2017. Interestingly, although IgG seroprevalence did not differ significantly at the first follow-up (28.3%, 158/559; 95% CI, 24.7 - 32.1) in relation to baseline (27.9%, 156/559; 95% CI, 24.3 - 31.8), we found evidence of an overall decline in CHIKV IgG levels. However, it remains unknown whether CHIKV IgG levels stabilized after the first follow-up or continued declining. Additionally, we did not evaluate neutralizing antibodies titers (NAb) against CHIKV, although there is evidence to support the notion that IgG levels are linked to NAb titers and protection against (re)infection24. We can also infer that the likelihood of CHIKV exposure post-outbreak was low due to the overall antibody contraction observed in our analysis. Conversely, a proportion of 20.4% (31/152) paired serum samples displayed increased IgG levels over time, suggesting either a possible subclinical infection due to re-exposure to the virus or casual variation at individual level. Noteworthy, we observed 4 cases of IgG seroreversion (IgG+ to IgG-), indicating the possibility of misinterpreting results in cross-sectional seroprevalence studies in cases when CHIKV antibody kinetics are not well characterized25. Interestingly, a report using plasma from patients (n=9) with acute CHIK demonstrated that total IgG titers decreased over time albeit still detectable neutralizing capacity at 21 months after post-illness onset26. It is yet to be addressed, however, the impact on further susceptibility, reason that makes it important to characterize long-last immunity development as a consequence of possible subsequent (re)infections rather than from primary infections only. Although we did not provide evidence of reinfection among the study participants, we found that seroreversion may occur, which pose the question about the duration of immunity in a post-epidemic period giving the major role of IgG antibodies against CHIKV reinfection and/or development of disease. Whether waning immunity could potentially contribute to a hypothetical scenario of reinfection in adults with prior history of CHIK (with or without symptoms) needs further investigation given the fact there is only one CHIKV serotype described so far16.
Moreover, CHIKV detection in Brazil occurred almost concomitantly with the emergence of ZIKV cases in the country (2014/2015)27. Our serological data regarding the kinetics of CHIKV antibody responses is similar to the data from a study in Salvador, where a rapid decline in antibody responses against ZIKV was found 1.5-2 years after the outbreak28. These observations are in line with other studies and suggest that the previous assumption of long-lasting immunity based on antibodies following natural arbovirus infection may not always be the case29-31. In addition, it has been previously demonstrated that homotypic DENV reinfections may occur, as well as the molecular detection of CHIKV in paired sera revealed that reinfection is common among infants32,33. To add complexity, cases of co-infections have been already reported in areas of co-circulation of DENV/ZIKV/CHIKV and how the immune responses are modulated in these scenarios requires better assessment34,35.
Lastly, we found no significant differences at the endpoint of the study in median (IQR) IgG levels between the age groups 15-24 years and 25-39 years as well as between those with self-reported CHIK diagnosis (symptomatic cases) and without CHIK diagnosis (asymptomatic cases) prior to enrollment in the cohort. In 2022, a third wave of CHIK was reported in the city of Fortaleza, higher than the first wave in 2016, with 20,227 confirmed cases10. The second follow-up is currently going on in a subsample of our cohort. The findings we reported here were obtained 4 years before this new outbreak and how it may be related to and can be influenced by antibody contraction and seroreversion are questions that need to be addressed in future studies on population-level effects of individual-level immunity.
Our study has several limitations that indicate further research is warranted. First, we were limited by the primary objectives and design of the original study. In fact, we did not know when the CHIKV infection actually occurred in the participants, a variable which may have implications for the interpretation of IgG levels in paired samples. Loss of follow-up compromised our sample size leading to the potential introduction of bias in our study. In line with this, estimation of the prevalence should be interpreted with caution due to the relative high rate of inconclusive results. Additionally, distribution of paired samples according to the collection intervals (baseline and follow-up) was also a limitation due to variations in the number of samples in the smaller or larger single intervals, which leads us to split into two broad intervals in order to provide better distribution of samples for further analyses. Importantly, the results should not be extrapolated to the whole population of Fortaleza, including males and other age groups not investigated in our study. Moreover, we did not perform neutralization assays to evaluate anti-CHIKV NAb titers. Considering clinical misdiagnosis and memory effects, self-reported diagnosis of CHIK by a physician prior to enrollment could be a source of bias due to the similarity of some symptoms also reported from individuals infected by DENV or ZIKV.

Conclusions
Despite these limitations, the findings of our study can provide a framework for further cohort studies with specific designs to understand the kinetics of antibody responses and waning immunity during CHIKV infection. This has important implications for accurate surveys of CHIKV seroprevalence in urban settings, critical information for vaccine development, and planning for epidemiological surveillance and new outbreaks.

Funding
This study was financed by the National Council for Scientific and Technological Development (CNPq), Brazil, grant #440778/2016-6, #312493/2021-5 (PQ), and #402403/2020-7 (EV1); Coordination for the Improvement of Higher Education Personnel (CAPES), Brazil, grant #8887.130.795/2016-00 and #88881.130806/2016-01; Foundation to Support Scientific and Technological Development of Ceará (FUNCAP), Ceará, Brazil, grant #3898920/2017; Coordination for the Improvement of Higher Education Personnel (CAPES), Brazil, Finance Code 001.

Competing interests
None declared
Ethical approval
The study was approved by the Institutional Ethics Committee of the Federal University of Ceará, Fortaleza, Brazil (No. 2.108.291). Written informed consent was obtained from each participant or their legally responsible guardians. This project was conducted according to the Brazilian Resolution 466/12 regarding research ethics involving human subjects.

Data availability
The data sets used to support the findings of this study are available upon reasonable request. Of note, they are not publicly available due to Brazilian legislation.

References
1. Bettis AA, L'Azou Jackson M, Yoon IK, Breugelmans JG, Goios A, Gubler DJ, Powers AM. The global epidemiology of chikungunya from 1999 to 2020: A systematic literature review to inform the development and introduction of vaccines. PLoS Negl Trop Dis 2022; 16(1):e0010069.

2. Yactayo S, Staples JE, Millot V, Cibrelus L, Ramon-Pardo P. Epidemiology of chikungunya in the Americas. J Infect Dis 2016; 214(suppl 5):S441–S445.

3. Fritzell C, Rousset D, Adde A, Kazanji M, Van Kerkhove MD, Flamand C. Current challenges and implications for dengue, chikungunya and Zika seroprevalence studies worldwide: A scoping review. PLoS Negl Trop Dis 2018; 12(7):e0006533.

4. Malvy D, Ezzedine K, Mamani-Matsuda M, Autran B, Tolou H, Receveur MC, Pistone T, Rambert J, Moynet D, Mossalayi D. Destructive arthritis in a patient with chikungunya virus infection with persistent specific IgM antibodies. BMC Infect Dis 2009; 9:200.

5. Silva LA, Dermody TS. Chikungunya virus: epidemiology, replication, disease mechanisms, and prospective intervention strategies. J Clin Invest 2017; 127(3):737-749.

6. Economopoulou A, Dominguez M, Helynck B, Sissoko D, Wichmann O, Quenel P, Germonneau P, Quatresous I. Atypical chikungunya virus infections: clinical manifestations, mortality and risk factors for severe disease during the 2005- 2006 outbreak on Réunion. Epidemiol Infect 2009; 137(4):534-541.

7. Rajapakse S, Rodrigo C, Rajapakse A. Atypical manifestations of chikungunya infection. Trans R Soc Trop Med Hyg 2010; 104(2): 89–96.

8. Nunes MR, Faria NR, de Vasconcelos JM, Golding N, Kraemer MU, de Oliveira LF, Azevedo Rdo S, da Silva DE, da Silva EV, da Silva SP, Carvalho VL, Coelho GE, Cruz AC, Rodrigues SG, Vianez JL Jr, Nunes BT, Cardoso JF, Tesh RB, Hay SI, Pybus OG, Vasconcelos PF. Emergence and potential for spread of Chikungunya virus in Brazil. Emergence and potential for spread of Chikungunya virus in Brazil. BMC Med 2015; 13:102.

9. Simião AR, Barreto FKA, Oliveira RMAB, Cavalcante JW, Lima Neto AS, Barbosa RB, Lins CS, Meira AG, Araújo FMC, Lemos DRQ, Alencar CH, Cavalcanti LPG. A major chikungunya epidemic with high mortality in northeastern Brazil. Rev Soc Bras Med Trop 2019; 52:e20190266.

10. Brasil. Secretaria Municipal de Saúde de Fortaleza. Monitoramento de casos de Dengue, Chikungunya e Zika até a Semana Epidemiológica 52 [Internet]. 2022. [acessado 2023 jan 15]. Disponível em: https://ms.dados.sms.fortaleza.ce.gov.br/BoletimArbovirose2022/ARBOVIROSE2022S52.pdf

11. Brasil. Ministério da Saúde (MS). Secretaria de Vigilância em Saúde. Monitoramento dos casos de arboviroses até a semana epidemiológica 50 [Internet]. 2022. [acessado 2023 jan 15]. Disponível em: https://www.gov.br/saude/pt-br/centrais-de-conteudo/publicacoes/boletins/epidemiologicos/edicoes/2022/boletim-epidemiologico-vol-53-no47/view.

12. Metcalf CJ, Farrar J, Cutts FT, Basta NE, Graham AL, Lessler J, Ferguson NM, Burke DS, Grenfell BT. Use of serological surveys to generate key insights into the changing global landscape of infectious disease. Lancet 2016; 388(10045):728-730.

13. Drakeley CJ, Corran PH, Coleman PG, Tongren JE, McDonald SL, Carneiro I, Malima R, Lusingu J, Manjurano A, Nkya WM, Lemnge MM, Cox J, Reyburn H, Riley EM. Estimating medium- and long-term trends in malaria transmission by using serological markers of malaria exposure. Proc Natl Acad Sci U S A 2005; 102(14):5108-5113.

14. Boni MF, Chau NV, Dong N, Todd S, Nhat NT, de Bruin E, van Beek J, Hien NT, Simmons CP, Farrar J, Koopmans M. Population-level antibody estimates to novel influenza A/H7N9. J Infect Dis 2013; 208(4):554-558.

15. Lum FM, Teo TH, Lee WW, Kam YW, Rénia L, Ng LF. An essential role of antibodies in the control of Chikungunya virus infection. J Immunol 2013; 190(12):6295-6302.

16. Auerswald H, Boussioux C, In S, Mao S, Ong S, Huy R, Leang R, Chan M, Duong V, Ly S, Tarantola A, Dussart P. Broad and long-lasting immune protection against various Chikungunya genotypes demonstrated by participants in a cross-sectional study in a Cambodian rural community. Emerg Microbes Infect 2018; 7(1):13.

17. Milligan GN, Schnierle BS, McAuley AJ, Beasley DWC. Defining a correlate of protection for chikungunya virus vaccines. Vaccine 2019; 37(50):7427-7436.

18. Roques P, Fritzer A, Dereuddre-Bosquet N, Wressnigg N, Hochreiter R, Bossevot L, Pascal Q, Guehenneux F, Bitzer A, Corbic Ramljak I, Le Grand R, Lundberg U, Meinke A. Effectiveness of CHIKV vaccine VLA1553 demonstrated by passive transfer of human sera. JCI Insight 2022; 7(14):e160173.

19. Yoon IK, Alera MT, Lago CB, Tac-An IA, Villa D, Fernandez S, Thaisomboonsuk B, Klungthong C, Levy JW, Velasco JM, Roque VG Jr, Salje H, Macareo LR, Hermann LL, Nisalak A, Srikiatkhachorn A. High Rate of Subclinical Chikungunya Virus Infection and Association of Neutralizing Antibody with Protection in a Prospective Cohort in The Philippines. PLoS Negl Trop Dis 2015; 9(5):e0003764.

20. Correia F, Kerr L, Frota C, Barreto I, Almeida R, Pamplona L, Araújo L, Myiajima F, Lima S, Araújo F, Simões L, Júnior F, Martins A, Dias L, Alcântara I, Silva A, Sanhueza C, Ribeiro M, Teixeira M, Meyer A, Júnior J, Vasconcelos L, Sousa P, Saraiva S, Wesson D, Kendall C. Factors associated with Chikungunya infection in a cohort of women aged 15-39 y in Fortaleza, Brazil. Trans R Soc Trop Med Hyg 2021; 115(9):1070-1079.

21. Dias JP, Costa MDCN, Campos GS, Paixão ES, Natividade MS, Barreto FR, Itaparica MSC, Goes C, Oliveira FLS, Santana EB, Silva NSJ, Brito CAA, Rodrigues LC, Sardi SI, Saavedra RC, Teixeira MG. Seroprevalence of Chikungunya Virus after Its Emergence in Brazil. Emerg Infect Dis 2018; 24(4):617-624.

22. Teixeira MG, Skalinski LM, Paixão ES, Costa MDCN, Barreto FR, Campos GS, Sardi SI, Carvalho RH, Natividade M, Itaparica M, Dias JP, Trindade SC, Teixeira BP, Morato V, Santana EB, Goes CB, Silva NSJ, Santos CAST, Rodrigues LC, Whitworth J. Seroprevalence of Chikungunya virus and living conditions in Feira de Santana, Bahia-Brazil. PLoS Negl Trop Dis 2021; 15(4):e0009289.

23. Anjos RO, Mugabe VA, Moreira PSS, Carvalho CX, Portilho MM, Khouri R, Sacramento GA, Nery NRR Jr, Reis MG, Kitron UD, Ko AI, Costa F, Ribeiro GS. Transmission of Chikungunya Virus in an Urban Slum, Brazil. Emerg Infect Dis 2020; 26(7):1364-1373.

24. Chua CL, Sam IC, Merits A, Chan YF. Antigenic Variation of East/Central/South African and Asian Chikungunya Virus Genotypes in Neutralization by Immune Sera. PLoS Negl Trop Dis 2016; 10(8):e0004960.

25. Loesche M, Karlson EW, Talabi O, Zhou G, Boutin N, Atchley R, Loevinsohn G, Chang JBP, Hasdianda MA, Okenla A, Sampson E, Schram H, Magsipoc K, Goodman K, Donahue L, MacGowan M, Novack LA, Jarolim P, Baden LR, Nilles EJ. Longitudinal SARS-CoV-2 Nucleocapsid Antibody Kinetics, Seroreversion, and Implications for Seroepidemiologic Studies. Emerg Infect Dis 2022; 28(9):1859-1862.

26. Kam YW, Lee WW, Simarmata D, Harjanto S, Teng TS, Tolou H, Chow A, Lin RT, Leo YS, Rénia L, Ng LF. Longitudinal analysis of the human antibody response to Chikungunya virus infection: implications for serodiagnosis and vaccine development. J Virol 2012; 86(23):13005-13015.

27. Silva MMO, Tauro LB, Kikuti M, Anjos RO, Santos VC, Gonçalves TSF, Paploski IAD, Moreira PSS, Nascimento LCJ, Campos GS, Ko AI, Weaver SC, Reis MG, Kitron U, Ribeiro GS. Concomitant Transmission of Dengue, Chikungunya, and Zika Viruses in Brazil: Clinical and Epidemiological Findings from Surveillance for Acute Febrile Illness. Clin Infect Dis 2019; 69(8):1353-1359.

28. Moreira-Soto A, de Souza Sampaio G, Pedroso C, Postigo-Hidalgo I, Berneck BS, Ulbert S, Brites C, Netto EM, Drexler JF. Rapid decline of Zika virus NS1 antigen-specific antibody responses, northeastern Brazil. Virus Genes 2020; 56(5):632-637.

29. Langerak T, Kasbergen LMR, Chandler F, Brinkman T, Faerber Z, Phalai K, Ulbert S, Rockstroh A, Bruin E, Koopmans MPG, Rockx B, van Gorp ECM, Vreden S. Zika Virus Antibody Titers Three Years after Confirmed Infection. Viruses 2021; 13(7):1345.

30. López L, Paul RE, Cao-Lormeau VM, Rodó X. Considering waning immunity to better explain dengue dynamics. Epidemics 2022; 41:100630.

31. Ribeiro GS, Hamer GL, Diallo M, Kitron U, Ko AI, Weaver SC. Influence of herd immunity in the cyclical nature of arboviruses. Curr Opin Virol 2020; 40:1-10.

32. Waggoner JJ, Balmaseda A, Gresh L, Sahoo MK, Montoya M, Wang C, Abeynayake J, Kuan G, Pinsky BA, Harris E. Homotypic Dengue Virus Reinfections in Nicaraguan Children. J Infect Dis 2016; 214(7):986-993.

33. Nyamwaya DK, Otiende M, Omuoyo DO, Githinji G, Karanja HK, Gitonga JN, R de Laurent Z, Otieno JR, Sang R, Kamau E, Cheruiyot S, Otieno E, Agoti CN, Bejon P, Thumbi SM, Warimwe GM. Endemic chikungunya fever in Kenyan children: a prospective cohort study. BMC Infect Dis 2021; 21(1):186.

34. Frota CC, Correia FGS, Alves Vasconcelos LR, de Sousa PRC, Ferreira MLDS, Saraiva SP, Mota Ferreira R, Romcy KAM, Pinheiro RF, de Oliveira RTG, Pinheiro Júnior FML, Martins AF, Sanhueza-Sanzana C, de Almeida RLF, Kendall C, Coelho ICB, Pires Neto RDJ, Aguiar ÍWO, Kerr LRFS. Positivity of dengue, chikungunya, and Zika infections in women in Northeast Brazil post-Zika epidemic. Pathog Glob Health 2022; 1-8.

35. Waggoner JJ, Gresh L, Vargas MJ, Ballesteros G, Tellez Y, Soda KJ, Sahoo MK, Nuñez A, Balmaseda A, Harris E, Pinsky BA. Viremia and Clinical Presentation in Nicaraguan Patients Infected With Zika Virus, Chikungunya Virus, and Dengue Virus. Clin Infect Dis 2016; 63(12):1584-1590.

Figure legends
Figure 1. Overview of the IgG seropositivity and anti-CHIKV IgG levels kinetics during the cohort study in Fortaleza, Brazil, 2018-2019. (A) Seropositivity of IgG antibodies against CHIKV. (B) Anti-CHIKV IgG ELISA ratio at baseline (day 0, D0) and after first follow-up. Each dot represents an individual participant. Dotted lines indicate the assay cutoff limits for IgG positivity. 2-tailed Wilcoxon matched-pairs signed-rank test was performed to evaluate statistical significance. Value that is significantly different is indicated by bar and asterisks as follow: ****, p<0.0001.
Source: Authors.

Figure 2. Influence of age and prior CHIK clinical diagnosis on the anti-CHIKV IgG antibodies kinetics among paired plasma samples from participants in the cohort study, Fortaleza, Brazil, 2018-2019. (A and C) Median ELISA ratios of anti-CHIKV IgG stratified by history of CHIK clinical diagnosis and interval of plasma collection. (B and D) Median ELISA ratios of anti-CHIKV IgG in different age groups in years. Each dot represents an individual participant. Dotted lines indicate the assay cutoff limits for IgG positivity. Mann-Whitney U test and 2-tailed Wilcoxon matched-pairs signed-rank test were used to evaluate differences between independent groups and paired samples, respectively. Asterisks indicate statistically significant differences as follows: **, p=0.0039 (Fig.2C), p=0.0056 (Fig.2D); ***, p=0.0002 (Fig.2B), p=0.0004 (Fig.2C), p=0.0003 (Fig.2D); ****, p<0.0001 (Fig.2A and 2B). ns = non-significant. D=days.
Source: Authors.