Introduction
The outcome of sexually transmitted Chlamydia trachomatis (Ct) and Mycoplasma genitalium (Mg) infections can be severe because of their potentially serious sequels on reproductive health, including pelvic inflammatory disease (PID), salpingitis, and tubal infertility [Reference Yu1,Reference Bébéar and de Barbeyrac2]. Globally, infections caused by Ct are the most prevalent sexually transmitted infections (STIs), second only to the parasite Trichomonas vaginalis (Tv) infections [Reference Bébéar and de Barbeyrac2]. According to global estimates, there are nearly 129 million new cases of Ct infections worldwide each year [Reference Rowley3]. To date, comprehensive global studies assessing the prevalence of Mg remain absent. However, a recent meta-analysis concluded the prevalence of Mg to be 1.3% and 3.9% in developed and developing countries, respectively [Reference Baumann4]. Importantly, both pathogens can cause asymptomatic infections, which are challenging in terms of diagnosis and treatment.
Ct and Mg present with analogous clinical behavior, causing similar clinical manifestations or being asymptomatic [Reference Yu1,Reference Bébéar and de Barbeyrac2,Reference Tuddenham, Hamill and Ghanem5,Reference Uusküla and Kohl6]. Case–control studies have shown that clinical carriage of Ct and Mg can be independent of one another [Reference Uusküla and Kohl6,Reference Keane7]; however, a study of female sex workers showed women with Ct and N. gonorrhea infection had an increased risk of Mg [Reference Cohen8]. This correlation is understandable as they share the same route of transmission. Additionally, a recent study at the National Sexual Health Clinic in Singapore found that Mg was strongly associated with Ct infection, present in 8.1% of cases compared to 2.4% of Ct-negative cases [Reference Hart9]. Nucleic Acid Amplification Tests (NAATs) or Polymerase Chain Reaction tests (PCRs) are reliable in the diagnosis of Ct and Mg, and in recent years, serological assays, among other methods, have been exploited to better understand the epidemiology of these infections [Reference Idahl10].
Quantification of different Ct and Mg serum antibodies can provide an accurate means to evaluate past or ongoing infections [Reference Idahl10–Reference Horner13]. Plasmid Gene Protein 3 (known as pgp3 or pGp3) is a highly immunogenic antigen during chlamydial infection [Reference Woodhall14–Reference Chen17], and notably, the antibodies produced against pGp3 exhibit a strong dependence on its protein conformation. In a UK-based study [Reference Horner13,Reference Wills18] that compared several assays in the same population, pGp3 serology was the most sensitive to detect a previously known Ct infection.
M. genitalium protein of adhesion (MgPa) is the major adhesin protein and the primary virulence factor of Mg, playing an essential role in mediating the attachment of the bacteria to host cells, thus facilitating their subsequent invasion [Reference Clausen11,Reference Li19,Reference Luo20]. Recombinant MgPa (rMgPa) is the C-terminal part of MgPa. Serological assays using MgPa and rMgPa antigens have shown promise in detecting Mg antibodies without cross-reactivity to Mycoplasma pneumoniae [Reference Clausen11,Reference Svenstrup21]. The interactions between Mg antigens and the host immune responses, however, remain largely unexplored.
The primary objective of this study was to cast further light on understanding the serological outcomes of Ct and Mg antibodies in young women and men and to exploit the possible co-factors role in Ct and Mg seropersistence.
Material and methods
Subjects
This study is based on the prospective Finnish HPV Family Study (FFHPV) cohort study jointly conducted at the Department of Obstetrics and Gynecology, Turku University Hospital (TUH), and the Institute of Dentistry, Faculty of Medicine, University of Turku, Finland [Reference Louvanto22]. The participants comprised 329 pregnant women and 135 fathers-to-be and their newborns who were enrolled (between 1998 and 2001) during the index pregnancy, at 36 weeks of pregnancy [Reference Louvanto22], and subsequently followed up for three years. The adult participants comprised spouses, however, the spouses of 194 women opted not to take part in the study, thus accounting for the difference in the number of male and female participants. All study participants were of Caucasian descent and shared the same ethnic background [Reference Vuorinen23]. The Research Ethics Committee of Turku University Hospital has approved this study’s design (#3/1998 and 2/2006, with amendments 45/1801/2018).
The participants were requested to complete a structured questionnaire at baseline and repeated as shorter versions at 3-year and 6-year visits. It encompassed over 60 meticulously designed questions, including information about demographics, sexual behavior, smoking habits, history of STIs, and overall health. Responses about the childbirth experience were collected from the mothers at approximately two months post-partum. To minimize the potential bias, the spouses participating in this study were deliberately kept uninformed regarding the content of his/her spouse’s questionnaires [Reference Louvanto22].
Serology
Blood samples were collected at the baseline and subsequently at 12-, 24- and 36-month follow-up visits from both spouses. The serological assays were performed in collaboration with the German Cancer Research Center (DKFZ), Heidelberg, Germany. For Ct, the serum IgG-antibodies to highly immunogenetic antigen pGp3 were assayed [Reference Woodhall14,Reference Li15]. Serology to Mg was analyzed using the primary virulence factor of this bacterium, MgPa, in two protein fragments, MgPa N-Term, and rMgPa [Reference Li19,Reference Luo20]. Studies about Mg serology are scanty, but antibodies targeting MgPa and rMgPa have been previously described [Reference Clausen11,Reference Trabert24]. The quantitative multiplex serology assay was used as previously described [Reference Trabert24,Reference Waterboer25]. This method is based on glutathione S-transferase (GST) capture ELISA combined with fluorescent-bead technology. The median fluorescence intensity (MFI) of at least 100 beads per antigen was measured, and the cut-off values for Ct (pGp3) seropositivity was MFI > 500 as well as for Mg MFI > 1,000 for both MgPa N-Term and rMgPa. Assay development and validation are described in detail earlier [Reference Trabert24].
Statistical analysis
The flowchart of the study design is illustrated in Supplementary Figure 1. Overall, 280 women and 115 men in the Ct antibody analysis and 264 women and 113 men in the Mg antibody analysis were included in the final statistical analyses. They all had at least two visits during the follow-up. Subjects with only one blood sample or no visits were excluded (50 women and 20 men). Additionally, we excluded participants with fluctuating serological results, e.g., seropositive at baseline, negative at the next follow-up, and then again positive (1 woman and 4 men from the Ct- and 17 women and 6 men from the Mg analyses).
The final cohorts were divided into subgroups according to the seropersistence to Ct and/or Mg. The always negative subgroup consisted of participants whose antibody levels remained below the defined cut-off value at every follow-up visit. The seropersistent subgroup consisted of the participants, who remained constantly seropositive to the given antigen(s) during the entire follow-up. The seroconversion group included individuals whose antibody levels transitioned from seronegative to seropositive, with an additional criterion of at least a two-fold increase over their previous serum measurement, after which all subsequent antibody levels for these individuals remained consistently seropositive throughout the follow-up period. Respectively, the serological decay subgroup consisted of those participants whose antibody titers were falling at least 50% from the previous serum titer below the cut-off values and remained negative until the end of the follow-up. For Mg, both antigens (Mg-Pa and rMgPa) had to increase/decrease the above amount to be included in the final serology outcomes groups.
Frequency tables were analyzed using the χ2 test, employing the likelihood ratio or Fisher’s exact test as appropriate for categorical variables. For the examination of differences in means of continuous variables, nonparametric tests (Mann–Whitney U test or Kruskal-Wallis) test were used. Crude and adjusted odds ratios (ORs) and their 95% confidence intervals (95%CI) were calculated by using logistic regression. In the adjusted model, age and all baseline co-factors that were statistically significant in the crude analysis were simultaneously included for mutual adjustment. All statistical analyses were performed using STATA MP17.0 (Stata Corp., College Station, TX, USA). All tests were run two-sided, and, in all analyses, probability values (p-values) of <0.05 were considered statistically significant.
Results
Ct seroprevalence at baseline was 32.4% among all women and 20.3% among all men. The corresponding results for Mg were 16.4% and 8.3%, respectively. The mean MFI antibody levels of pGp3 (Ct) and MgPa and rMgPa (Mg) among the seropersistent women and men stratified by their FU-visit are shown in Figure 1. The mean MFI levels of all three antibodies remained stable during the whole FU period of the seropersistent women. In men, however, variation of the levels of two IgG antibodies (MgPa and rMg) of Mg was seen, although the number of Mg seropositive men was much lower (only 5 to 6 men at each follow-up point).
Figure 1. Mean IgG-antibody levels to Chlamydia trachomatis antigen pGp3 and Mycoplasma genitalium antigens MgPa and rMgPa among the seropersistent women and men followed for 3 years. (a) Mean antibody levels to C. trachomatis in the seropersistent subgroup stratified by follow-up visit. The light gray bars represent women, and the darker gray bars represent men; (b) Mean antibody levels to M. genitalium in women (lighter gray for MgPa and darker gray for rMgPa) and (c) Mean antibody levels to M. genitalium in men (lighter gray for MgPa and darker gray for rMgPa). n = stands for the number of seropositive individuals at each visit.
The majority of both women and men remained seronegative to Ct, 65.7% and 81.7%, respectively. The same was true for Mg, to which 85.2% of women and 92.0% of men tested constantly seronegative. Persistent seropositivity to both Ct and Mg was more common in women (30.4% and 13.3%) than in men (17.4% and 5.3%), respectively. Seroconversion or serological decay were rare events. Additionally, there were only 20 women and 3 men who were persistently seropositive to Ct and Mg during the 3-year follow-up. Regarding the serological outcomes between couples, there were 72 couples in Ct- and 92 couples in Mg analyses with corresponding serological profiles, either seropositive or negative. These data are summarized in detail in Supplementary Table 1.
The demographic data stratified by the different serological outcomes of Ct and Mg among women are shown in Supplementary Table 2. The mean age of women varied between 23 to 27 years among the serological outcome groups. Women with persistent Ct antibodies were older than those always seronegative (26 years vs. 25 years, p = 0.016).
The potential co-factors associated with seropersistence for Ct and Mg in women during the follow-up were first evaluated with crude ORs (Supplementary Table 3), followed by Table 1, which has all significant baseline co-factors adjusted with each other. The increasing number of lifetime sexual partners was associated with persistent (pGp3) seropositivity to Ct, as the OR increased in parallel with the increasing number of lifetime sexual partners, up to crude OR 12.1 (95%CI 1.29–32.79) in those who reported more than 10 lifetime sex partners. This association stayed significant also with the adjusted model with the 6–10 and > 10 partners, with an OR of 5.5 and 5.6 (95%CI range of 1.39 to 22.29). Seropersistence to Mg was significantly associated only with more than 10 lifetime sex partners, crude OR 15.5 (95%CI 3.36–59.71). However, this did not stay significant after adjustment. Sexual debut at the age of 16 years or later was associated with lower odds for seropersistence to Ct and Mg separately or in combination, but after adjustment, none of these remained their significance. Women reporting more than 10 sexual partners before the age of 20 years had relations to becoming persistently seropositive to Ct: crude OR 5.2 (95%CI 1.82–14.72) or Mg: crude OR 14.6 (95%CI 3.57–59.71) or both: crude OR 27.7 (95%CI 4.18–182.92). Interestingly, after adjustment, only persistent Mg remained significant with an adjusted OR of 12.3 (95%CI 1.52–100.90).
Table 1. Age and all potential significant baseline co-factors* adjusted with each other of persistent seropositivity versus seronegativity to Chlamydia trachomatis and/or Mycoplasma genitalium among women in the Finnish Family HPV study
* Crude ORs shown in Supplementary Table 3.
† Participants consistently testing negative/positive to both C. trachomatis and M. genitalium.
In women, occasional anal intercourse was connected to Ct seropositivity (crude OR 2.1, 95%CI 1.07–4.04) and with the Mg and Ct combination seropositivity (crude OR 3.4, 95%CI 1.14–10.12), while oral sex had no such association. After adjustment, the occasional and regular anal intercourse remained significant with Ct only, adjusted OR 2.1 (95%CI 1.00–4.58) and 15.3 (95%CI 1.18–197.12), respectively. As expected, reported STIs were associated with Ct seropersistence in both crude and adjusted models but did not influence Mg seropersistence. In our study, only 28.2% of the women and 45.0% of the men seropersistent to Ct reported a history of diagnosed infection.
Also, when evaluating factors at the end of the study follow-up, a higher number of deliveries seemed to be protective against seropersistence to Ct. There was nearly a 15-fold increased likelihood (95%CI 1.88–112.17) for seropersistence to Ct when the use of contraceptive pills was started at the age of 14 to 16 years. The association was also evident among women who had changed their partner since entering the cohort, with OR ranging from 2.7 (95%CI 1.24–6.05) to 4.9 (95%CI 1.44–16.52).
Table 2 and Supplementary Table 4 summarize the corresponding data of the men. For men, no separate adjusted OR table was done due to only one significant finding after adjustment, as further clarified under. The mean age varied between 27 and 36 years among the different serogroups of Ct and Mg. Men with 6–10 sex partners by the age of 20 years were more likely seropersistent to Ct (<5 as the reference group) crude OR of 3.4 (95%CI 1.05–11.14), but this did not stay significant after adjustment with overall sex partners. Men reporting >5 lifetime sexual partners were significantly associated with Ct seropersistence, which remained significant also after adjustment, adjusted OR 12.6 (95%CI 1.55–102.49). Importantly, men who reported having had several partners during the index pregnancy were more likely to have simultaneous seropositivity to both bacteria, crude OR 36.0 (95%CI 1.61–805.20). Interestingly, none of the co-factors included in our analysis showed any significant association with Mg seropersistence in men.
Table 2. Potential co-factors of persistent seropositivity versus seronegativity to Chlamydia trachomatis and/or Mycoplasma genitalium among men during the three-year follow-up. Significant crude associations highlighted in bold and those that remained significant after adjustment are underlined
a Participants consistently testing negative/positive to for both C. trachomatis and M. genitalium.
b Adjusted OR 12.6 (1.55–102.49).
As the female and male subjects of our study represent marital couples, we also made a pair-wise analysis of the serological outcomes among the seropersistent couples. However, only a limited number of couples demonstrated co-existent seropersistence for Ct (n = 10) or Mg (n = 4). As measured by the mean seropositivity among these couples, we could not disclose any significant correlations between the couples at any time point (using a scatterplot). As expected, the scatterplot yielded a limited number of measurements with notable dispersion. The same limitation hampered the analysis of the potential co-factors in these couples, including the meaningful calculation of odds ratios. Noteworthy, however, is the fact across all three categories (Ct, Mg, and antibody co-occurrence), none of the couples with seropersistence reported a history of five or fewer lifetime sexual partners (data not shown).
Discussion
Our family study focused on the serological outcomes of Ct and Mg and also assessed the potential co-factors associated with their seropersistence, followed up for three years. The persistence of Ct antibodies was associated with a variety of sexual practices among women but less significantly among men. Among the women in our cohort, both adolescent and lifetime number of sexual partners were associated with both Ct and Mg seropersistence. Among the men in our study, no such associations were disclosed with Mg seropersistence. Until now, the majority of the risk factors for Ct and Mg have been focused on seropositivity, not seropersistence. However, it is important to understand i) whether the acquired IgG antibody levels remain stable, ii) which are the potential co-factors that predict the persistent seropositivity in a longitudinal setting, and iii) which are the possible consequences that might be attributed to seropersistence.
Previous studies have shown that both Ct and Mg are usually contracted during young adulthood. Ct usually peaks under the age of 25 in both genders [Reference Torrone26] while Mg peaks in males at 25–34 years and in females at 16–19 years [Reference Baumann4]. Microbe-specific serum IgG antibodies signify past or ongoing infection, though factors like antibody specificity, assay type, time since infection, and patient characteristics can affect IgG response and serological assay efficacy, particularly for prior Ct infections [Reference Wills18,Reference Geisler27]. It is known that mucosal Ct infections typically elicit an observable rise in IgG responses when suitable serological assays are used. Thus, antibody levels can reflect both past or ongoing Ct infection [Reference Horner13,Reference Wills18,Reference Geisler27,Reference Albritton28], and in a clinical setting to confirm the infection, additional characterizations of the pathogen’s nucleic acids are needed.
In a Finnish study, Ct IgG antibodies to MOMP (chlamydial major outer membrane protein) were detected in 65.5% of women within three months of infection onset, with approximately one-third remaining seropositive for 3–10 years after the initial infection [Reference Öhman29]. Another study of a female cohort using the same antigen as in the present study reported that pGp3 antibodies persisted for up to 12 years, and the antibody prevalence in the female cohort was higher than in their male counterparts [Reference Wills18]. Another study based on 9.695 biobank samples with a wider age distribution showed that pGp3 seroprevalence was 25.7% among women and 15.9% in men [Reference Mentzer30]. Our study confirms these observations on the stability of Ct antibodies and reports a similar prevalence of these antibodies (being higher in women), although the participants in our study were significantly younger.
As both Ct and Mg are STIs, we analyzed sexual practices, but also non-sexual co-factors as predictors of seropersistence. The present analysis revealed that an increasing number of lifetime sexual partners and early onset of sexual activities were significantly associated with seropersistence to both bacteria. Previous studies have confirmed the association between an increased number of sexual partners and the prevalence of Ct and Mg. In the study by Horner and colleagues [Reference Horner13], the OR increased in parallel with an increasing number of sexual partners, and in participants who reported having had 20 or more partners, OR for pGp3 seropositivity ranged from 49.1 to 58.7, depending on whether the participants had reported past Ct infection. Early sexual debut was also associated with seropositivity, consistent with our results. As for Mg infections [Reference Oakeshott31], it was reported that their prevalence was more common in women reporting two or more sexual partners in the previous year (adjusted OR 2.2). In another survey, a strong association of Mg with sexual risk behaviors was found in both genders [Reference Sonnenberg32]. Noteworthy is the fact that these studies focused on NAAT detection, being different from our approach.
The prevalence of Mg infection closely parallels the prevalence of Ct among women with high-risk sexual behavior, such as multiple partners and inconsistent condom use [Reference Huppert33]. Studies on simultaneous serology of Ct and Mg are scarce, and studies on seropersistence measured with different Mg antigens are nearly lacking. In our cohort, seropersistence to both pathogens was more common among women (11.1%) than men (3.5%). Persistence of Mg infection has been demonstrated before [Reference Cohen8,Reference Oakeshott31], although those studies used NAATs and had shorter follow-up periods compared to our serological approach. Additional adequately powered studies are needed to acquire a comprehensive view of Mg infections and their clinical implications. Given their shared STI pathogenesis, it is likely that the factors driving Mg transmission and serological persistence parallel those identified for Ct.
Ct infections are extensively studied due to their serious reproductive health consequences when untreated, including increased risks of adverse pregnancy and fertility outcomes [Reference Bébéar and de Barbeyrac2,Reference Torrone26]. Recently, it was shown that Mg seropositivity was also more common among women of infertile couples (5.4%) than fertile couples (1.6%) with an OR of 3.5 (95%CI 1.10–10.75) [Reference Idahl12]. However, no such association was observed in male partners [Reference Idahl12]. In our study, all women were pregnant at baseline, and only one fertility-associated co-factor could be identified; a higher parity was a protective factor against Ct. This likely reflects more stable relationships and reduced exposure to Ct due to fewer lifetime sexual partners and lower engagement in high-risk behaviors. Notably, among women with 10 or more lifetime sexual partners, having a parity of three or more was associated with a stronger protective effect, adjusted OR 0.3 (95%CI 0.09–0.89).
It can be speculated that protective factors against Mg seropersistence, including a lower number of sexual partners, could be similar to those of Ct. However, our study did not substantiate this hypothesis. A limitation of this study is the relatively small cohort size, which restricted our ability to explore the causes of seroconversion or decay in greater depth. Additionally, reliance on self-reported data for variables such as STI history, partner count, and sexual behaviors may have introduced potential non-differential misclassification, adding a degree of vulnerability to our findings. An important strength of our study is its longitudinal design, with repeated serum sampling throughout the follow-up period, allowing for a detailed analysis of changes over time.
Taken together, persistent Ct antibodies exhibit a strong association with well-defined high-risk sexual practices among women, less significantly in males. It can be concluded that appropriate preventive measures are of vital importance in combating STIs, irrespective of gender. Notably, seropositivity to Mg was predominantly correlated with the early sexual behavior of women but not those of men. Further studies of both genders are warranted to comprehensively assess the factors associated with Mg infection and the development of serological persistence.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S095026882500007X.
Data availability statement
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Author contribution
Data curation: T.W., B.M., J.B., K.L., K.S., S.G., S.S.; Formal analysis: T.W., B.M., J.B., K.S., N.K.; Investigation: T.W., B.M., J.B., K.L., K.S., S.G., S.S., N.K.; Methodology: T.W., B.M., J.B., K.L., S.G., S.S.; Software: T.W., B.M., J.B., K.L., K.S., S.S.; Writing – review & editing: T.W., B.M., J.B., K.L., K.S., S.G., S.S., N.K.; Conceptualization: K.L., S.G., S.S.; Funding acquisition: K.L., S.S.; Project administration: K.L., S.S.; Resources: K.L., S.S.; Supervision: K.L., S.S.; Validation: K.L., K.S., S.S.; Visualization: K.L., S.S., N.K.; Writing – original draft: N.K.
Financial support
Since its onset in 1998, the FFHPV study has been financially supported by the Academy of Finland [SS, KL]; Päivikki and Sakari Sohlberg Foundation [SS, SG, KL]; Sigrid Jusélius Foundation [KL]; and the Finnish Medical Foundation [KL]. The authors declare no conflict of interest.
Competing interest
The authors declare no conflict of interest.
Open access
