Downloaded from http://sti.bmj.com/ on April 19, 2015 - Published by group.bmj.com
STI Online First, published on April 17, 2015 as 10.1136/sextrans-2014-051839 Basic science
Kinetics of circulating antibody response to Trichomonas vaginalis: clinical and diagnostic implications Phuong Anh Ton Nu,1 Paola Rappelli,2 Daniele Dessì,2 Vu Quoc Huy Nguyen,3 Pier Luigi Fiori2 1
Department of Parasitology, Huè University of Medicine and Pharmacy, Huè City, Vietnam 2 Department of Biomedical Sciences, University of Sassari, Sassari, Italy 3 Department of Obstetrics and Gynecology, Huè University of Medicine and Pharmacy, Huè City, Vietnam Correspondence to Dr Pier Luigi Fiori, Department of Biomedical Sciences, University of Sassari, Viale S. Pietro 43B, Sassari 07100, Italy; ﬁ[email protected]
Received 3 September 2014 Revised 1 March 2015 Accepted 23 March 2015
ABSTRACT Objectives Persistence of antibodies against pathogens after antimicrobial treatment is a marker of therapy failure or evolution to a chronic infection. The kinetics of antibody production decrease following antigen elimination is highly variable, and predicting the duration of soluble immunity in infectious diseases is often impossible. This hampers the development and use of immunoassays for diagnostic and seroepidemiological purposes. In the case of Trichomonas vaginalis infection, the kinetics of antibody levels decrease following therapy has never been studied. We thus investigated the clearance of circulating anti-T. vaginalis IgGs after pharmacological treatment in patients affected by trichomoniasis. Methods 18 female patients affected by acute trichomoniasis were enrolled in this study. After metronidazole therapy administration, subjects were followed up monthly up to 5 months, and serum levels of anti-T. vaginalis IgGs were measured by ELISA. Results We showed that a successful therapy is characterised by a relatively fast decline of speciﬁc antibodies, until turning into negative by ELISA in 1–3 months. In a few patients we observed that the persistence of anti-T. vaginalis antibodies was associated with an evolution to chronic infection, which may be due to treatment failure or to reinfection by untreated sexual partners. Conclusions Our results describe the direct correlation between the decline of a speciﬁc humoral antiT. vaginalis response and an effective antimicrobial therapy. These ﬁndings may facilitate the follow-up approach to circumvent limitations in developing new diagnostic tools and techniques routinely used in microbiology laboratories to assess the presence of T. vaginalis in clinical samples.
To cite: Nu PAT, Rappelli P, Dessì D, et al. Sex Transm Infect Published Online First: [please include Day Month Year] doi:10.1136/sextrans2014-051839
Trichomonas vaginalis is the causative agent of trichomoniasis, the most common non-viral sexually transmitted disease in humans, causing vaginitis in women and urethritis and prostatitis in men. More than 75% of trichomoniasis in men and 50% in women are asymptomatic, generating a subclinical infection accompanied by chronic local inﬂammation, which can last for years.1 Moreover, T. vaginalis is able to establish in vivo a symbiotic relationship with Mycoplasma hominis, which contributes to upregulate the protozoan-induced inﬂammation,2 thus potentially affecting disease severity. Acute infection
is associated with HIV transmission,3 infertility and pregnancy and postpartum complications.4 Recent studies have positively associated trichomoniasis with aggressive cervical5 and prostate cancers.6 7 Routine diagnosis of trichomoniasis is based on direct microscopic examination of wet mount preparations and on culture-based systems, but both are limited by a low sensitivity (from 38% to 82%), being strongly dependent on the number and on the viability of protozoa.8 Molecular techniques have been proposed in recent years,8 9 but in many countries they have not been included yet in the routine diagnosis. Indirect immunological techniques represent an effective alternative to direct identiﬁcation. The clinical importance of serological conﬁrmation of transmissible diseases is widely accepted, especially during chronic infections: in these cases, the parasite load can be below the detection limits assessed by common identiﬁcation methods. In addition, detection of circulating-speciﬁc immunoglobulins is crucial in seroepidemiological studies. When infection is cleared by the combined actions of the host defence mechanisms and a successful antimicrobial treatment, the antigenic stimulation ceases, and the peak of antibody production declines. In a different scenario, an acute infection may evolve to a chronic disease, with B lymphocytes constantly stimulated to secrete antibodies by the persisting antigen exposure. One of the major limitations of indirect diagnosis is the difﬁculty in discriminating between previous and current infections. Persistence of speciﬁc IgG in patient sera is a feature that varies from pathogen to pathogen. The lack of understanding about timing and kinetic mechanisms of circulating IgG decrease following microorganisms eradication represents for a number of transmissible diseases, a major issue related to antibody detection for diagnostic or epidemiological purposes. In the case of T. vaginalis infections, the antibody titre decrease following therapy has never been investigated. Moreover, no test to assess therapy outcomes in individual patients complaining of aspeciﬁc symptoms following metronidazole treatment is currently available. In order to study the kinetics of circulating IgG after metronidazole treatment, a group of patients affected by trichomoniasis was tested before and after therapy, until disappearance of anti-T. vaginalis antibodies.
Nu PAT, et al. Sex Transm Infect 2015;0:1–3. doi:10.1136/sextrans-2014-051839
Copyright Article author (or their employer) 2015. Produced by BMJ Publishing Group Ltd under licence.
Downloaded from http://sti.bmj.com/ on April 19, 2015 - Published by group.bmj.com
Basic science METHODS Patients Patients with trichomoniasis were selected among a group of 249 women with symptoms of vaginitis (age ranging from 20 to 60 years, mean 38±10) attending the Gynecological Clinic of Huè University Hospital (Vietnam) between September 2010 and June 2012. Informed consent was obtained from all subjects and samples were processed anonymously. The collection of samples was approved by the Bioethics Committee of Huè University of Medicine and Pharmacy, Vietnam (21 October 2010). After clinical examination, diagnosis of trichomoniasis was made by direct microscopic examination of wet mount preparations and conﬁrmed by samples cultivation for up to 7 days in trypticase-yeast extract-maltose (TYM) liquid medium. A total of 18 women testing positive to T. vaginalis infection were selected; none of them reported previous trichomonad infection. All positive patients were treated with 500 mg metronidazole per os, twice a day for 7 days. All patients were clinically re-evaluated 4 months after therapy administration to assess the persistence of vaginitis symptoms. During the ﬁrst and the last visit, vaginal samples were collected to verify the protozoan presence using multiplex PCR, according to the protocol previously described.10
Serodiagnostic follow-up A sample of blood was collected from each T. vaginalis-positive patient before starting pharmacological treatment and then monthly for 5 months after therapy. Sera were stored at −80°C until use. All sera from patients with trichomoniasis were tested to detect anti-T. vaginalis IgG by ELISA following the protocol previously described,11 with some modiﬁcations. Brieﬂy, 50 000 trichomonad cells of the reference strain G3 were added to each well of a microtiter plate, air-dried and then ﬁxed with ice cold 95% ethanol. Sera, diluted 1:100 in phosphate buffered saline plus 5% bovine serum albumin, were added to each well and incubated at 37°C for 1 h. Alkaline phosphatase-labeled antihuman IgG was used as secondary antibody. Plates were read at 405 nm with a Multiskan FC plate reader (Termo Scientiﬁc). All sera were tested in duplicate. Sera from a group of 20 healthy Vietnamese children (age ranging from 2 to 10 years) were used as negative controls.
Cut-off value deﬁnition To deﬁne the seropositivity against T. vaginalis a cut-off value was determined on a group of 56 sera from Vietnamese women with no history of trichomoniasis, matching with the positive group for age and geographic origin. An additional group of 53 sera from Italian women, that tested negative in ELISA in a previous work (7), was also used. All control sera were tested in ELISA as described, and the SD was calculated on optical density (OD) values obtained: under our experimental conditions it accounted for 9.3% of OD mean (m=0.098). We considered as negative all sera with ODsm+3OD (+27.9%). Samples with OD ranging between m+2SD and m+3SD were classiﬁed as borderline. ELISA tests were performed every month, to follow-up the seropositivity of patients. Thus, in order to avoid variability between different experiments, we decided to express results as sample/control ratio. Control was calculated as the mean value of negative sera from 20 Vietnamese children, which were tested in each experiment. An OD ratio >1.28 was considered positive (>m+3OD). 2
RESULTS Diagnosis of trichomoniasis was made on 18 of 249 symptomatic Vietnamese women and metronidazole therapy was administered. Results of the serological follow-up are reported in table 1: in 14 of 18 positive patients the titre of circulating anti-T. vaginalis antibodies progressively declined over time. All these 14 patients became asymptomatic within 1–2 months after therapy, and eradication of infection was conﬁrmed by PCR on vaginal samples, 5 months after treatment. On the contrary, patients #2, #3, #7 and #8 tested positive in ELISA for 4–5 months (table 1). Three of them (#2, 3, and 8) reported persistent vaginal symptoms several weeks after metronidazole administration. Serological data are consistent with a chronic antigenic stimulation and may be explained either by a failure of the drug treatment at clearing the infection or by recurrent reinfections by sexual partners who were not treated. In particular, patient #2 was likely infected with a metronidazole-resistant isolate. Indeed, during the clinical re-evaluation 4 months after the metronidazole therapy, while reporting no sexual activity, the patient complained of the persistence of symptoms and was therefore treated with tinidazole (1 g/day for 7 days per os). This resulted in a subsequent signiﬁcant drop in antibody titre (OD ratio: from 3.59 to 1.27) after 1 month, while vaginal sample tested PCR negative at the ﬁfth month, demonstrating the clearance of protozoa. Similarly, patient #7 was likely infected by a metronidazoleresistant isolate since, despite therapy, protozoa were detected by PCR after 5 months, and the antitrichomonad IgG titre was constantly positive during the follow-up, even if the patient reported no sexual activity. In contrast with subject #2, the patient #7 was asymptomatic at clinical re-evaluation 4 months after therapy, and therefore no alternative therapy was
Serological follow-up of patients with trichomoniasis OD ratio (sample/control) Months
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
2.01 1.89 2.64 1.54 2.90 2.29 2.16 2.23 1.34 1.38 1.35 2.43 1.37 1.70 1.97 1.35 1.51 1.63
(+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+) (+)
2.17 1.98 4.57 1.54 4.28 1.21 1.92 1.89 1.30 1.41 1.39 2.04 1.12 0.96 1.75 1.26 ND 1.13
1.63 2.83 3.35 0.90 ND 1.25 1.87 1.73 1.36 ND ND 1.01 ND 0.86 ND ND ND 0.98
1.60 1.94 6.95 0.97 1.33 1.18 1.80 1.76 1.37 0.99 0.98 0.71 0.95 ND ND ND 1.09 0.95
0.95 3.59 2.55 0.95 1.27 1.12 1.73 1.82 0.97 ND ND ND ND ND 0.90 1.11 0.99 ND
1.03 (−) 1.27 (−) 2.94 (+) 0.97 (−) 1.09 (−) 0.84 (−) 1.73 (+) 0.88 (−) ND 0.90 (−) ND ND ND 0.98 (−) 0.84 (−) 0.94 (−) 0.94 (−) ND
Eighteen positive women were followed up for 5 months after diagnosis and anti-trichomonas therapy administration; sera were harvested monthly and tested in ELISA to detect anti-Trichomonas vaginalis IgG class antibodies. Results are shown as an OD ratio of samples (mean values of duplicates) and controls (mean value of 20 children sera). >1.28 OD: positive (in bold), 1.186 borderline and