Abstract

Author summary

Citation: Chen X, Munoz B, Mkocha H, Gaydos CA, Dize L, Quinn TC, et al. (2022) Risk of seroconversion and seroreversion of antibodies to Chlamydia trachomatis pgp3 in a longitudinal cohort of children in a low trachoma prevalence district in Tanzania. PLoS Negl Trop Dis 16(7): e0010629. https://doi.org/10.1371/journal.pntd.0010629

Editor: Prashant Kumar, University of Kansas School of Pharmacy, UNITED STATES

Received: February 14, 2022; Accepted: July 1, 2022; Published: July 13, 2022

Copyright: © 2022 Chen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: This work was supported by the Task Force for Global Health, Inc. [1491 to SKW], the El-Maghraby chair at the Wilmer Eye Institute [to SKW], and the Division of Intramural Research, National Institutes of Allergy and Infectious Diseases [to TCQ]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Trachoma, the leading infectious cause of blindness worldwide, is a chronic conjunctivitis caused by repeated infection with Chlamydia trachomatis (Ct) [1]. The World Health Organization (WHO) set one of the elimination goals as sustained district-level prevalence of trachomatous inflammation–follicular (TF) in children aged 1–9 years less than 5% [2]. This is measured in two surveys at least two years apart with no intervening Mass Drug Administration (MDA). As more endemic regions approach the elimination goal, it becomes critical to identify proper methods of detecting re-emergence. If surveys find that trachoma increases (or goes above 5% in the case of trachoma surveillance surveys), more resource may need to be mobilized to manage resurgence.

However, the precision of TF grading becomes problematic when trachoma prevalence is low. There is tendency to overcall follicles [3], and there is clearly intergrader variability even when agreement is acceptable [4]. At the same time, a test of infection for surveillance purposes is also controversial as sensitive commercial tests are expensive, there are no guidelines or standardization on usage, and evidence shows that a low level of infection does not necessarily lead to trachoma re-emergence [5,6].

Serologic testing for chlamydial pgp3 antigen is a promising tool. Compared to surveys of TF and infection, which are snapshots of the current prevalence, early data suggest that seropositivity measures cumulative exposure to ocular Ct [7]. In areas where TF remains hyperendemic, antibody prevalence is high with increases in seropositivity with age, indicating ongoing transmission [7,8]. In surveillance surveys where TF was <5%, with low or almost absent infection, seropositivity rates were low, with zero to modest rates of increase with age [3,9,10]. A recent study identified re-emergence of trachoma based on TF of 7.1%, verified by infection and photo confirmation of clinical grades. There was a seroprevalence of 18.2% that increased with age; notably, a seropositivity rate of 6.7% in 1-2-year-olds, who were born after cessation of MDA, was strong evidence for ongoing transmission [11].

While serologic data provide useful information, there is limited understanding of the long-term stability of seropositivity in children in formerly endemic districts. To fully model district rates of seroconversion and possible transmission, this understanding is critical. A small study in a trachoma hyperendemic village found that seropositivity in a cohort of children was stable over six months. While MDA lowered the Median Fluorescence Intensity value of the antibodies, no child lost seropositivity [7]. However, in a district with TF of 5.2%, we followed a cohort of children in the absence of MDA over a year and found 6.4% of seropositive children at baseline seroreverted at one year; seroreverters were also more likely to live in communities with TF<5% [12]. Six months after one round of MDA, the seroreversion rate was 4% (estimated yearly rate of 8%) [13]. However, the stability of these changes over many years is unknown.

In this study, we followed a cohort of children over three years in a low endemic district and determined the stability of their antibody status and trajectories of change in serostatus. We hypothesized that seroconversion and seroreversion were more common in younger children, and that children residing in communities with TF≥5% were more likely to seroconvert while children residing in communities with TF<5% were more likely to serorevert.

Methods

Results

A total of 2536 children were surveyed at baseline and 1719 (67.8%) showed up to all four surveys (Fig 1). The cohort consisted of these 1719 children.

The percent of females was greater in the cohort compared to those who missed one follow-up (52.2% vs. 44.3%, P = 0.0023, Table 1). There was no difference by age, baseline TF, infection, or seropositivity.

The cross-sectional TF, infection, and seropositivity proportions of all the participants that showed up to each survey are shown in Table 2. Overall, TF decreased significantly (P<0.005) as did infection (P = 0.01). Seropositivity increased over the three years (P<0.001).

The age-specific seropositivity rates for each survey are shown in S1 Fig, where the increasing age of the cohort means that the youngest age groups with the lowest seropositivity rate are not represented at the three-year survey. S2 Fig shows the same age group at each survey point, suggesting the increase in seropositivity is driven by increases in the children aged five years and younger. The 9-year-olds at second and third follow-ups, who were 6-7-year-olds at baseline, had lower seropositivity rates than 9-year-olds at baseline (P = 0.047 and P = 0.0022, S2G Fig), likely because they lived through periods of lower overall TF and lower overall infection than the baseline group of 9-year-olds.

The cohort was divided into seven groups, based on distinct individual pattern of change in serostatus (Table 3). Of those who were seropositive at baseline, most children (n = 450, 86.2%) remained seropositive at all follow-ups; 63 (12.1%) were seroreverters. In a multivariable model adjusted for clustering (Table 4), seroreverters were younger (Odds Ratio [OR] = 0.88 for every one-year increase in baseline age, 95% CI = 0.79–0.99) and more likely to be female (OR = 1.77, 95% CI = 1.02–3.07). They also tended to live in a village with a baseline TF<5% (OR = 1.27, 95% CI = 0.68–2.39), although the result was not statistically significant. We did not have enough power to study the effect of infection rate on seroreversion as 96% of children lived in a village with zero infection at baseline. A test of interaction between age and sex was not significant. In sensitivity analyses including 96 participants lost to one follow-up who were either seronegative throughout or seroreverters, the results did not change.

Of those in the cohort who were seronegative at baseline, 1011 (84.5%) remained seronegative throughout the study; 156 (13.0%) were seroconverters. In a multivariable model adjusting for clustering (Table 4), seroconverters were younger (OR = 0.92 for every one-year increase in baseline age, 95% CI = 0.87–0.98). They also tended to be female (OR = 1.35, 95% CI = 0.94–1.95) and more likely to live in a village with a baseline TF≥5% (OR = 1.37, 95% CI = 0.78–2.39), although the latter two results were non-significant. In sensitivity analyses including the 203 participants lost to one follow-up who were either seronegative throughout or seroconverters, female sex was significantly associated with seroconversion (OR = 1.51, 95% CI = 1.08–2.11). The odds of seroconverters residing in a village with estimated TF ≥5% increased to 1.52 but was not statistically significant (P = 0.13). Overall, younger children and female children had more fluidity in serostatus, and residence in a village with TF≥5% likely increases the risk of seroconversion.

We examined the MFI-BG in seroreverters and the effect of village TF prevalence at baseline on the change in intensity of antibody response. As shown in Fig 2, the change in log MFI-BG was similar in children in both sets of villages. We evaluated the significance in children whose pre- and post- seroreversion survey used the same bead sets, and the mean decrease in log MFI-BG was 0.54 (P<0.001).

The change in MFI-BG was similarly evaluated among children who were seroconverters (Fig 3). There was no effect on the intensity of change according to estimated village TF prevalence at baseline. We evaluated the significance in children whose pre- and post- seroconversion survey used the same bead sets, and the mean increase in log MFI-BG was 2.49 (P<0.001).

There were 31 (2%) children who experienced a single or transient seroconversion or seroreversion (Groups 3 and 6). We examined the effect of MFI-BG values falling in the indeterminate range on these findings. Only one child (3% of the 31) had the single change with a value of MFI-BG in the indeterminate range. The other children in this group did not have MFI-BG values in the indeterminate ranges. The data suggests that these transient changes in serostatus are real and not due to values in the indeterminate ranges. Twenty of the 26 children (76.9%) who had one transient seroconversion were under age 6.

Of the longitudinal cohort, only eight participants (0.5%) had unstable serostatus (Group 7), frequently changing between seropositive and seronegative. The apparent instability of their serostatus can largely be explained by MFI-BG values in the indeterminant ranges, which affected five of the eight children (Table 5). Notably, all children in the unstable group were over five years old and lived in villages with TF≤5%. No one in the unstable group had TF/TI or Ct infection.

Discussion

In our cohort of children in a formerly trachoma hyperendemic district, we found that serostatus was largely stable, with approximately 85% of children retaining baseline serostatus over the three-year period. This was not 100%, as observed 6 months following MDA in a hyperendemic community [7], but we followed our cohort for a longer period, in a low endemic district but with some communities having TF prevalence estimated above and below 5%. Although one might expect to see more seroreversion after MDA, the two rounds of MDA in our study period did not, for the most part, seem to affect the overall stability of serostatus in children. This is consistent with our short-term results that showed a modest impact of MDA on MFI-BG, and a slight but not striking increase in seroreversion rates after MDA [13]. Less than 1% of children had truly unstable serostatus, largely due to MFI-BG values in the indeterminate range. Stability is a desirable quality when estimating seroconversion rates from age specific prevalence, if the assumption is made that once antibodies are acquired, they are not lost. However, such was not the case in our study, as we found strong evidence for seroreversion.

Seroconversion and seroreversion occurred at rates of 13% and 12% respectively during the three-year period. Very few children had MFI-BGs in the indeterminate ranges around the cut-offs, so the observed seroreversion and seroconversion are not artifacts from indeterminate ranges. The previous study in a hyperendemic community reported no seroreversion following MDA but was likely related to the short follow-up period and the high level of infection prior to MDA [7]. Previous modeling estimated the half-life of seroreversion to be 26 years [18], while our study suggested shorter timelines.

Our observation of seroreversion, and what seems to be real transient seroconversions, makes biological sense. While it is unclear how many infections are needed to maintain seropositivity, a single infection is not enough [19]. The higher cumulative antigen load over time boosts memory B cells and long-lived plasma cells that secrete high numbers of circulating antibodies [20], which then help to maintain seropositivity, even in the absence of continued disease exposure. Therefore, it is not surprising that we found younger children to have more fluidity in serostatus than older children. Older children have had more episodes of Ct exposure in their lifetime, so they are more likely to develop long-lasting immune memory and stable serostatus. This is also demonstrated in the group with transient changes in serostatus, most of whom were under age six. Our data support a recommendation that modeling efforts to estimate seroconversion should also account for age specific estimates of seroreversion in order not to underestimate seroconversion. There might be concern that seropositivity in the youngest children is due to transfer of maternal antibodies to urogenital Ct strains and that the waning of maternal antibodies contributed to the loss of seropositivity. However, the decline of maternal antibodies occurs between 6–12 months [21], which should only contribute to seroreversion in children <1 year old. No child in our study was under age one.

Seroconversion was also observed in this cohort, despite residing in a district where estimated trachoma and infection rates were low. Living in villages where the estimated TF rate based on our sample was ≥5% was associated with greater odds of seroconversion but it was not statistically significant. The intensity of the seroconversion did not differ by whether the estimated TF was less than 5% compared to 5% or more. As with seroreversion, the risk of seroconversion was significantly higher in the younger children. The greater risk of seroconversion in younger children likely reflects two factors. First, younger children are less likely to be seropositive, so as a group when they encounter an infection, there are more of them eligible to seroconvert. Second, in general, younger children have more infection and trachoma than older children suggesting their risk of exposure is greater than that in older children.

Monitoring or surveillance for trachoma should be for the purpose of detecting resurgence or re-emergent disease over a specific time frame. Our data suggest that the younger age groups, specifically 5 years and under, were better suited as the sample to reflect the greatest risk of seroconversion from repeated exposure to infection compared to the older age groups. In another study, following a district impact survey which found TF prevalence to be <5%, a subsequent survey found that the age group born after this impact assessment provided the serologic evidence that re-emergence had occurred [11].

We found that female children were more likely to seroconvert, and this association was statistically significant in sensitivity analyses. One possible reason is that young girls often assist in looking after younger siblings, which puts them in constant contact with other children. As young children are the main reservoir for Ct, female children are at higher risks of Ct infection than their male counterparts [22]. The levels of infection in this study were so low that further exploring risk factors in the girls was not possible.

Female children were also more likely to serorevert, suggesting more instability in their serostatus. While adult females tend to show greater antibody responses than males, the sex-based variation in immune responses can be more ambiguous in childhood before sex steroids start to exert major effects [23]. In fact, B cell numbers are comparable in male and female children [24,25]. In a study on the serum immunoglobulin levels in Nigeria, female children aged 5–12 were found to have equivalent IgG levels as male children, while female adults had significantly higher values of IgG compared to male adults [26]. The relative fluidity in female children’s response to trachoma antigen could be due to some sex difference in immunity that is yet unknown. To our knowledge, there is scarce research on the sex difference in serology in children, especially in those under 10 years old. More studies in this area are needed.

Only eight out of the 1719 children were in the unstable group, largely explained by MFI-BG values falling in the indeterminate range. If the serostatus instances where the values fall in the indeterminate range are instead counted as the alternate, then these children fall into a more regular pattern. Participants 2 and 3 in this group would be seroreverters, and participants 7 and 8 would have transient seroconversions. Only three children were truly alternating between seropositivity and seronegativity.

There are limitations to the study. First, the data reflect a formerly hyperendemic district, and the results might not apply in districts that have eliminated trachoma a long time ago, where seropositivity even in older ages may be low. However, the low TF and low infection in the district do reflect the condition expected during surveillance for trachoma re-emergence, where serology would be of most use. Second, two rounds of MDA within a short period occurred during the study period. Without MDAs, there would likely have been more seroconversions. Additionally, about 70% of children showed up to all follow-ups. Lost to follow up may theoretically lead to biases if there is differential inclusion of those more likely to seroconvert or serorevert. There were more baseline seropositive children in those who were lost to two or three follow-ups. However, there was no age or sex difference between seropositives who were lost to follow-ups and seropositives who remained in the study, so they were not more or less prone to seroreversion. Our sensitivity analyses using those missing only one follow-up did not affect any of our findings, except to add more children and make some associations statistically significant. Another limitation is the use of two different bead sets. We have previously reported the reproducibility of serostatus determination using different bead sets with different cut-off values for seropositivity, suggesting the use of two bead sets had limited impact on our findings [27].

In summary, our study in a cohort of children demonstrated the dynamics of seroconversion and seroreversion in this formerly endemic district with overall low rates of infection and disease. The finding of overall long-term stability in serostatus over time provides reassurance that cross-sectional studies likely reflect seropositivity rates over at least a three-year period. However, there is a dynamic interplay of seroconversion and seroreversion, especially in young children. Modeling efforts in serology in children should account for seroreversion to avoid underestimating the seroconversion rate. In addition, the dynamic nature of change in serostatus is most pronounced in the youngest children, so for monitoring purposes, programs should consider using the youngest age group as the target population. Our data suggest using the age group 5 years and younger, but the age group may be tailored to the specific epidemiology of the district and other factors, such as time since MDA cessation.

Supporting information

References

1. 1. Taylor HR, Burton MJ, Haddad D, West S, Wright H. Trachoma. Lancet. 2014;384: 2142–2152. pmid:25043452
   • View Article
   • PubMed/NCBI
   • Google Scholar
2. 2. World Health Organization. (2015). Technical consultation on trachoma surveillance: meeting report. September 11–12, 2014, Task Force for Global Health, Decatur, USA. World Health Organization. Available from: https://apps.who.int/iris/handle/10665/174085. Accessed October 8, 2021.
3. 3. Butcher R, Sokana O, Jack K, Sui L, Russell C, Last A, et al. Clinical signs of trachoma are prevalent among Solomon Islanders who have no persistent markers of prior infection with Chlamydia trachomatis. Wellcome Open Res. 2018;3: 14. pmid:29588922
   • View Article
   • PubMed/NCBI
   • Google Scholar
4. 4. Naufal F, West SK, Brady CJ. Utility of photography for trachoma surveys: A systematic review. Surv Ophthalmol. 2021: S0039-6257(21)00177–6. pmid:34425127
   • View Article
   • PubMed/NCBI
   • Google Scholar
5. 5. West SK, Munoz B, Weaver J, Mrango Z, Dize L, Gaydos C, et al. Can We Use Antibodies to Chlamydia trachomatis as a Surveillance Tool for National Trachoma Control Programs? Results from a District Survey. PLoS Negl Trop Dis. 2016;10(1): e0004352. pmid:26771906
   • View Article
   • PubMed/NCBI
   • Google Scholar
6. 6. Burton MJ, Holland MJ, Makalo P, Aryee EA, Alexander ND, Sillah A, et al. Re-emergence of Chlamydia trachomatis infection after mass antibiotic treatment of a trachoma-endemic Gambian community: a longitudinal study. Lancet. 2005;365(9467): 1321–1328. pmid:15823382
   • View Article
   • PubMed/NCBI
   • Google Scholar
7. 7. Goodhew EB, Morgan SM, Switzer AJ, Munoz B, Dize L, Gaydos C, et al. Longitudinal analysis of antibody responses to trachoma antigens before and after mass drug administration. BMC Infect Dis. 2014;14: 216. pmid:24755001
   • View Article
   • PubMed/NCBI
   • Google Scholar
8. 8. Cama A, Müller A, Taoaba R, Butcher RMR, Itibita I, Migchelsen SJ, et al. Prevalence of signs of trachoma, ocular Chlamydia trachomatis infection and antibodies to Pgp3 in residents of Kiritimati Island, Kiribati. PLoS Negl Trop Dis. 2017;11(9): e0005863. pmid:28898240
   • View Article
   • PubMed/NCBI
   • Google Scholar
9. 9. West SK, Zambrano AI, Sharma S, Mishra SK, Muñoz BE, Dize L, et al. Surveillance Surveys for Reemergent Trachoma in Formerly Endemic Districts in Nepal From 2 to 10 Years After Mass Drug Administration Cessation. JAMA Ophthalmol. 2017;135(11): 1141–1146. pmid:28973295
   • View Article
   • PubMed/NCBI
   • Google Scholar
10. 10. Nash SD, Astale T, Nute AW, Bethea D, Chernet A, Sata E, et al. Population-Based Prevalence of Chlamydia trachomatis Infection and Antibodies in Four Districts with Varying Levels of Trachoma Endemicity in Amhara, Ethiopia. Am J Trop Med Hyg. 2021;104(1): 207–215. pmid:33200728
    • View Article
    • PubMed/NCBI
    • Google Scholar
11. 11. Odonkor M, Naufal F, Munoz B, Mkocha H, Kasubi M, Wolle M, et al. Serology, infection, and clinical trachoma as tools in prevalence surveys for re-emergence of trachoma in a formerly hyperendemic district. PLoS Negl Trop Dis. 2021;15(4): e0009343. pmid:33861754
    • View Article
    • PubMed/NCBI
    • Google Scholar
12. 12. West SK, Munoz B, Kaur H, Dize L, Mkocha H, Gaydos CA, et al. Longitudinal change in the serology of antibodies to Chlamydia trachomatis pgp3 in children residing in a trachoma area. Sci Rep. 2018;8(1): 3520. pmid:29476106
    • View Article
    • PubMed/NCBI
    • Google Scholar
13. 13. West SK, Munoz B, Mkocha H, Gaydos CA, Quinn TC. The effect of Mass Drug Administration for trachoma on antibodies to Chlamydia trachomatis pgp3 in children. Sci Rep. 2020;10(1): 15225. pmid:32938957
    • View Article
    • PubMed/NCBI
    • Google Scholar
14. 14. Ervin AM, Mkocha H, Munoz B, Dreger K, Dize L, Gaydos C, et al. Surveillance and Azithromycin Treatment for Newcomers and Travelers Evaluation (ASANTE) Trial: Design and Baseline Characteristics. Ophthalmic Epidemiol. 2016;23(6): 347–353. pmid:27820670
    • View Article
    • PubMed/NCBI
    • Google Scholar
15. 15. Thylefors B, Dawson CR, Jones BR, West SK, Taylor HR. A simple system for the assessment of trachoma and its complications. Bull World Health Organ. 1987;65(4): 477–483. pmid:3500800
    • View Article
    • PubMed/NCBI
    • Google Scholar
16. 16. Dize L, West S, Quinn TC, Gaydos CA. Pooling ocular swab specimens from Tanzania for testing by Roche Amplicor and Aptima Combo 2 assays for the detection of Chlamydia trachomatis: accuracy and cost-savings. Diagn Microbiol Infect Dis. 2013;77(4): 289–291. pmid:24079951
    • View Article
    • PubMed/NCBI
    • Google Scholar
17. 17. Goodhew EB, Priest JW, Moss DM, Zhong G, Munoz B, Mkocha H, et al. CT694 and pgp3 as serological tools for monitoring trachoma programs. PLoS Negl Trop Dis. 2012;6(11): e1873. pmid:23133684
    • View Article
    • PubMed/NCBI
    • Google Scholar
18. 18. Pinsent A, Solomon AW, Bailey RL, Bid R, Cama A, Dean D, et al. The utility of serology for elimination surveillance of trachoma. Nat Commun. 2018;9(1): 5444. pmid:30575720
    • View Article
    • PubMed/NCBI
    • Google Scholar
19. 19. Martin DL, Saboyà-Díaz MI, Abashawl A, Alemayeh W, Gwyn S, Hooper PJ, et al. The use of serology for trachoma surveillance: Current status and priorities for future investigation. PLoS Negl Trop Dis. 2020;14(9): e0008316. pmid:32970672
    • View Article
    • PubMed/NCBI
    • Google Scholar
20. 20. Palm AE, Henry C. Remembrance of Things Past: Long-Term B Cell Memory After Infection and Vaccination. Front Immunol. 2019;10: 1787. pmid:31417562
    • View Article
    • PubMed/NCBI
    • Google Scholar
21. 21. Leuridan E, Hens N, Hutse V, Aerts M, Van Damme P. Kinetics of maternal antibodies against rubella and varicella in infants. Vaccine. 2011;29(11): 2222–2226. pmid:20558248
    • View Article
    • PubMed/NCBI
    • Google Scholar
22. 22. Congdon N, West S, Vitale S, Katala S, Mmbaga BB. Exposure to children and risk of active trachoma in Tanzanian women. Am J Epidemiol. 1993;137(3): 366–372. pmid:8452144
    • View Article
    • PubMed/NCBI
    • Google Scholar
23. 23. Klein SL, Flanagan KL. Sex differences in immune responses. Nat Rev Immunol. 2016;16(10): 626–638. pmid:27546235
    • View Article
    • PubMed/NCBI
    • Google Scholar
24. 24. Lee BW, Yap HK, Chew FT, Quah TC, Prabhakaran K, Chan GS, et al. Age- and sex-related changes in lymphocyte subpopulations of healthy Asian subjects: from birth to adulthood. Cytometry. 1996;26(1): 8–15. pmid:8809475
    • View Article
    • PubMed/NCBI
    • Google Scholar
25. 25. Lisse IM, Aaby P, Whittle H, Jensen H, Engelmann M, Christensen LB. T-lymphocyte subsets in West African children: impact of age, sex, and season. J Pediatr. 1997; 130(1): 77–85. pmid:9003854
    • View Article
    • PubMed/NCBI
    • Google Scholar
26. 26. Obiandu C, Okerengwo AA, Dapper DV. Levels of serum immunoglobulins in apparently healthy children and adults in Port Harcourt, Nigeria. Niger J Physiol Sci. 2013; 28(1): 23–27. pmid:23955402
    • View Article
    • PubMed/NCBI
    • Google Scholar
27. 27. Kaur H, Dize L, Munoz B, Gaydos C, West SK. Evaluation of the reproducibility of a serological test for antibodies to Chlamydia trachomatis pgp3: A potential surveillance tool for trachoma programs. J Microbiol Methods. 2018; 147: 56–58. pmid:29501689
    • View Article
    • PubMed/NCBI
    • Google Scholar