Proposed panel of diagnostic tools for accurate

Journal of Immunological Methods xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Journal of Immunological Methods journal homepage: www.elsevier.com/locate/jim

Research paper

Proposed panel of diagnostic tools for accurate temporal classification of symptomatic T. gondii infection Geisa Baptista Barrosa, Elenice Moreira Lemosb,1, Priscila Pinto e Silva-dos-Santosb, Reynaldo Dietzeb, Eliana Zandonadec, José Roberto Mineod, Deise Aparecida de Oliveira Silvad, Ana Cláudia Marquez Pajuabad, Matheus de Souza Gomese, Laurence Rodrigues do Amarale, Jordana Grazziela Coelho-dos-Reisf,⁎, Olindo Assis Martins-Filhof,2, José Carlos Serufog,2 a

Departamento de Pediatria, Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil Núcleo de Doenças Infecciosas, Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil Departamento de Estatística, Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil d Laboratório de Imunoparasitologia, Universidade Federal de Uberlândia, Uberlândia, Minas Gerais, Brazil e Laboratório de Bioinformática e Análise Molecular, Universidade Federal de Uberlândia, Campus Patos de Minas, Patos de Minas, MG, Brazil f Grupo Integrado de Pesquisas em Biomarcadores, Centro de Pesquisas René Rachou, FIOCRUZ, Belo Horizonte, Minas Gerais, Brazil g Departamento de Clínica Médica, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil b c

A B S T R A C T Serological tests available for the diagnosis of acute Toxoplasma gondii infection have limitations in establishing the temporal diagnosis of acute toxoplasmosis. The present analytical-descriptive investigation comprises of a prospective longitudinal cohort study to search for accurate biomarkers to distinguish acute, early and late convalescent T. gondii infection. Classic methods (immunofluorescence-IFA along with Enzyme-linked immunosorbent-ELISA and fluorescent-ELFA assays) for IgM, IgA, IgG and IgG avidity were employed in parallel with flow cytometry-based anti-fixed T. gondii tachyzoites serology (FC-AFTA-IgM, IgG, IgG avidity and IgG subclasses). The results reemphasized the limitations of IgM & IgG IFA, IgG ELFA, IgG & IgG subclasses FC as well as IgA ELISA biomarkers for the temporal diagnosis of acute toxoplasmosis. Receiver Operating-characteristics features (ROC-curves) were employed to adjust conventional cut-offs aiming at establishing a novel protocol to discriminate more accurately the different phases of toxoplasmosis. Conversely, IgM presented high diagnostic co-positivity for acute toxoplasmosis (97% for ELISA, 96% for ELFA and 95% for FC-AFTA) along with moderate co-negativity for detection of late convalescent toxoplasmosis (82%, 76% and 79%, respectively). IgG avidity (ELFA and FC-AFTA) outstand with the highest performance indices with 91% and 96% co-negativity for assessing acute toxoplasmosis and 91% and 98% co-positivity for late convalescent toxoplasmosis, respectively. Multivariate analysis generated a three-step algorithm comprising IgM ELFA screening followed by ELFA and FCAFTA IgG avidity with high accuracy in discriminating acute from late convalescent infection. Together, these findings demonstrate the applicability of the proposed panel of diagnostic tools for accurate temporal classification of T. gondii infection.

1. Introduction Approximately one-third of the world's population is infected with Toxoplasma gondii, an intracellular protozoan belonging to the Apicomplexa phylum, subclass Coccidia (Robert-Gangneux and Dardé, 2012). Infection can be acquired through the ingestion of raw or

underdone meat containing tissue cysts or food and water contaminated by oocysts. The disease can still be transplacentally transmitted if the mother acquires the primary infection during pregnancy. Other less common forms of transmission are transfusion of blood products and organ transplants (Robert-Gangneux and Dardé, 2012). Acute infection is asymptomatic/oligosymptomatic in about 80% of

⁎ Corresponding author at: Grupo Integrado de Pesquisas em Biomarcadores, Centro de Pesquisas René Rachou, Fundação Oswaldo Cruz, Avenida Augusto de Lima, 1715 Barro Preto, Belo Horizonte, Minas Gerais 30190-002, Brazil. E-mail address: jordana.reis@cpqrr.fiocruz.br (J.G. Coelho-dos-Reis). 1 In Memoriam: Elenice Moreira Lemos. 2 These authors contributed equally for the supervision of this study.

http://dx.doi.org/10.1016/j.jim.2017.08.010 Received 14 June 2017; Received in revised form 8 August 2017; Accepted 8 August 2017 0022-1759/ © 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Barros, G.B., Journal of Immunological Methods (2017), http://dx.doi.org/10.1016/j.jim.2017.08.010


G.B. Barros et al.

2. Methods

individuals. The clinical picture is non-specific and the most commonly found symptom is isolated adenomegaly. We may also find pictures of malaise accompanied by fever, headache, myalgia, sore throat and symptoms related to the lung, liver, heart, brain and eyes (RobertGangneux and Dardé, 2012). The clinical manifestation is usually benign. However, in immunocompromised patients, there may be recrudescence of the infection, particularly in those with acquired immunodeficiency syndrome (AIDS) and individuals undergoing immunosuppressive drugs (Ajzenberg et al., 2009). Considering the possibility of absence of symptoms during acute infection, the diagnosis of primary infection is based on serological reactions (Robert-Gangneux and Dardé, 2012). The relevance of diagnosing acute infection occurs mainly in two clinical situations: i) when the patient is symptomatic and there is a need for diagnostic confirmation and ii) in pregnant women, to whom serology will generally be the only source of information for decision making during clinical practice (Desmonts and Couvreur, 1974; Bobić et al., 1991; Liesenfeld et al., 2001; Sroka et al., 2016). In these cases, assessment of time of acquisition of primary infection is important, because fetal infection occurs almost exclusively after postconceptional infection (Desmonts and Couvreur, 1974). The laboratory diagnostic is usually performed through the quantitative evaluation of IgG and IgM. A positive IgG test establishes that the individual had prior contact with T. gondii. The detection of specific IgM antibodies, on the other hand, has been the most frequently used serological marker for the diagnosis of acute or recent infection. However, the high percentage of false-positive IgM results and the persistence of IgM for prolonged periods, sometimes for several years (Bobić et al., 1991), support the concept that the isolated presence of anti-T. gondii IgM antibodies does not confirm the diagnosis of acute infection. Supplementary methods, such as the detection of specific IgA and IgE antibodies, also markers of acute infection, present limitations related to sensitivity and disappearance in the serum at variable intervals (4–8 months). In about 25% of individuals, IgA can persist for up to 12 months and both IgA and IgE diagnostic tests are not widely available. Currently, IgG avidity is the most useful auxiliary or confirmatory test in positive IgM cases, determining with some degree of safety, in cases of high avidity, that the first infection occurred for at least 3–4 months. One of the most useful utilities of measuring the IgG avidity is based on the hypothesis that high-avidity results rule out the risk of fetal transmission during pregnancy. Sroka et al. (2016) have compared the efficiency of commercial kits for ELFA IgM, IgG and avidity IgG in determining the phase of T. gondii infection during pregnancy. These authors indicated the high usefulness of the commercial ELFA in serodiagnostics of toxoplasmosis in pregnant women. However, special attention should be paid to parallel detection of IgM antibodies and the IgG avidity, since the IgG avidity test has two important limitations: the late maturation phenomenon, in which 25% of the 12-month infection cases show low avidity IgG and the large percentage of cases (26.4%) with doubtful results (Liesenfeld et al., 2001). Aiming at overcoming the remaining challenges still observed in determining the phase of T. gondii infection, the present investigation was designed to search for accurate biomarkers to distinguish acute, early and late convalescent T. gondii infection. Classic methods (immunofluorescence-IFA along with Enzyme-linked immunosorbentELISA and fluorescent-ELFA assays) for IgM, IgA, IgG and IgG avidity were employed in parallel with flow cytometry-based anti-fixed T. gondii tachyzoites serology (FC-AFTA-IgM, IgG, IgG avidity and IgG subclasses). In this study, we describe a longitudinal study that shows the dynamics of anti-T. gondii antibodies and explores the diagnostic performances of the combination of conventional serological tests with adjusted cut-off points along with flow cytometry in assessing time of primary toxoplasmosis infection. Using a multivariate analysis approach, we have generated a three-step algorithm comprising IgM ELFA screening followed by ELFA and FC-AFTA IgG avidity with high accuracy to identify acute from late convalescent T. gondii infection.

2.1. Study population and blood sampling This longitudinal, prospective, analytical-descriptive cohort study was conducted at the Infectious Diseases Nuclei at Universidade Federal do Espírito Santo (NDI/UFES) in collaboration with the Immunology Laboratory at Universidade Federal de Uberlândia (UFU) e o Grupo Integrado de Pesquisas em Biomarcadores at Centro de Pesquisas René Rachou (CPqRR/FIOCRUZ). The enrollment period for this study was 25 months (June 2005 to June 2007). Volunteers with clinical suspect of acute toxoplasmosis were screened by infectious diseases Specialist at private or public outpatient clinics from the Metropolitan region of Vitória (Espírito Santo State, Brazil). After screening, the patients received medical care from one of us (GBB), for reassessment based on the inclusion and exclusion criteria. The inclusion criteria were: patients of both genders; age range between 7 and 80 years; presence of at least two of the following signs or symptoms compatible with acute T. gondii infection, including malaise, adynamia, presence of lymph glands, headache, fever and arthralgia; patients who agreed with the study protocol and signed the informed consent form. The exclusion criteria were: presence of HIV infection; pregnancy; impossibility to comply with the protocol. The prospective study enrolled 31 patients with symptomatic T. gondii infection using the date of symptoms onset as day zero (baseline) and all patients were followed for one year. Sequential blood sampling (10 mL) was collected without anticoagulant from all participants at 1, 2, 3, 4, 6, 8, 10 and 12 months after baseline (n = 248 paired samples, 8 time-points from 31 patients). Subsequently, the patients were categorized into 3 groups based on the reactivity board of serologic tests for timeline clustering and the subgroups referred as acute (1, 2 and 3 months, n = 93 paired samples from 31 patients), early (4, 6 and 8 months, n = 93 paired samples from 31 patients) and late convalescent (10 and 12 months, n = 62 paired samples from 31 patients). If the patient was considered eligible, a free written informed consent form was signed prior to the first blood sampling. This work complied with resolution number 196/1996 from the National Health Council for research involving humans and was approved by the Ethical Committee at Universidade Federal do Espírito Santo, Vitória, Espírito Santo, Brazil during the ordinary meeting of March 30th 2005. 2.2. Classic serological methods for T. gondii infection diagnosis 2.2.1. Detection of anti-T. gondii IgM, IgG and IgG avidity by Enzymelinked fluorescent assay (ELFA) The anti-T. gondii IgM reactivity was determined by the VIDAS® TOXO IgM II test (bioMérieux, Lyon, France), a fully automated quantitative analysis, employing a immunocapture method and fluorescence detection (ELFA), according to the manufacturer's instructions. The results lower than 0.55 IU/mL were considered negative, while those higher than 0.65 IU/mL were considered positive. Adjustment of cut-off point was further defined by ROC curve analysis to interpret IgM ELFA as described in the Results section. The VIDAS® TOXO IgG II test (bioMérieux, Lyon, France) was employed to measure the anti-T. gondii IgG reactivity, using a sandwich immunoenzymatic method in two stages with final fluorescence detection (ELFA), according to the manufacturer's instructions. Results lower than 4.0 IU/mL were considered negative and those higher or equal than 8.0 IU/mL were considered positive. The IgG avidity test was used by two-step enzyme immunoassay sandwich immunoassay with fluorescent final detection system (ELFA VIDAS®, bioMérieux, Lyon, France), according to the manufacturer's instructions. The results lower than 0.2 were considered low avidity and those higher or equal 0.3 were considered high avidity. Adjustment 2


G.B. Barros et al.

azino-bis-3-ethylbenzothiazoline sulphonic acid (ABTS) was added to each well at 0.01 M in 0.07 M citrate-phosphate buffer (pH 4.2). Optical density (OD) was assessed at 405 nm in a plate reader. Positive and negative controls were included on each experimental batch. The results were expressed as ELISA reactivity indices (EI), as follows: EI = ODsample/ODcut-off, where EI values higher than 1.2 was considered positive.

of cut-off point was further defined by ROC curve analysis to interpret ELFA IgG avidity as described in the Results section. 2.2.2. Indirect immunofluorescence assay (IFA) for anti-T. gondii IgM and IgG The immunofluorescence assay for anti-T. gondii IgM and IgG was carried out as described previously by Mineo et al. (1980). Briefly, slides were prepared with pre-fixed T. gondii tachyzoites and stored at − 20 °C for up to three months. Serial dilutions of test samples were used for IgM (1:16 to 1/4096) and IgG (1/64 to 1/65,536) detection, together with positive and negative controls. Aliquots of 50 μL of prediluted test sample were incubated for 30 min at 37 °C in a humidified chamber. After incubation the slides were washed once with PBS, dried at room temperature and re-incubated under the same conditions with anti-human IgG (Fc specific) or anti-human IgM (μchain specific) conjugated with fluorescein isothiocyanate (FITC) (Sigma, St Louis, MO, USA). After one washing step with PBS, two independent technicians performed the independent reading of antibody reactivity using the immunofluorescence microscope (Olympus CBA2). The serum titer was determined as the last dilution, where 50% of the parasites showed fluorescence. Titers of IgM below 1:64 and IgG below 1:128 were considered negative.

2.3. Non-conventional serological methods for T. gondii infection diagnosis 2.3.1. Flow cytometry-based anti-fixed T. gondii tachyzoites serology (FCAFTA) for detection of IgM, IgG, IgG subclasses Flow cytometric detection of anti-fixed T. gondii tachyzoites IgM and IgG were performed according to the protocol described by Silva-dosSantos et al. (2012). Briefly, in 96-well U-bottom microplates (Nunc, Denmark), aliquots of 50 μL of the pre-diluted serum samples (1:16,000 for IgM and 1:512,000 for IgG) were incubated with 50 μL of the prefixed T. gondii tachyzoites (RH strain) suspension (5.0 × 106/mL) and incubated at 37 °C for 30 min. Following incubation, the parasites were washed twice with 200 μL PBS supplemented with 3% of Fetal Bovina Serum (PBS-3%FBS), by centrifugation at 1000g, 4 °C for 10 min. Subsequently, the parasites were incubated with 50 μL of anti-human IgM or anti-human IgG labeled with FITC, pre-diluted in PBS-3%SFB at 37 °C for 30 min (1:2000 for IgM and 1:20,000 for IgG). For the detection of anti-T. gondii IgG subclass (IgG1, IgG2, IgG3 and IgG4), following incubation with pre-diluted serum samples (1:2,048,000 for IgG1; 1:4000 for IgG2; 1:4000 for IgG3 and 1:500 for IgG4), the parasites were washed and re-incubated with 50 μL of biotin-labeled anti-human IgG subclasses pre-diluted in PBS-3%FBS (1:20,000 for IgG1; 1:2000 for IgG2; 1:1000 for IgG3 and 1:2000 for IgG4), followed by two washes with PBS-3%FBS. The parasites were then incubated at 37 °C for 30 min in the presence of 20 μL of streptoavidin conjugated with phycoerythrin (SAPE), pre-diluted 1:400 in PBS-3%FBS. After incubation, the parasites were washed twice and the supernatant discarded. The stained parasites were then fixed with 200 μL of FACS fix solution (10 g/L of paraformaldehyde, 10.6 g/L sodium cacodilate, 6.63 g/L sodium chloride, pH 7.2) and kept at 4 °C up to 24 h prior flow cytometric acquisition. For each experimental batch, an internal control was performed to monitor nonspecific binding of the secondary antibody, where the parasites were incubated in the absence of human serum, however in the presence of the secondary antibody. In all tests samples of positive and negative sera for toxoplasmosis were included. A total of 5000 events were acquired for each tested sample. The FITC or SAPE relative fluorescence intensity of labeled parasites was evaluated on single histograms and data expressed as the percentage of positive fluorescent parasites (PPFP). The threshold for unspecific binding was established at PPFP < 2% according to the internal control and maintained for each experimental batch to obtain the PPFP values for individual samples. The adjusted cut-off to interpret IgM detection by flow cytometry was further defined by ROC curve analysis and employed for data analysis as provided in Results section.

2.2.3. In house capture Enzyme-linked immunosorbent assay (ELISA) for anti-T. gondii IgM and IgA Capture ELISA was performed as described previously by Mineo et al. (1986), modified as follows: briefly, 96-well polystyrene microplates (Immulon 2, Dynex Technologies, Chantilly, VA, USA) were covered with anti-human IgM or anti-human IgA (Sigma, St Louis, Mo, USA) at 10 μg/mL in carbonate-bicarbonate buffer (0.6 M, pH 9.6) for 18 h at 4 °C. After incubation, plates were washed three times with phosphate buffered saline supplemented with 0.05% Tween 20 (PBS-T) and blocked with PBS-T supplemented with 5% skim powder milk (PBSTM) for 1 h at room temperature. After three washing steps, the plates were incubated with testing samples pre-diluted in PBS-TM (1:16) for 2 h at 37 °C. After additional washing steps, the plates were incubated with soluble T. gondii tachyzoite antigen (STAg) from RH strain, prepared according to Benevides et al. (2013), at 100 μg/mL in PBS-TM for 2 h at 37 °C. Bound STAg was detected by rabbit anti-T. gondii F(ab’)2 fragment antibody conjugated with peroxidase, prepared according to Wilson and Nakane (1976), diluted 1:50 in PBS-TM. Following incubation for 1 h at 37 °C, plates were washed six times with PBS-T and the enzymatic substrate 0.03% H2O2 and 2,2′-azino-bis-3-ethylbenzothiazoline sulphonic acid (ABTS) was added to each well at 0.01 M in 0.07 M citrate-phosphate buffer (pH 4.2). Optical density (OD) was determined at 405 nm in plate reader. Positive and negative controls were included on each experimental batch. The cut-offs were determined by the mean OD of the negative controls plus 3-fold standard deviations. The results were expressed as ELISA reactivity indices (EI), as follows: EI = ODsample/ODcut-off, where EI values higher than 1.2 was considered positive. Adjustment of cut-off point was further defined by ROC curve analysis to interpret IgM ELISA as described in the Results section.

2.3.2. Flow cytometry-based anti-fixed T. gondii tachyzoites serology (FCAFTA) for assessing IgG avidity The IgG avidity analysis by flow cytometry was performed as described previously by Silva-dos-Santos et al. (2012). Two parallel batches of tests were performed simultaneously, one batch performed in the absence of urea (U −) and the other batch carried out in the presence of urea (U +). Briefly, in 96-well round bottom polystyrene plates, 50 μL of pre-diluted (1:32,000, to1:2,048000) were incubated with 50 μL of the pre-fixed T. gondii tachyzoites (RH strain) suspension (5.0 × 106/mL) at 37 °C for 30 min. After incubation, in the U − and U + protocols, the parasites were washed once with 200 μL PBS-3%FBS by centrifugation at 1000× g, 4 °C for 10 min. In the U − batch, the parasites were re-incubated with 200 μL of PBS-3%FBS for 5 min

2.2.4. Indirect ELISA for anti-T. gondii IgG Indirect ELISA was performed as described previously by Mineo et al. (1980), modified as follows: High affinity 96 well polystyrene microplates (Corning cost 3590, Sigma) were sensitized with STAg (10 μg/mL) diluted in 0.06 M carbonate-bicarbonate buffer (pH 9.6) for 18 h at 4 °C. After incubation, the plates were washed three times with PBS-T and blocked with PBS-TM for 1 h at room temperature. After three steps with PBS-T, the plates were incubated with pre-diluted test samples (1:64) in PBS-TM for 1 h at 37 °C. After six washing steps, the plates were incubated with goat anti-human IgG antibody conjugated with peroxidase (1:2000) for 1 h at 37 °C. After a second round of six washes with PBS-T, the enzymatic substrate 0.03% H2O2 and 2,2′3


G.B. Barros et al.

whereas in the U+ batch the parasites were incubated with 200 μL of 6 M urea solution in PBS-3%FBS for 5 min. Following incubation, parasites from both batches were washed once with 200 μL PBS-3%FBS and reincubated at 37 °C for 30 min in the presence of 50 μL of prediluted anti-human IgG (1/20,000) conjugated with FITC. The stained parasites were washed twice with 200 μL PBS-3%FBS and fixed with 200 μL FACS fix solution. Parasites were maintained up to 24 h at 4 °C before flow cytometric acquisition. The results were expressed as avidity index (AI) that were evaluated at the last serum dilution displaying PPFP higher than 30% in the U − test, and calculated as the ratio between the PPFP value obtained from U+ test by the paired PPFP value yielded from the U − test as follows: AI = PPFP(U +)/PPFP (U −) × 100. The cut-off to interpret IgG avidity by flow cytometry followed the criteria proposed originally by Silva-dos-Santos et al. (2012) and further confirmed by ROC curve analysis and employed for data analysis as provided in Results section.

detected by ELFA and FC was evident, with significant differences amongst 1, 2 and 3 months as compared to that observed at 6, 8, 10 and 12 months after baseline (Fig. 1). The results reemphasized the limitations of IgM & IgG IFA, IgG ELFA, IgG & IgG subclasses FC as well as IgA ELISA biomarkers for the temporal diagnosis of acute toxoplasmosis. 3.2. Panoramic overview of serologic tests for temporal diagnosis throughout T. Gondii infection Aiming at identifying appropriate cut-offs to discriminate more accurately the different phases of toxoplasmosis, Receiver Operatingcharacteristics features (ROC-curves) were assessed for the pre-selected biomarkers (IgM ELFA, IgM ELISA, IgM FC, IgG avidity ELFA and IgG avidity FC) to adjust the conventional thresholds for establishing innovative diagnostic tools (Fig. 2). The results indicated that further adjustments were necessary to increase accuracy of segregation between acute (1 month) and convalescent (12 months) T. gondii infection for all tests, except for IgG avidity FC (Fig. 2 – inserted table). Specifically, the conventional criteria proposed for IgM ELFA (negative < 0.55 and positive > 0.65 IU/mL) shifted up to 2.4 IU/mL; the conventional cut-off proposed for IgM ELISA (negative for Index ≤ 1.0 and positive for Index > 1.2) was set as negative for Index < 2.5 and positive for Index ≥ 2.5; the IgM FC threshold changed from PPFP ≤ 15% to PPFP ≤ 32% and IgG avidity ELFA limit, originally defined as low (< 0.2) and high (≥ 0.3) shifted to 0.15 (Fig. 2 – inserted table).

2.4. Statistical analysis The descriptive analysis for each serological parameter was performed using the SPSS software version 15.0 (SPSS, Inc., Chicago, IL, USA). The ANOVA test was performed to evaluate the association between the results of repeated measurements through the Statistica program, version 7.0 (Statsoft Inc., Tulsa, OK, USA), considering the time (1, 2, 3, 4, 6, 8, 10 and 12 months) as a dependent variable. The means were tested for all times and also for the times individually by the Tukey test. In all cases, significant differences were considered significant when p < 0.05. The performance of selected serological tests applied to temporal diagnosis T. gondii infection were carried out using two approaches, including: Receiver-operating characteristic curve (ROC curve), the combination of two tests and discriminant analysis. The ROC curves were constructed using the MedCalc Statistical Software, version 7.0 (Ostend, Belgium) and employed to define adjusted cut-offs for temporal segregation of T. gondii infection and assess the overall accuracy of a given test. For all methods, two cut-off points were evaluated, referred as conventional criteria and adjusted cut-off points (Fig. 2 - inserted table). Discriminant analysis was carried out using the SPSS program, version 15.0 (SPSS, Inc., Chicago, IL, USA). The “ENTER” and “STEP WISE” methods were used to identify the serological test with the best discriminant power. Cross-validation approach was employed to evaluate the correctness percentages as proposed by Johnson and Wichern (1992).

3.3. Reactivity board of serologic tests for timeline clustering of acute, early and late convalescent T. Gondii infection Although the pre-selected biomarkers (IgM ELFA, IgM ELISA, IgM FC, IgG avidity ELFA and IgG avidity FC) displayed a promising profile for temporal diagnosis of T. gondii infection, the overall analysis of these parameters could not be applied for the laboratorial diagnosis at specific time-points after symptoms onset. In fact, the pre-selected biomarkers were efficient to cluster significant differences only amongst early and late time-points after symptoms onset as shown in Fig. 1. In order to solve this limitations but still providing a reliable approach for the temporal diagnosis of T. gondii infection, a reactivity board of the pre-selected serologic tests was assembled to define the clusters of time-points presenting similar antibody reactivity. Using this approach, three sets of time-points were identified to categorize the T. gondii infection into distinct phases, further referred as acute (1, 2 and 3 months), early (4, 6 and 8 months) and late convalescent (10 and 12 months). Data analysis demonstrated that the acute subgroup presented homogeneous serological pattern characterized by positive IgM (ELFA, ELISA and FC) along with low IgG avidity (ELFA and FC), while late convalescent infection rendered an opposite profile comprising of negative IgM (ELFA, ELISA and FC) together with high IgG avidity (ELFA and FC) as presented in Fig. 3. A transitional profile was observed for early convalescent T. gondii infection.

3. Results 3.1. Panoramic overview of serologic tests for temporal diagnosis of T. Gondii infection The Fig. 1 provides an overview of classic and nonconventional approaches aiming at determining novel biomarkers to the accurate temporal diagnosis of T. gondii infection. This set of analytical-descriptive results was obtained from prospective longitudinal cohort study performed at distinct time-points (1, 2, 3, 4, 6, 8, 10 and 12 months after baseline date of symptoms onset). Classic methods comprised of Enzyme-linked fluorescent-ELFA for IgM, IgG and IgG avidity, Enzyme-linked immunosorbent-ELISA assays for IgM, IgA and IgG along with immunofluorescence-IFA for IgM and IgG analysis. In parallel, non-conventional methods as flow cytometry-based anti-fixed T. gondii tachyzoites serology (FC-AFTA-IgM, IgG and IgG subclasses) along with IgG avidity were also employed. Descriptive data analyses pointed out IgM ELFA, IgM ELISA and IgM FC as promising biomarkers to discriminate clearly the decreasing antiT. gondii antibody reactivity observed at 1, 2 and 3 months after symptoms onset in comparison to that observed at 8, 10 and 12 months after baseline. Conversely, a progressive increase of IgG avidity

3.4. Performance of serologic tests for temporal diagnosis of acute, early and late convalescent T. Gondii infection Aiming at further characterizing the applicability of the pre-selected methods for temporal diagnosis of T. gondii infection (IgM by ELFA, ELISA and FC along with IgG avidity by ELFA and FC), scatter plot distribution of individual samples was assembled for each method and the performance indices co-positivity and co-negativity presented in Fig. 4. Data analysis demonstrated that IgM ELFA presented the most robust performance indices (97%; 46%; 82%) as compared to IgM ELISA (96%; 44%; 76%) and IgM FC (95%; 48%; 79%). In parallel, IgG 4


G.B. Barros et al.

Fig. 1. Panoramic overview of serologic tests for temporal diagnosis of T. gondii infection. Classic and nonconventional approaches were employed to identify novel biomarkers to the accurate temporal diagnosis of T. gondii infection. Analytical-descriptive analysis was carried out at distinct time-points (1, 2, 3, 4, 6, 8, 10 and 12 months after baseline date of symptoms onset) in a prospective longitudinal cohort study. Classic methods comprised of Enzyme-linked fluorescent-ELFA for IgM, IgG and IgG avidity ( ); Enzyme-linked immunosorbent-ELISA assays for IgM, IgA, IgG and IgG avidity ( ) and Immunofluorescence-IFA for IgM and IgG analysis ( ). Non-conventional methods included: flow cytometry-based anti-fixed T. gondii tachyzoites serology (FC-AFTA) for IgM, IgG and IgG subclasses, along with IgG avidity ( ). Data are reported as scatter distribution of individual values of IgM, IgG and IgA obtained by ELFA (IU/mL), ELISA (Index), IFA (Titers), FC (PPFP) along with IgG avidity Index detected by ELFA and FC at distinct time-points after symptoms onset (baseline). Gay background highlighted the promising biomarkers able to discriminate clearly the anti-T. gondii antibody reactivity at 1, 2 and 3 months in comparison to that observed at 8, 10 and 12 months after baseline (IgM ELFA, IgM ELISA and IgM FC along with of IgG avidity detected by ELFA and FC). Statistical differences were assessed by Tukey test using Statistica program, version 7.0 (Statsoft Inc., Tulsa, OK, USA) as described in Methods. In all cases, significance was considered at p < 0.05.

from late convalescent T. gondii infection, with moderate global accuracy (75%; LOOCV = 75%). 3D–scatter plot analysis further illustrated the applicability of the proposed panel of diagnostic tools for an accurate temporal classification of T. gondii infection.

avidity FC (96%; 59%; 98%) displayed higher co-positivity and co-negativity as compared to IgG avidity ELFA (91%; 71%; 91%) (Fig. 4). 3.5. Multivariate analysis of serologic tests applied to the temporal diagnosis of acute, early and late convalescent T. Gondii infection

4. Discussion In order to provide a feasible algorithm applicable to the temporal diagnosis of T. gondii infection, data mining strategies of multiparametric analysis was applied using the results generated by the preselected methods (IgM by ELFA, ELISA and FC along with IgG avidity by ELFA and FC). Multivariate analysis have generated a three-step algorithm comprising IgM ELFA screening followed by ELFA and FC-AFTA IgG avidity with high accuracy in discriminating acute from late convalescent infection (Fig. 5). The results confirmed that IgM ELFA and IgG avidity were the top-two methods leading to elevated performance indices (sensitivity = 92% and specificity = 80%) to distinguish acute

Serological tests are indirect diagnostic methods that evaluate the humoral immunity developed by the host in response to infection by a given microorganism. The evaluation of the specific immune response of immunocompetent individuals in cases of primary acquired T. gondii infection is performed with tests, which detect and/or quantify classes of immunoglobulins, such as IgG and its subclasses, IgM, IgA and IgE, or that evaluate aspects peculiar to the reactivity of these antibodies, such as IgG avidity. The characteristics of the inputs, in particular the antigens, and the methodology used in each diagnostic test are important 5


G.B. Barros et al.

Fig. 2. Cut-off Adjustments for Serologic Tests Applied to the Temporal Diagnosis of T. gondii Infection. Scatter plot distribution of individual values were employed to verify the need of cut-off adjustments of conventional criteria proposed the pre-selected biomarkers, including IgM ELFA and IgG avidity ELFA ( ), IgM ELISA ( ) along with IgM FC and IgG avidity FC ( ). The reactivity profile obtained acute (1 month) and convalescent (12 months) T. gondii infection were employed and the reference standards to define the cut-off adjustments of conventional criteria. The Receiver Operating-characteristics features (ROC-curves) were assessed to adjust the conventional thresholds for establishing innovative diagnostic tools. Performance indices obtained with the adjusted cut-offs, including the global accuracy determined by the area under the curve, sensitivity and specificity are provided within the ROC curves. Conventional criteria and adjusted cut-off are reported in the inserted table.

et al., 2009). The authors considered partial discrepancy when one result of one test was doubtful and the other positive or negative. They concluded that if a test is doubtful, a second test could, if not, eliminate acute infection, avoiding anxiety and unnecessary treatment of pregnant women. In fact, the authors suggest that results of high IgG avidity from either test are exclusion criteria for the diagnosis of acute infection. Cases presenting low or intermediate avidity may require additional tests including an extended panel of antibodies for further clarification and better evaluation of the temporal diagnosis of the disease. Roberts et al. (2001) described a multicenter study to define strategies for serodiagnosis of primary infection with T. gondii that explored the diagnostic performances of 20 antibody assays and their combinations in assessing the time of primary infection. The study included sera samples from three groups of subjects: I- acute infection, < 3 months; II- convalescent infection, 3–12 months and III- past infection, > 12 months. No assay combination was able to distinguish acute from convalescent infection (3–12 months). For diagnostic purposes, the best assay combination found was IgM and IgG avidity. In agreement with these findings, recently, Laboudi and Sadak (2017) tested the performance of IgG avidity test for early diagnosis of during pregnancy. The authors proposed that the avidity test is useful to

for the analysis of the results, since, to a large extent, these elements determine the type of antibody identified. Different methods have been proposed to detect distinct antibodies, deserving analysis in light of their specificities and avidity when applied to the temporal diagnosis of acquired toxoplasmosis. Correlation studies between antibody titers were found by the Sabin & Feldman method as well as indirect immunofluorescence and direct agglutination tests (van Nunen and van der Veen, 1965; Desmonts and Remington, 1980). These three assays use whole parasites, with reading determined in titers (reciprocal of highest dilution with positive result). The qualitative correlations (concordance between positive and negative), most commonly used between tests with different methodologies, are the most frequently found. They suffer less influence from factors such as time, which is directly related to the levels of antibodies, reflecting the evolution of the immune response of the host to the parasitic infection. These tests today are less and less used because they do not allow automation. A recent study compared the results of two commercial IgG avidity tests (Platelia Toxo IgG Avidity, Bio-Rad, France and Liaison Toxo IgG Avidity II Diasorin, Italy), and found 75.3% of total concordance, 23.6% of partial discrepancy and 1.1% of total discordance (Lachaud 6


G.B. Barros et al.

Fig. 3. Reactivity board of serologic tests for timeline clustering of acute, early and late convalescent T. gondii infection. Reactivity board assemblage of the pre-selected serologic tests (IgM ELFA, IgM ELISA, IgM FC along with IgG avidity ELFA and IgG avidity FC) was built aiming at defining clusters of time-points presenting similar antibody reactivity. For this ≤ 2.4 IU/mL; > 2.4 IU/mL), IgM ELISA ( Index ≤ 2.5; purpose, serological reactivity were respectively categorized as negative or positive for IgM ELFA ( Index > 2.5) and IgM FC ( PPFP ≤ 32%; PPFP > 32%) and as low or high for IgG avidity ELFA ( < 0.15; ≥ 0.25) and IgG avidity FC ( ≤ 60; > 60). Three sets of time-points with homogeneous profile were identified, allowing the categorization of T. gondii infection into distinct phases, referred as acute (1–2–3 months = ), early (4–6–8 months = ) and late convalescent (10–12 months = ).

IFI and ISAGA, improved over decades, since the development of the first diagnostic test by Sabin and Feldman (1948). Classically, the IgG result accepted as positive for the diagnosis of acute T. gondii infection is one that shows a significant increase in antibody titers in serial samples, obtained two to four weeks apart and analyzed in parallel. In the case of serological test for diagnosis and IFI, it is considered significant the increase of the titers in four times between paired samples. Our results showed that, on average, there was an increase in IgG ELISA, ELFA and FC antibody levels in the first two months of infection. The IFI technique did not show the significant increase of the titers at this time. Probably, the increase of IgG measured by IFI happens earlier during infection. After the first trimester, the IFI titers remained high, varying individually during the first year after the infection, not being useful to differentiate the time of evolution of the infection. A timerelated study on the anti-T. gondii serological pattern during toxoplasmosis in the Brazilian population including for IgG and IgM antibodies measured by IFIs, hemagglutination and complement fixation showed that IgG IFI presented high titers in the first month of the evaluation (above 1: 8000), unlike hemagglutination, which increases after weeks or months of infection (Camargo and Leser, 1976). Apparently, tests that use less soluble components of the parasite wall, such as the IFI, show a rapid increase in titers when compared to others

exclude the diagnosis of recent-acquired toxoplasmosis when serum samples test positive for IgM in the first trimester of pregnancy, contributing effectively for the prompt therapeutic intervention. In the present study, results from different techniques including anti - T. gondii ELISA, ELFA and FC tests for detecting IgA, IgM and IgG as well as IgG avidity were evaluated to identify the most promising methods applicable to temporal diagnosis of acquired toxoplasmosis. Several statistical strategies were employed for data analysis, which allowed the pre-selection of the biomarkers IgM ELFA, IgM ELISA, IgM FC, IgG avidity ELFA and IgG avidity FC. These biomarkers displayed a promising performance in the time-sensitive diagnosis of T. gondii infection. However, the overall analyses of these parameters, taken alone, were not applicable to the laboratorial diagnosis at specific time-points after symptoms onset. In fact, the pre-selected biomarkers were only useful to identify acute and late convalescent patients after symptoms onset but not the early convalescent group. In this regard, anti-IgG antibodies are detected one to three weeks after the onset of symptoms, depending on the serologic technique used, and reach peak values over a variable period of time (6 to 8 weeks), remaining detectable throughout life. They are widely used in screening for past infection (Remington et al., 2006). There are several laboratory tests available for IgG detection. The most used are ELISA, 7


G.B. Barros et al.

Fig. 4. Performance of Serologic Tests for Temporal Diagnosis of Acute, Early and Late Convalescent T. gondii Infection. Scatter plot distribution of individual values were employed to display the reactivity profile of T. gondii infected-patients for the pre-selected biomarkers, including IgM ELFA, IgM ELISA, IgM FC, IgG avidity ELFA and IgG avidity FC. The reactivity profile obtained for the three sets of T. gondii infected patients were compared, which were referred as acute (1–2–3 months = ○), early (4–6–8 months = ) and late convalescent (10–12 months = ). The percentages of patients below or above the adjusted cut-offs are displayed next to each group.

point between low and high avidity and their correlation with time, items that make the comparison of results unclear. In addition, the results of the ELFA IgG avidity showed a gradual and consistent evolution during the 12 months of serological follow-up applicable for temporal diagnosis of acute toxoplasmosis, but flow cytometry was a more striking and sensitive technique. IgG avidity assessed by Flow Cytometry can be useful for the diagnosis of T. gondii infection, with potential to refine the IgG avidity evaluation during acute toxoplasmosis. Other studies found similar kinetics (Montoya et al., 2004; Fricker-Hidalgo et al., 2006), but with greater amplitude of variation between results. Our study identified appropriate cut-offs to discriminate more accurately the different phases of toxoplasmosis, adjusted from the conventional thresholds (Fig. 2), establishing innovative diagnostic tools for the evaluation of the temporal diagnosis of acquired toxoplasmosis. The utilization of the adjusted cut-off points increased accuracy of segregation between acute (1 month) and convalescent (12 months) T. gondii infection for all tests, except for IgG avidity FC. In conclusion, multivariate analysis generated a three-step algorithm comprising IgM ELFA screening followed by ELFA and FC-AFTA IgG avidity with high accuracy in discriminating acute from late convalescent infection. Together, these findings demonstrate the applicability of the proposed panel of diagnostic tools for accurate temporal classification of T. gondii infection.

that use more soluble constituents, especially of cytoplasmic origin such as hemagglutination, and the ELISA and ELFA used in our study. The evaluation of the dynamics of antibodies requires the formation of a panel of sera collected longitudinally, according to the design of the present study. This was not observed in several studies that analyzed the titers of different patients at different times of the infection (Beghetto et al., 2003; Durlach et al., 2003; Kodym et al., 2007; Flori et al., 2008). Results of several studies have shown that antibodies of high avidity of IgG are predominantly found in cases of chronic infection (Beghetto et al., 2003; Montoya et al., 2004; Kodym et al., 2007). However, the same consensus is not obtained regarding the detection of low avidity antibodies, since they are identified in a variable percentage of individuals with more than one year of infection (Beghetto et al., 2003; Montoya et al., 2004; Lefevre-Pettazzoni et al., 2006). Montoya et al. (2004) followed a cohort of 73 consecutive patients (104 serum samples) with diagnosis of toxoplasmic lymphadenopathy (LT), with the objective of studying the kinetics of IgG avidity and its clinical utility, using the ELFA technique. The data showed that 97% of the low avidity results were within the first 6 months of infection and that 82% of individuals who had high avidity antibodies had > 7 months of disease. Paul (1999) showed that only 53% of individuals infected with parasitic diseases with high avidity results had > 5 months of illness. In a systematic survey on IgG avidity in subjects with documented seroconversion (pregnant or immunocompetent patients), LefevrePettazzoni et al. (2007) found 0 to 66% of low avidity in chronic infections amongst the 11 published studies. They also showed heterogeneity in the execution of the tests carried out in the studies, such as the use of in-house techniques, the way of calculating the index, large variation in the period of follow-up of patients, definition of the cut-off

Acknowledgements This study was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação Oswaldo Cruz (FIOCRUZ), Fundação de Amparo à Pesquisa do Estado de Minas 8


G.B. Barros et al.

Fig. 5. Multivariate Analysis of Serologic Tests Applied to the Temporal Diagnosis of Acute, Early and Late Convalescent T. gondii Infection. Three sets of statistical analysis including decision tree, discriminant components and 3D–Scatter plots were performed in order to understand the interaction amongst the pre-selected biomarkers tested: IgM ELFA, IgG avidity index (AI) for IgG ELFA and AI IgG FC applied to the temporal diagnosis of T. gondii infection. The statistical analyses were performed considering the three groups of T. gondii-infected patients categorized previously as acute (1–2–3 months = ), early (4–6–8 months = ) and late convalescent (10–12 months = ). The percentages and ratios of patients below or above the adjusted cut-offs in the three-step algorithm drawn by the decision tree analysis are displayed in each rectangle. The results for the discriminant analysis including the leaveone-out-cross-validation are included in the inserted table. 3D–Scatter plot was performed for the IgM ELFA, AI IgG ELFA and AI IgG FC.

the sabbatical leave. OAMF thank CNPq for fellowships (PQ) and FAPEAM for the PVS program (PECTI-AM/PG#019/2013). JGCdR received fellowship from Programa Nacional de Pós-Doutorado (PNPD/ CAPES).

Gerais (FAPEMIG) (APQ-01877-12, APQ-01004-13 and INV-00106-13) and Fundação de Amparo à Pesquisa do Espírito Santo (FAPESPPSUS#50820249/2010). The authors thank the program for technological development in tools for health-PDTIS-FIOCRUZ for the use of its facilities and special thanks to the Central Sorológica de Vitória and MICRA Laboratory for technical support. GBB is thankful to the Departmento de Pediatria at Universidade Federal do Espírito Santo for 9


G.B. Barros et al.

contamination. Diagn. Microbiol. Infect. Dis. 64 (3), 267–274. Lefevre-Pettazzoni, M., Le Cam, S., Wallon, M., Peyron, F., 2006 Nov. Delayed maturation of immunoglobulin G avidity: implication for the diagnosis of toxoplasmosis in pregnant women. Eur. J. Clin. Microbiol. Infect. Dis. 25 (11), 687–693. Lefevre-Pettazzoni, M., Bissery, A., Wallon, M., Cozon, G., Peyron, F., Rabilloud, M., 2007 Mar. Impact of spiramycin treatment and gestational age on maturation of toxoplasma gondii immunoglobulin G avidity in pregnant women. Clin. Vaccine Immunol. 14 (3), 239–243. Liesenfeld, O., Montoya, J.G., Kinney, S., Press, C., Remington, J.S., 2001 Apr 15. Effect of testing for IgG avidity in the diagnosis of Toxoplasma gondii infection in pregnant women: experience in a US reference laboratory. J Infect Dis 183 (8), 1248–1253. Mineo, J.R., Camargo, M.E., Ferreira, A.W., 1980 Feb. Enzyme-linked immunosorbent assay for antibodies to Toxoplasma gondii polysaccharides in human toxoplasmosis. Infect. Immun. 27 (2), 283–287. Mineo, J.R., Camarco, M.E., Ferreira, A.W., Almeida, G., 1986 Jan-Feb. Research on IgM anti-Toxoplasma gondii antibodies by using a reverse immunoenzymatic technic. Rev. Inst. Med. Trop. Sao Paulo 28 (1), 6–11. Montoya, J.G., Huffman, H.B., Remington, J.S., 2004 Oct. Evaluation of the immunoglobulin G avidity test for diagnosis of toxoplasmic lymphadenopathy. J. Clin. Microbiol. 42 (10), 4627–4631. van Nunen, M.C., van der Veen, J., 1965 Sep. Examination for toxoplasmosis by the fluorescent antibody technique. Trop. Geogr. Med. 17 (3), 246–253. Paul, M., 1999 Jul. Immunoglobulin G avidity in diagnosis of toxoplasmic lymphadenopathy and ocular toxoplasmosis. Clin. Diagn. Lab. Immunol. 6 (4), 514–518. Remington, J.S., McLeod, R., Thulliez, P., Desmonts, G., 2006. Toxoplasmosis. In: Remington, J.S., Klein, J.O., Wilson, C.B., Baker, C.J. (Eds.), Infectious Diseases of Fetus and Newborn Infant. Elsevier Sauders, Philadelphia, pp. 947–1091. Robert-Gangneux, F., Dardé, M.L., 2012 Apr. Epidemiology of and diagnostic strategies for toxoplasmosis. Clin. Microbiol. Rev. 25 (2), 264–296. Roberts, A., Hedman, K., Luyasu, V., Zufferey, J., Bessières, M.H., Blatz, R.M., Candolfi, E., Decoster, A., Enders, G., Gross, U., Guy, E., Hayde, M., Ho-Yen, D., Johnson, J., Lécolier, B., Naessens, A., Pelloux, H., Thulliez, P., Petersen, E., 2001 Jul. Multicenter evaluation of strategies for serodiagnosis of primary infection with toxoplasma gondii. Eur. J. Clin. Microbiol. Infect. Dis. 20 (7), 467–474. Sabin, A.B., Feldman, H.A., 1948 Dec 10. Dyes as microchemical indicators of a new immunity phenomenon affecting a protozoon parasite (toxoplasma). Science 108 (2815), 660–663. Silva-dos-Santos, P.P., Barros, G.B., Mineo, J.R., de Oliveira Silva, D.A., Menegaz, M.H., Serufo, J.C., Dietze, R., Martins-Filho, O.A., Lemos, E.M., 2012 Apr 30. Flow cytometry-based algorithm to analyze the anti-fixed toxoplasma gondii tachyzoites IgM and IgG reactivity and diagnose human acute toxoplasmosis. J. Immunol. Methods 378 (1–2), 33–43. Sroka, J., Wójcik-Fatla, A., Zajac, V., Sawczyn, A., Cisak, E., Karamon, J., Dutkiewicz, J., Bojar, I., 2016 Dec 23. Comparison of the efficiency of two commercial kits - ELFA and Western blot in estimating the phase of Toxoplasma gondii infection in pregnant women. Ann. Agric. Environ. Med. 23 (4), 570–575. Wilson, M.B., Nakane, P.K., 1976. The covalent coupling of proteins to periodate-oxidized sephadex: a new approach to immunoadsorbent preparation. J. Immunol. Methods 12 (1–2), 171–181.

References Ajzenberg, D., Yera, H., Marty, P., Paris, L., Dalle, F., Menotti, J., Aubert, D., Franck, J., Bessières, M.H., Quinio, D., Pelloux, H., Delhaes, L., Desbois, N., Thulliez, P., RobertGangneux, F., Kauffmann-Lacroix, C., Pujol, S., Rabodonirina, M., Bougnoux, M.E., Cuisenier, B., Duhamel, C., Duong, T.H., Filisetti, D., Flori, P., Gay-Andrieu, F., Pratlong, F., Nevez, G., Totet, A., Carme, B., Bonnabau, H., Dardé, M.L., Villena, I., 2009 Apr 15. Genotype of 88 toxoplasma gondii isolates associated with toxoplasmosis in immunocompromised patients and correlation with clinical findings. J Infect Dis 199 (8), 1155–1167. Beghetto, E., Buffolano, W., Spadoni, A., Del Pezzo, M., Di Cristina, M., Minenkova, O., Petersen, E., Felici, F., Gargano, N., 2003 Dec. Use of an immunoglobulin G avidity assay based on recombinant antigens for diagnosis of primary Toxoplasma gondii infection during pregnancy. J. Clin. Microbiol. 41 (12), 5414–5418. Benevides, L., Cardoso, C.R., Milanezi, C.M., Castro-Filice, L.S., Barenco, P.V., Sousa, R.O., Rodrigues, R.M., Mineo, J.R., Silva, J.S., Silva, N.M., 2013 Sep 24. Toxoplasma gondii soluble tachyzoite antigen triggers protective mechanisms against fatal intestinal pathology in oral infection of C57BL/6 mice. PLoS One 8 (9), e75138. Bobić, B., Sibalić, D., Djurković-Djaković, O., 1991. High levels of IgM antibodies specific for Toxoplasma gondii in pregnancy 12 years after primary toxoplasma infection. Gynecol. Obstet. Investig. 31 (3), 182–184. Camargo, M.E., Leser, P.G., 1976 Jul-Aug. Diagnostic information from serological tests in human toxoplasmosis. II Evolutive study of antibodies and serological patterns in acquired toxoplasmosis, as detected by hemagglutination, complement fixation, IgG and IgM-immunofluorescence tests. Rev. Inst. Med. Trop. Sao Paulo 18 (4), 227–238. Desmonts, G., Couvreur, J., 1974 May. Congenital toxoplasmosis. A prospective study of 378 pregnancies. N. Engl. J. Med. 290 (20), 1110–1116. Desmonts, G., Remington, J.S., 1980 Jun. Direct agglutination test for diagnosis of toxoplasma infection: method for increasing sensitivity and specificity. J. Clin. Microbiol. 11 (6), 562–568. Durlach, R.A., Kaufer, F., Carral, L., Hirt, J., 2003 Jul. Toxoplasmic lymphadenitis—clinical and serologic profile. Clin. Microbiol. Infect. 9 (7), 625–631. Flori, P., Bellete, B., Crampe, C., Maudry, A., Patural, H., Chauleur, C., Hafid, J., Raberin, H., Tran Manh Sung, R., 2008 Mar. A technique for dating toxoplasmosis in pregnancy and comparison with the Vidas anti-toxoplasma IgG avidity test. Clin. Microbiol. Infect. 14 (3), 242–249. Fricker-Hidalgo, H., Saddoux, C., Suchel-Jambon, A.S., Romand, S., Foussadier, A., Pelloux, H., Thulliez, P., 2006 Oct. New Vidas assay for toxoplasma-specific IgG avidity: evaluation on 603 sera. Diagn. Microbiol. Infect. Dis. 56 (2), 167–172. Johnson, R.A.E., Wichern, D.W., 1992. Applied Multivariate Statistical Analysis. PrenticeHall, New Jersey. Kodym, P., Machala, L., Rohácová, H., Sirocká, B., Malý, M., 2007 Jan. Evaluation of a commercial IgE ELISA in comparison with IgA and IgM ELISAs, IgG avidity assay and complement fixation for the diagnosis of acute toxoplasmosis. Clin. Microbiol. Infect. 13 (1), 40–47. Laboudi, M., Sadak, A., 2017 Aug. Serodiagnosis of toxoplasmosis: the effect of measurement of IgG avidity in pregnant women in Rabat in Morocco. Acta Trop. 172, 139–142. Lachaud, L., Calas, O., Picot, M.C., Albaba, S., Bourgeois, N., Pratlong, F., 2009 Jul. Value of 2 IgG avidity commercial tests used alone or in association to date toxoplasmosis

10