Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Validation of Multiplex Serology for human hepatitis viruses B and C, human T-lymphotropic virus 1 and Toxoplasma gondii

  • Nicole Brenner ,

    Roles Formal analysis, Investigation, Visualization, Writing – original draft (NB); (TW)

    Affiliations Infections and Cancer Epidemiology, Infection, Inflammation and Cancer Research Program, German Cancer Research Center (DKFZ), Heidelberg, Germany, Faculty of Biosciences, Heidelberg University, Heidelberg, Germany

  • Alexander J. Mentzer,

    Roles Formal analysis, Investigation, Project administration, Writing – review & editing

    Affiliations The Wellcome Centre for Human Genetics, University of Oxford, Oxford, United Kingdom, Big Data Institute, Li Ka Shing Centre for Health Information and Discovery, University of Oxford, Oxford, United Kingdom

  • Julia Butt,

    Roles Methodology, Writing – review & editing

    Affiliation Infections and Cancer Epidemiology, Infection, Inflammation and Cancer Research Program, German Cancer Research Center (DKFZ), Heidelberg, Germany

  • Kathrin L. Braband,

    Roles Investigation, Writing – review & editing

    Affiliation Infections and Cancer Epidemiology, Infection, Inflammation and Cancer Research Program, German Cancer Research Center (DKFZ), Heidelberg, Germany

  • Angelika Michel,

    Roles Methodology, Writing – review & editing

    Affiliation Infections and Cancer Epidemiology, Infection, Inflammation and Cancer Research Program, German Cancer Research Center (DKFZ), Heidelberg, Germany

  • Katie Jeffery,

    Roles Resources, Writing – review & editing

    Affiliation Department of Microbiology, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom

  • Paul Klenerman,

    Roles Resources, Writing – review & editing

    Affiliations Department of Microbiology, Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom, NIHR Biomedical Research Centre, Oxford, United Kingdom

  • Barbara Gärtner,

    Roles Resources, Writing – review & editing

    Affiliation Institut für Medizinische Mikrobiologie und Hygiene, Universität des Saarlands, Homburg, Germany

  • Paul Schnitzler,

    Roles Resources, Writing – review & editing

    Affiliation Center for Infectious Diseases, Virology, University Hospital of Heidelberg, Heidelberg, Germany

  • Adrian Hill,

    Roles Resources, Writing – review & editing

    Affiliations The Wellcome Centre for Human Genetics, University of Oxford, Oxford, United Kingdom, The Jenner Institute, University of Oxford, Oxford, United Kingdom

  • Graham Taylor,

    Roles Resources, Writing – review & editing

    Affiliation Molecular Diagnostic Unit, Imperial College Healthcare NHS Trust, London, United Kingdom

  • Maria A. Demontis,

    Roles Resources, Writing – review & editing

    Affiliation Molecular Diagnostic Unit, Imperial College Healthcare NHS Trust, London, United Kingdom

  • Edward Guy,

    Roles Resources, Writing – review & editing

    Affiliation Toxoplasma Reference Unit, Public Health Wales Microbiology, Swansea, United Kingdom

  • Stephen J. Hadfield,

    Roles Resources, Writing – review & editing

    Affiliation Toxoplasma Reference Unit, Public Health Wales Microbiology, Swansea, United Kingdom

  • Rachael Almond,

    Roles Conceptualization, Writing – review & editing

    Affiliation UK Biobank, Stockport, United Kingdom

  • Naomi Allen,

    Roles Conceptualization, Writing – review & editing

    Affiliations UK Biobank, Stockport, United Kingdom, Nuffield Department of Population Health, Medical Sciences Division, University of Oxford, Oxford, United Kingdom

  • Michael Pawlita,

    Roles Methodology, Supervision, Writing – review & editing

    Affiliation Molecular Diagnostics of Oncogenic Infections Division, Infection, Inflammation and Cancer Research Program, German Cancer Research Center (DKFZ), Heidelberg, Germany

  •  [ ... ],
  • Tim Waterboer

    Roles Methodology, Project administration, Supervision, Writing – review & editing (NB); (TW)

    Affiliation Infections and Cancer Epidemiology, Infection, Inflammation and Cancer Research Program, German Cancer Research Center (DKFZ), Heidelberg, Germany

  • [ view all ]
  • [ view less ]

Validation of Multiplex Serology for human hepatitis viruses B and C, human T-lymphotropic virus 1 and Toxoplasma gondii

  • Nicole Brenner, 
  • Alexander J. Mentzer, 
  • Julia Butt, 
  • Kathrin L. Braband, 
  • Angelika Michel, 
  • Katie Jeffery, 
  • Paul Klenerman, 
  • Barbara Gärtner, 
  • Paul Schnitzler, 
  • Adrian Hill


Multiplex Serology is a high-throughput technology developed to simultaneously measure specific serum antibodies against multiple pathogens in one reaction vessel. Serological assays for hepatitis B (HBV) and C (HCV) viruses, human T-lymphotropic virus 1 (HTLV-1) and the protozoan parasite Toxoplasma gondii (T. gondii) were developed and validated against established reference assays. For each pathogen, between 3 and 5 specific antigens were recombinantly expressed as GST-tag fusion proteins in Escherichia coli and tested in Monoplex Serology, i.e. assays restricted to the antigens from one particular pathogen. For each of the four pathogen-specific Monoplex assays, overall seropositivity was defined using two pathogen-specific antigens. In the case of HBV Monoplex Serology, the detection of past natural HBV infection was validated based on two independent reference panels resulting in sensitivities of 92.3% and 93.0%, and specificities of 100% in both panels. Validation of HCV and HTLV-1 Monoplex Serology resulted in sensitivities of 98.0% and 95.0%, and specificities of 96.2% and 100.0%, respectively. The Monoplex Serology assay for T. gondii was validated with a sensitivity of 91.2% and specificity of 92.0%. The developed Monoplex Serology assays largely retained their characteristics when they were included in a multiplex panel (i.e. Multiplex Serology), containing additional antigens from a broad range of other pathogens. Thus HBV, HCV, HTLV-1 and T. gondii Monoplex Serology assays can efficiently be incorporated into Multiplex Serology panels tailored for application in seroepidemiological studies.


Multiplex Serology is a high-throughput methodology allowing for the simultaneous detection of serum antibodies against multiple pathogens in one reaction vessel by performing different pathogen-specific assays in parallel. This is achieved by bead-based presentation of selected antigens to primary serum antibodies [1]. Formation of immunocomplexes consisting of the respective antigen, bound primary antibodies and anti-human detection antibodies can be quantitatively measured using a Luminex flow cytometer distinguishing up to 100 bead sets (i.e., antigens). The term ‘Monoplex Serology’ is used to describe pathogen-specific serological assays including only antigens of the respective pathogen, while Multiplex Serology is defined as the combination of at least two Monoplex Serology assays. We have successfully developed Monoplex Serology assays for human papillomaviruses, Helicobacter pylori, human polyomaviruses, and human herpesviruses in order to perform seroepidemiological studies [1, 2, 3, 4]. To expand our portfolio of serological assays for infectious agents known or suspected to be involved in cancer and / or neurodegenerative disease development, we developed and validated pathogen-specific Monoplex Serology assays for hepatitis B (HBV) and C (HCV) viruses, human T-cell lymphotropic virus 1 (HTLV-1) and the protozoan parasite Toxoplasma gondii (T. gondii) for application in Multiplex Serology.

HBV is a hepatotropic DNA virus belonging to the family of Hepadnaviridae, while HCV is a hepatotropic RNA virus and is classified as a member of the Flaviviridae family [5]. Transmission of HBV and HCV mainly occurs via contaminated blood products, vertical transmission, injection drug use or sexual intercourse [6, 7]. Infection by HBV or HCV may be asymptomatic or symptomatic. Acute HBV infection is usually cleared by the adult host but may lead to chronic infection in approximately 5% of cases, whereas 90% of those infected in the neonatal period develop a chronic infection [8, 9]. Approximately 60–90% of HCV infected individuals develop a chronic HCV infection [10, 11, 12]. Both HBV and HCV can cause chronic liver disease in persistently infected individuals, some of whom may develop liver cirrhosis and / or hepatocellular carcinoma (HCC). Thus, HBV and HCV have been classified as group I human carcinogens for HCC [13, 14]. In addition, both viruses have been associated with non-Hodgkin lymphoma and cholangiocarcinoma [13, 14]. The prevalence of chronic HBV and HCV infection and their attributable fractions in liver cancer vary by region and human development index [15, 16]. The worldwide prevalence of chronic HBV infection is 3.6%, with estimates ranging from less than 0.5% in some European and American countries like Norway, France, the UK or US, to more than 20% in South Sudan [15]. Global viraemic HCV prevalence was estimated to be 1.1% ranging between 0.4% in some European countries such as Germany, Denmark or the UK, to 10% in Egypt and Gabon [17]. In 2012, the global attributable fractions of liver cancer were 56% and 20% for HBV and HCV, respectively [16]. Chronic HBV or HCV infection can be treated with different antiviral drugs, which in case of HCV can now produce long-term viral eradication in most cases suggesting a reduction in mortality [18, 19]. Eradication of chronic HBV is not achievable at present [20]. Prophylactic vaccination is available for HBV but not for HCV [13, 21, 22]. Infection history for both viruses can be detected via serology [7, 23]. In the case of HCV, additional tests to detect HCV RNA or antigen are needed to accurately distinguish between current and past infection [7].

HTLV-1 belongs to the group of delta-type retroviruses. Transmission may occur via contaminated blood, sexual intercourse, injection drug use and vertical transmission [24]. This retrovirus has been classified as a group I carcinogen for adult T-cell leukaemia/lymphoma in humans [13, 14]. Additionally, it is known to cause HTLV-1-associated myelopathy/tropical spastic paraparesis [13, 24, 25]. Global HTLV-1 prevalence is generally low (<1%), except for endemic areas, such as certain regions in southwestern Japan, sub-Saharan Africa, South America and the Caribbean islands where a HTLV-1 prevalence of up to 20% was reported [13, 24]. Currently, neither a vaccine against HTLV-1 infection nor any treatment for virus clearance is available. Treatment focuses on HTLV-1 induced diseases, which only occur in a small fraction of infected individuals (≤5%) [13, 26, 27, 28]. HTLV-1 infection is currently diagnosed by a combination of a serological assay and a confirmation assay such as PCR or immunoblotting [24].

T. gondii is an obligate intracellular protozoan parasite infecting warm-blooded animals including humans as intermediate hosts. The parasite’s definitive hosts are felines. Multiple routes of human infection have been reported, including ingestion of contaminated water or undercooked meat, blood products, transplants and congenital infection [29, 30]. Upon infection, T. gondii causes toxoplasmosis that may result in flu-like symptoms in immunocompetent individuals or severe disease in immunocompromised individuals, neonates and congenitally infected foetuses [29, 31]. In addition, infection by T. gondii has been associated with impacts on human behaviour and disorders such as schizophrenia [32, 33]. The necessity for treatment depends on the severity of toxoplasmosis and the immune status of the infected individual [34, 35]. No prophylactic or therapeutic vaccine against T. gondii is available. It has been estimated that the global T. gondii prevalence is approximately 30% [29, 36]. Serological methods are often used for diagnosis of toxoplasmosis. In addition, molecular methods such as PCR have been developed to detect DNA in different biospecimens [31].

The above mentioned pathogens have one common denominator–infected individuals can be detected by serology. In case of HBV, serology allows assessing an individual’s infection and vaccination history. Serological assays can be deployed in seroepidemiological studies to investigate associations between infections and diseases, such as certain types of cancer or neurodegenerative diseases. Thus, we developed pathogen-specific Monoplex Serology assays to detect natural HBV, HCV, HTLV-1 and T. gondii infections. As a secondary aim, HBV Monoplex Serology was aimed to additionally investigate HBV vaccination history.

The developed assays were validated against reference assays used for clinical diagnosis. Selection of a minimum number of pathogen-specific antigens reaching optimum statistical characteristics in comparison to the respective reference assays was pursued for efficient application in Multiplex Serology. Here, we report the successful validation of Monoplex Serology assays against HBV, HCV, HTLV-1 and T. gondii comprising as little as two antigens each and their incorporation into Multiplex Serology.

Materials and methods

Antigen selection and cloning for recombinant antigens expressed in Escherichia coli

In Table 1, antigens selected for development and validation of pathogen-specific Monoplex Serology assays are shown. Proteins and their amino acid sequences were chosen based on reported immunogenicity in the literature. Recombinant proteins were expressed in Escherichia coli (E. coli) without reported signal peptides or transmembrane regions to facilitate efficient bacterial expression. The sequences for T. gondii protein bag-1 and the hepatitis B surface antigen (HBs) were codon-optimized for expression in E. coli and assembled into a modified pGEX4T3 vector (pGEX4T3tag, described below) by gene synthesis (eurofins Genomics, Ebersberg, Germany) [37]. Sequences for the cloned antigens were obtained from DNA vectors or extracted from parasite cells (Table 1). Antigen expression for HCV was previously described [38]. For cloning of antigen sequences into pGEX4T3tag vectors, corresponding primers were designed and ordered from eurofins Genomics. The antigen constructs were then amplified in E. coli DH5α. The inserts were sequenced (eurofins Genomics / GATC Biotech, Konstanz, Germany) and aligned with reference sequences (NCBI nucleotide database) for validation of correct sequences (amino acid changes listed in Table 1).

Table 1. Characteristics of selected pathogen-specific serology antigens.

Antigen expression in E. coli

Non-commercial recombinant antigens were expressed in either E. coli strain BL21 or BL21 Rosetta (Table 1). Antigens were expressed as GST-X-tag fusion proteins with an N-terminal GST and a C-terminal SV40-tag consisting of the last 11 amino acids of the large T-antigen (tag) [37]. The expression vector pGEX4T3tag contains an ampicillin resistance gene (ampR) enabling positive selection of transformed colonies. Transcription is inducible by IPTG. Bacterial cells were lysed and cleared. Glycerol (50% (v/v)) was added to the lysates, and they were stored at -20°C. Lysates underwent quality control as described before including Coomassie staining, western blot staining for N- and C-terminal tags and GST capture ELISA as described elsewhere [37, 38].

Covalent coupling of HBs antigens to polysterene beads

Three different HBs products were obtained from external providers (Table 1). These lack an N-terminal GST-tag and were thus covalently coupled to polystyrene beads as described elsewhere [1]. The commercial HBs purified from serum, or recombinantly expressed in S. cerevisiae were coupled in a concentration of 100 μg/ml. Virus-like particles (VLPs, purified into phosphate buffered saline; Indian Immunologicals LTD., Hyderabad, Telangana, India) were coupled in a concentration of 50 μg/ml to the beads. Prior to coupling, confirmation of VLP structure was undertaken using dynamic light scattering (Zetasizer nano ZS, Malvern Panalyical Ltd., Malvern, UK). The average measured polydispersity index of the six duplicate samples of HBs antigen (HBsAg) was around 0.3 which suggested a degree of polydispersity (using a polydispersity index of less than 0.3 suggesting monodispersed samples). Volume size distribution demonstrated that the majority (>98%) of particles were in the size range of 30 nm confirming preserved VLP structure.

Reference panels and reference assays

Pathogen-specific Monoplex Serology assays were validated against commercial reference assays based on reference panels (RP Ia,b; II-IV; Va,b,c) listed in Table 2. Sera for RP Ia and Va were obtained from the routine diagnostic laboratory of the Saarland University Medical School. The sera were stored at -20°C. Both serum panels are overlapping and comprise n = 230 sera in total of which five sera were not tested for anti-HBs antibodies at the German Cancer Research Center (DKFZ, Heidelberg). Of the 139 anti-HBc negative sera of panel Ia, 110 were tested anti-HBs negative and 29 were tested anti-HBs positive by the Saarland University Medical School. Of the 91 anti-HBc positive sera, 29 were tested negative for anti-HBs and 62 tested positive for anti-HBs by the reference laboratory.

Anonymised sera for RP Ib and II were obtained from the routine diagnostic laboratory of Oxford University Hospitals; each sample had been tested for HBV or HCV, respectively. These sera were stored at -20°C and were reaching the end of their retention period. RP Ib comprised samples tested negative to anti-HBc (n = 100) that were also tested negative for the presence of HBsAg. Of the 57 anti-HBc positive reference samples, 54 were tested negative for HBsAg and 3 tested positive for HBsAg. HCV positive sera of RP II (n = 50) were tested double positive by the ARCHITECT Anti-HCV assay and the Murex anti-HCV assay. These sera were additionally tested for HCV antigen (ARCHITECT HCV Ag, Abbott) of which 22 tested HCV antigen negative and 28 HCV antigen positive. Four sera tested positive by the ARCHITECT Anti-HCV assay are considered false-positive, as the confirmatory assay (Murex anti-HCV assay) was negative.

RP III consisted of sera donated by patients who attended the National Centre for Human Retrovirology to the Communicable Diseases Research Tissue Bank (approved by the National Research Ethics Service: Ref 15/SC/0089). Processing and storage was conducted by the department of Retrovirology and GU Medicine, Division of Infectious Diseases, Imperial College London.

The sera of RP IV were selected from a prospective serum cohort tested with the Sabin-Feldman Dye Test. Negative or positive serostatus was determined with the dye test (cut-off level of 2 IU/ml), with the assay calibrated against the WHO ToxoG International Standard Preparation [41]. The sera were stored at -20°C.

RP Vb and Vc were received from the routine diagnostic testing of the Center for Infectious Diseases at the University Hospital of Heidelberg. The reference sera were tested at the University Hospital of Heidelberg in 2014 (Vb) and 2016 (Vc) and stored at -20°C until testing at the DKFZ.

Monoplex and Multiplex Serology

Serum antibodies against selected antigens (Table 1) were measured using pathogen-specific Monoplex Serology and Multiplex Serology as previously reported [1]. In brief, antigens were expressed as GST-X-tag fusion proteins and in situ affinity purified from crude lysates on glutathione casein-coated fluorescence-labelled polystyrene beads (carboxylated xMAP Microspheres, Luminex Corp. Austin, Texas, USA) except for externally obtained HBs antigens. These were covalently coupled to fluorescence-labelled polystyrene beads as described above. The polystyrene beads contain two fluorescent dyes in different ratios. Thus, up to 100 different bead sets can be distinguished by the Luminex flow cytometer. Antibodies against multiple antigens were measured in one reaction vessel by loading each antigen onto another bead set and subsequent combination into a single bead mix. In Monoplex Serology, only the pathogen-specific antigens described in Table 1 plus antigen GST for background correction were included in the bead mix. In Multiplex Serology, antigens from multiple pathogen-specific Monoplex Serology assays plus antigen GST were combined into the bead mix. Serum was incubated with the antigen-loaded beads and primary antibodies were detected using a biotinylated goat-α-human IgM/IgG/IgA secondary antibody (1:1000, #109-065-064, Jackson Immunoresearch, West Grove, PA, USA). Subsequent incubation with a reporter dye, streptavidin-R-phycoerythrin (1:750, PE-Streptavidin Conjugate, MOSS Inc., Pasadena, CA, USA), allowed quantitative measurement of the formed immunocomplexes in a Luminex 200 flow cytometer. Measuring at least 100 beads per bead set, Median Fluorescence Intensities (MFI) were calculated. Pathogen-specific Monoplex Serology was performed in dilutions 1:100 and 1:1000 on each reference panel. For HBV, HCV and HTLV-1, results for dilution 1:1000 are reported, while for T. gondii the optimum dilution 1:100 is reported.

Multiplex Serology incorporating HBV, HCV, HTLV-1 and T. gondii Monoplex Serology assays was performed at a 1:1000 serum dilution including additional pathogen-specific assays (e.g. human herpesviruses 1–8, human polyomaviruses, human papillomaviruses, human immunodeficiency virus 1).

Statistical analyses

Sera were received and tested in a blinded fashion. After testing, the Wellcome Centre for Human Genetics served as trusted third party, and, following confirmation that preliminary sensitivity and specificity metrics were greater than 85%, merged testing results obtained at DKFZ with the previously provided reference data, thus unblinding the analysis. Antibody measurements were then assessed by reference serostatus and refined cut-offs were determined in a stepwise approach. Analogous to Receiver Operation Characteristics analysis, sensitivity and specificity were optimized to achieve at least 85% by gradually increasing the cut-off starting from a minimum of 30 MFI at serum dilution 1:1000, and 50 MFI at 1:100. Final cut-offs were determined favouring specificity until an additional increase in specificity resulted in a disproportionate loss in sensitivity, thus maximizing agreement as a secondary criterion.

In addition to sensitivity and specificity, Cohen’s kappa (k) plus corresponding 95% confidence intervals (CI) were calculated using SAS 9.4. The overall serostatus for each pathogen based on multiple pathogen-specific antigens was evaluated by systematic combination (e.g. AND/OR algorithms) of the antigens optimizing the statistical characteristics using SAS 9.4. Kappa statistics were evaluated as follows: 0.01≤k≤0.20: slight agreement, 0.21≤k≤0.40: fair agreement, 0.41≤k≤0.60: moderate agreement, 0.61≤k≤0.80 substantial agreement, and 0.81≤k≤0.99: almost perfect agreement [42]. Statistical performance of Multiplex Serology compared to the corresponding reference panel was calculated analogous to the procedure described for Monoplex Serology.

Comparison of Monoplex and Multiplex Serology performance for the corresponding reference panels was conducted by calculating Intraclass Correlation Coefficients (ICCs) using R 3.5.0, package ‘psych’ [43]. ICC(3,1) plus corresponding 95% CI. ICCs were evaluated as follows: 0.01≤ICC≤0.49: poor reliability, 0.50≤ICC≤0.74: moderate reliability, 0.75≤ICC≤0.89: good reliability, 0.90≤ICC≤1.00: excellent reliability [44]. The numbers of sera slightly differed for statistical comparisons from the numbers given in Table 2 by 0–4 sera for RP Ia/b-III and by 27 sera for RP IV due to insufficient bead count during measurements or insufficient sample volume.


Antigen selection

The HBV, HCV, HTLV-1 and T. gondii pathogen-specific Monoplex Serology assays initially comprised 3–5 preselected antigens. For all pathogens, selection took place according to reported antigen immunogenicity (e.g. [38, 45, 46, 47, 48, 49, 50, 51, 52]) and commercial assays. For HBV, infection and/or vaccination status and the associated serological antibody patterns were additionally taken into account. Characteristics of the selected antigens are shown in Table 1. All non-commercial recombinant antigens were expressed as GST-fusion proteins as described previously [37]. Parental plasmids used for antigen expression were sequenced to confirm correct amino acid sequences. Most antigens did not show any variation from the reference sequence. Amino acid changes resulting from nucleotide mismatches are reported in Table 1.Three amino acid changes were found for HTLV-1 rex in comparison with the reported NCBI reference and one for T. gondii bag1 (Table 1).

Comparison of pathogen-specific Monoplex Serology assays with reference serostatus

Five reference serum panels (denoted RP Ia, Ib and II-IV) were used for the evaluation of the Monoplex Serology assays. Measured antibody reactivities (MFI) stratified by reference assay status are shown in Fig 1. In each Monoplex Serology assay, at least three different antigens were included. Cut-offs for single antigen performance were determined by optimizing specificity and sensitivity. Statistical characteristics of the single antigens are shown in Table 3. Overall pathogen-specific seropositivity was evaluated by systematic combination of all included antigens for each pathogen-specific assay. Cut-offs for overall pathogen-specific performance were determined identically as for single antigens, and statistical characteristics for overall seropositivity in Monoplex Serology are shown in Table 4.

Fig 1. Comparison of quantitative antibody measurements (MFI) with reference serostatus in pathogen-specific Monoplex Serology for HBV (Ia, Ib), HCV (II), HTLV-1 (III) and T. gondii (IV).

Ia, Ib, II-IV indicate corresponding reference panels. red lines: optimized cut-offs for single antigen performance; cut-offs were determined by optimizing specificity and sensitivity analogous to Receiver Operator Characteristics analysis. No cut-offs were determined for antigens that showed a reduced capacity to distinguish between reference assay positives and negatives. MFI: Median Fluorescence Intensities.

Table 3. Single antigen performance in Monoplex Serology in corresponding reference panels.

Table 4. Overall pathogen-specific statistical performance in Monoplex Serology in corresponding reference panels.

HBV Monoplex Serology validation.

HBV Monoplex Serology for the detection of natural infection history based on antigens HBc and HBe was validated using two independent reference panels (Ia and Ib). Stratification of measured antibody reactivities by reference assay status revealed high concordance for HBc and HBe with both reference assays (Fig 1 Ia/Ib). For RP Ia, two false-positives (1%) for HBe and six (7%) and seven (8%) false-negatives for HBc and HBe, respectively, were observed. Similarly, only one false-positive (1%) for HBe and three false-negatives (5%) each for HBe and HBc were detected for RP Ib. Statistical characteristics for the single antigens reveal successful assay validation with both reference panels (Table 3). Sensitivities ranged between 92.3% and 93.4%, and specificities between 98.6% and 100.0%. Kappa statistics ranged between 0.92 and 0.94 showing almost perfect agreement. Overall HBV seropositivity was best determined by a combination of both HBc and HBe, or HBc alone. For both reference panels, the combination of HBc and HBe resulted in 100.0% specificity and a sensitivity of 92.3% in case of RP Ia and 93.0% on RP Ib (Table 4). Similar statistical characteristics were observed by an AND/OR combination of HBc and HBe (not shown) consistent with almost perfect correlation between measured HBc and HBe antibody reactivities (rRPIa = 0.98, rRPIb = 0.99, S1 Fig Ia/Ib).

In addition, anti-HBs serology was pursued to be established and validated based on RP Va. However, multiple attempts to establish and to successfully validate anti-HBs measurements in Monoplex Serology failed on our platform (S2 Fig). These attempts included using a GST-HBs-tag fusion protein expressed in E. coli, purified HBs antigen from human plasma (not shown due to high background reactivity), recombinant HBs antigen from Saccharomyces cerevisiae, and HBV virus-like particles (VLP) also used in vaccination programmes to protect against HBV (S2A–S2C Fig). Due to insufficient sensitivity and/or high background reactivity in reference assay negatives, no cut-offs were applied and no statistical characteristics calculated. In addition to RP Va, two additional reference panels (Vb, Vc) were used to evaluate the performance of the covalently coupled HBV VLPs. However, this experiment confirmed the poor capacity to distinguish between anti-HBs reference positives and negatives already observed on RP Va (S2D Fig).

HCV Monoplex Serology validation.

HCV Monoplex Serology was previously validated by Dondog et al. against a meanwhile outdated reference assay [38]. Here, we report a contemporary validation based on a different reference panel (RP II) using the ARCHITECT Anti-HCV assay in combination with a confirmatory ELISA (Murex anti-HCV, Diasorin) used to detect false-positives. HCV Monoplex Serology is based on the Core protein and non-structural (NS) proteins NS3, NS4A, NS5A and NS5B expressed from HCV subtype 1a. HCV antigens Core and NS3 showed high capacity to discriminate between reference assay negatives and positives (Fig 1II). However, antigen Core was less sensitive compared to antigen NS3, detecting eight (16%) compared to one (2%) false-negatives, respectively. Both antigens detected only two false-positives (2%). Thus, both Core and NS3 showed high specificities of 98.1% and sensitivities of 98.0% in the case of NS3 and 84.0% for Core (Table 3). This resulted in almost perfect agreement with the reference assay for both antigens (kNS3 = 0.96, kCore = 0.85). NS4A, NS5A and NS5B were less capable of distinguishing between reference assay positives and negatives and no cut-offs were applied. Only a subset of reference assay positive sera showed antibody reactivity against these antigens. However, detection of antibodies by the additional non-structural proteins was highly specific as almost all reactivities in reference assay negative sera were <100 MFI while reference assay positive sera yielded up to 10,000 MFI (Fig 1II).

Different criteria for overall HCV seropositivity based on multiple HCV antigens were applied (Table 4). Optimum statistics were reached by an AND/OR combination of Core and NS3 yielding a specificity of 96.2% and a sensitivity of 98.0%. In Dondog et al., an AND combination of Core and NS3, or a combination of NS3 with two other non-structural proteins were identified as optimal for overall HCV seropositivity [38]. These criteria were also applied but shown to be less sensitive for analysis of RP II (Table 4, S3 Fig, combinations with other non-structural proteins not shown).

HTLV-1 Monoplex Serology validation

The initial HTLV-1 Monoplex Serology panel included antigens gag, env, tax, rex and HBz and was validated against a combination of two reference assays used as standard procedure in the UK to diagnose HTLV-1 infection on RP III (Table 2). All antigens except HBz distinguished reference assay positive from negative serum samples (Fig 1 III). For gag, env and tax, zero to three (0–3%) false-positives were detected. The rex antigen detected a higher number of false-positives (n = 8; 8%). The number of false-negatives ranged between four for gag (4%) and 19 for tax (19%). Of the 100 reference assay positives, 76 were seropositive to all four informative HTLV-1 antigens (gag, env, tax, rex), six sera to three out of four antigens, and nine sera to gag and env. Only two reference assay positive sera (2%) did not react with any of the included HTLV-1 antigens. The resulting specificity was high for all four antigens ranging between 92.0% for rex and 100.0% for env and tax (Table 3). Sensitivity was lowest for tax (81.0%) and highest for gag (96.0%). Kappa statistics varied from substantial agreement for antigen rex (k = 0.75) to almost perfect agreement for the other antigens (k = 0.81–0.93). Two criteria to determine overall HVTL-1 seropositivity improved the single antigen statistics of gag: either inclusion of env alone, or inclusion of env, tax and rex. Using a combination of gag and env for determination of overall HTLV-1 seropositivity, 100.0% specificity and 94.0% sensitivity were reached (Table 4). With the additional inclusion of tax and rex one additional serum was detected increasing the sensitivity to 95% (not shown) while maintaining specificity of 100%.

Potential cross-reactivity of HTLV-1 with HIV-1 antibody measurements was assessed by measuring antibody reactivities against HIV-1 gag, env, tat and rev (Kranz et al., submitted) in parallel with their homologs HTLV-1 gag, env, tax and rex. Of the 200 HTLV-1 reference sera, 10 showed detectable antibody reactivities against any of the HIV-1 antigens. As antibody patterns and responses did not correlate between HIV-1 and HTLV-1 homologs (r = -0.01 to r = 0.09), no cross-reactivity between HIV-1 and HTLV-1 could be detected based on RP III (S4 Fig).

T. gondii Monoplex Serology validation.

T. gondii Monoplex Serology initially comprised antigens p22, sag1 and bag1, and was validated against the Sabin-Feldman dye test on reference panel IV. Both antigens p22 and sag1 showed high capacity to distinguish reference assay serostatus, with overlapping reactivities for borderline reacting sera (range: 30–110 MFI; Fig 1IV). This resulted in detection of six false-positives (12%) out of 50 reference assay negative sera for both p22 and sag1. Ten (7%) and 23 (16%) false-negatives were detected out of 148 reference assay positive sera for p22 and sag1, respectively. This resulted in a specificity of 86.0% for both p22 and sag1, and a sensitivity of 92.6% and 84.5%, respectively (Table 3). Kappa statistics of k = 0.77 and k = 0.64 for p22 and sag1 provided evidence of substantial agreement. Antigen bag1 did not show to be informative for the discrimination between reference assay negatives and positives. An AND/OR combination of T. gondii antigens p22 and sag1 to define overall T. gondii seropositivity improved specificity to 92.0% with comparable sensitivity of 91.2% (Table 4, S5 Fig). This resulted in substantial agreement with the reference assay (k = 0.79).

Comparison of performance in Monoplex and Multiplex Serology

To compare the assay performance between monoplex and multiplex formats, HBV, HCV, HTLV-1 and T. gondii Monoplex Serology assays were incorporated into a broad Multiplex Serology panel and re-tested at the optimum serum dilution for the majority of incorporated assays (1:1000). Assay performance in multiplex format was reassessed based on the described reference panels (Ia, Ib, II-IV). The statistical performance (specificity, sensitivity, Cohen’s kappa) was re-evaluated against the reference assay serostatus and compared to the corresponding pathogen-specific Monoplex Serology results (Fig 2). For some of the pathogen-specific assays, slight variations in sensitivity and specificity were detected within overlapping confidence intervals. Overall statistical performance in Multiplex Serology was maintained. Sensitivity and specificity for overall seropositivity for HBV, HCV and HTLV-1 exceeded 90% in both monoplex and multiplex assay format, maintaining high concordance with the corresponding reference assay (k ≥ 0.85). Sensitivity, specificity and kappa statistics results for T. gondii at the sub-optimal serum dilution of 1:1000 are shown and compared in Fig 2. Specificity was high (>90%) in both Monoplex and Multiplex Serology settings. However, sensitivity was impaired (<80%) also strongly reducing Cohen’s kappa in both settings (kMonoplex: 0.63; kMultiplex: 0.52).

Fig 2. Statistical performance of pathogen-specific assays (overall seropositivity to HBV, HCV (AND/OR), HTLV-1, T. gondii) in monoplex (blue) and multiplex (orange) format.

To enhance visualization, Cohen’s kappa statistics are illustrated as percentages. Sensitivity, specificity and Cohen’s kappa statistics are illustrated by boxes while the corresponding 95% CI is indicated by horizontal lines. Performance of Monoplex and Multiplex Serology was directly compared on the corresponding reference panel using ICCs. The comparison yielded moderate to excellent reliability between monoplex and multiplex format: ICCHBV RPIa: 0.94 (95% CI 0.92–0.95), ICCHBV RPIb: 0.97 (95% CI 0.96–0.98), ICCHCV: 0.97 (95% CI 0.96–0.98), ICCHTLV-1: 0.94 (95% CI 0.92–0.95), ICCT.gondii: 0.69 (95% CI 0.61–0.76). kappa: Cohen’s kappa. ICC: Intraclass correlation coefficient. CI: confidence interval. RP: reference panel.

A direct comparison of Monoplex versus Multiplex Serology performance was conducted using ICCs showing excellent reliability (ICC: 0.94–0.97) for HBV, HCV and HTLV-1. The ICC for comparison of T. gondii Monoplex and Multiplex Serology showed moderate reliability (ICC: 0.69). This was strongly influenced by the loss in sensitivity at testing dilution 1:1000 and a reduced number of samples tested in multiplex format (n = 172 as opposed to n = 198).


Here, we report the successful development and validation of Monoplex Serology assays for HBV, HCV, HTLV-1 and T. gondii. Overall seropositivity for each pathogen yielded a sensitivity between 91.2% and 98.0%, and a specificity between 92.0% and 100.0% in comparison to reference assays in clinical use. Each Monoplex Serology assay requires only two antigens, and can thus be efficiently incorporated into Multiplex Serology, which makes simultaneous measurement of antibody reactivities and resulting detection of seropositivities against multiple pathogens very efficient. The pathogen-specific Monoplex Serology assays were additionally combined and tested in Multiplex Serology on the respective reference serum panels including a broader infectious disease assay panel (e.g. human herpesviruses 1–8, human polyomaviruses, human papillomaviruses, human immunodeficiency virus 1). Statistical performance was generally maintained (ICC: 0.69–0.97).

We used two independent reference panels for validation of HBV Monoplex Serology; the corresponding reference assays are closely related and both measured anti-HBc. This explains the very good agreement between anti-HBc Monoplex Serology and both reference assays. In general, different combinations of anti-HBc, anti-HBs and HBsAg in blood allow distinguishing between susceptible, vaccinated, acutely or chronically HBV infected individuals, or individuals who are immune after natural infection [45, 53]. During current (i.e., acute or chronic) HBV infection or after cleared HBV infection, IgG antibodies against HBc are present. Thus, seropositivity to antigen HBc in HBV Monoplex Serology indicates an HBV infection that might be resolved, acute or chronic.

Almost perfect concordance of anti-HBe with both reference assays was observed; in addition, the anti-HBc and anti-HBe measurements in HBV Monoplex Serology showed very high correlation. In both tested reference panels, most individuals were observed to be double seropositive for HBc and HBe by HBV Monoplex Serology. This explains the high concordance of anti-HBe measurements with the reference assays. However, neither reference panel had been tested for antibodies against HBe by the reference institutions. Thus, the validity and potential meaning of anti-HBe measurements in HBV Monoplex Serology need to be evaluated in the context of known seropatterns during HBV infection. During chronic HBV infection, or in the course of clearance of acute infection, individuals develop anti-HBe antibodies [23, 53]. Thus, individuals with a cleared HBV infection are anti-HBe seropositive. In RP Ib, the majority of HBV seropositive sera (95%) were positive for anti-HBc and negative for HBsAg by the reference institution. This pattern points to a past and cleared infection which is accompanied by HBe antibodies [53]. In RP Ia, approximately one third of HBV seropositive sera was positive for anti-HBc and negative for anti-HBs marking either a phase during acute infection, or chronic infection. The remaining two thirds of HBV seropositive sera were double seropositive for anti-HBc and anti-HBs indicating past infection. In these two subgroups of HBV infected individuals, antibodies against antigen HBe may occur [53]. Thus, although not directly validated against an anti-HBe reference assay, the presence of antibodies against HBe in the two independent anti-HBc reference panels is plausible, and supports the method’s validity.

HBV vaccination is based on HBsAg alone, and thus detection of only anti-HBs antibodies marks immunity due to vaccination. Inclusion of anti-HBs measurements in HBV Monoplex Serology would not only facilitate the discrimination of the exact HBV infection status in anti-HBc positives but would also help to identify vaccinated individuals in anti-HBc seronegatives [23, 53]. Despite multiple attempts, incorporation of anti-HBs measurements in HBV Monoplex Serology was not successful. Even covalent coupling of vaccine-derived HBsAg VLPs to the beads did not enable specific detection of anti-HBs antibodies based on three independent reference serum panels. Most likely, the specific immunogenic structure of the HBsAg complexes was destroyed during the rigid covalent coupling procedure to the beads.

HCV Monoplex Serology was successfully validated against a combination of the ARCHITECT Anti-HCV assay and confirmatory ELISA Murex anti-HCV. The antigen achieving the highest concordance with the reference assay was NS3 (k = 0.96). All other antigens showed very high specificities, but impaired sensitivities (Fig 1II, Table 3). The ARCHITECT Anti-HCV reference assay is based on 2 antigens representing a combination of recombinant structural and non-structural HCV antigens. The recombinant antigen HCr43 is expressed in E. coli and was reported to consist of parts of NS3 and the Core antigen. The second recombinant antigen, c100-3, comprises regions of NS3 and NS4 expressed by S. cervisiae [54]. The ARCHITECT Anti-HCV assay is a chemiluminescent microparticle immunoassay (CMIA) simultaneously presenting both recombinant antigens on microparticles to the primary antibodies in the serum. Seropositivity is defined by an overall anti-HCV signal and does not differentiate between individual contributions of the recombinant antigens to the signal. Similarly, the confirmatory ELISA Murex anti-HCV uses an antigen mixture of purified HCV antigens (NS3, rAg, Core, NS4, NS5) to detect anti-HCV antibodies [55]. The principle used in both reference assays represents a considerably different approach to data analysis in HCV Monoplex Serology. High specificity of all antigens included in HCV Monoplex Serology indicates very good concordance with both reference assays for HCV serology negatives. Impaired sensitivity for all antigens except NS3 suggests two points. First, at least in this serum reference panel, antibodies against NS3 are the most frequent in HCV seropositives compared to the other HCV Monoplex Serology antigens. Secondly, the seemingly impaired sensitivity of the other HCV antigens might be indicative of true seropatterns in HCV seropositives as the reference assay does not distinguish between seropositivity against different antigens. Thus, HCV Monoplex Serology potentially provides a higher degree of detail that cannot be evaluated in comparison to this HCV reference assay.

The reference laboratory further characterised sera tested positive by the ARCHITECT Anti-HCV assay and the confirmatory ELISA (Murex anti-HCV) using the ARCHITECT HCV Ag assay (Abbott) to distinguish between past infection (ARCHITECT Anti-HCV and Murex anti-HCV double positive) and current HCV infection (triple positive: ARCHITECT Anti-HCV, Murex anti-HCV, ARCHITECT HCV Ag assay). Although HCV Monoplex Serology could not differentiate between past and current HCV infection according to the described reference assay combination (not shown), all 28 sera diagnosed with current HCV infection were correctly classified by the NS3 antigen. In addition, 26 (93%) of these sera were also seropositive against antigen Core and 21 (75%) reacted with at least two other non-structural proteins in addition to NS3. Of the 22 sera classified as past HCV infection by the reference institution, 21 (96%) were seropositive to NS3 in HCV Monoplex Serology, 17 (77%) were seropositive to antigen Core and 11 (50%) reacted with at least two other non-structural proteins in addition to NS3. These antibody patterns indicate lower seroreactivity against Core and other non-structural HCV proteins than against NS3 in individuals with past instead of current HCV infection.

Four sera were tested positive by the reference assay (ARCHITECT Anti-HCV), but negative by the confirmatory ELISA (Murex anti-HCV, Diasorin) and were thus regarded as false-positives by the reference laboratory. Of these four sera, only one reacted with antigen NS3 in HCV Monoplex Serology and none with antigen Core or at least two non-structural proteins other than NS3. Although not being representative due to low numbers, this indicates that HCV Monoplex Serology may detect less false-positives than the ARCHITECT Anti-HCV assay.

Dondog et al. validated HCV Monoplex Serology against the Abbott AxSYM HCV version 3.0 microparticle enzyme immune assay plus a confirmatory assay (RT-PCR or Western blot) [38]. The AxSYM assay was a precursor of the ARCHITECT Anti-HCV assay and was based on a partly overlapping set of antigens [56]. Differential performance of HCV Monoplex Serology and the resulting optimum criterion for definition of HCV seropositivity in comparison with the previous validation may be due to different characteristics of the reference assays such as the antigen panel and/or the serum panel.

HTLV-1 Monoplex Serology was successfully validated against a combination of two reference assays used for HTLV-1 diagnosis in the UK (Table 2). Single antigen performance of HTLV-1 gag and env, an AND/OR combination of gag and env, as well as a combination of all informative HTLV-1 antigens (gag, env, tax, rex) resulted in high concordance (k: 0.88–0.95) with the standard diagnosis algorithm in the UK. Inclusion of antigens tax and rex to determine overall HTLV-1 seropositivity classified one additional reference assay seropositive serum correctly in comparison to the gag AND/OR env combination. However, this algorithm is very detailed and likely suffers from overfitting to the underlying serum panel. Thus, it is questionable whether it is directly transferable to other studies. Therefore, for inclusion of a robust and efficient HTLV-1 assay in Multiplex Serology, antigens gag and env are considered sufficient.

Alignment of the HTLV-1 antigen rex sequence with the corresponding sequence of the NCBI database revealed 3 amino acid changes. However, no reference sequence of strain MT-2 was available for rex. Thus, these mismatches could also represent differences in the parental DNA. Statistical performance of HTLV-1 antigen rex suggests no major impairment of immunogenic epitopes.

To assess potential cross-reactivity of the included HTLV-1 antigens, the reference sera were also tested for antibodies against homologous HIV-1 antigens. However, no correlation was observed (r: -0.01 to 0.09, SF 4). Antibody patterns in HTLV-1 reference sera reacting against at least one HIV-1 antigen suggest that two HTLV-1 negative individuals are true HIV-1 positives and 8 individuals were double seropositive for HTLV-1 and HIV-1 (not shown).

Toxoplasma gondii Monoplex Serology was validated against the Sabin-Feldman dye test. Single antigen performance could be improved by a sag1 AND/OR p22 combination to define overall T. gondii seropositivity. The reference assay uses T. gondii tachyzoites. During this life cycle stage, five major surface antigens are expressed including p22 and sag1 [57, 58]. The 14 observed false-negatives by the sag1 AND/OR p22 criterion might be seropositive to at least one of the other three surface antigens and are thus missed by T. gondii Monoplex Serology. The observation that a considerable number of reference assay positive sera are seropositive to p22, but not against sag1 indicates that upon infection with T. gondii, the host might not elicit antibodies against all surface antigens. Thus, T. gondii Monoplex Serology could potentially be improved by including further surface antigens not expressed in this study.

T. gondii serology at both serum dilutions was shown to be highly specific (>90%). Reduced sensitivity in Monoplex and Multiplex Serology for T. gondii was observed at a 1:1000 serum dilution in comparison to 1:100. Thus, T. gondii Serology (both in monoplex and multiplex format) is recommended to be performed at serum dilution 1:100. For incorporation of T. gondii Monoplex Serology into Multiplex Serology at dilution 1:1000, it should be noted that despite missing true-positives, the detection of false-positives is expected to be very limited. Therefore, depending on the aim of the study, inclusion of T. gondii serology into Multiplex Serology at serum dilution 1:1000 might be feasible.

Our successful assay validation for HBV, HCV, HTLV-1 and T. gondii implies robustness of their use in future large-scale seroepidemiological studies. In fact, all assays have already been incorporated in Multiplex Serology [59, 60, 61].

Supporting information

S1 Fig. Correlation of anti-HBc and anti-HBe antibody measurements in Monoplex Serology based on RP Ia and Ib.

rRPIa = 0.98, rRPIb = 0.99.

Refstat: reference status.


S2 Fig.

Anti-HBs measurements stratified by reference serum status based on RP Va (Panels A-C) and Vb (Architect) and Vc (Centaur) (Panel D). Except for the GST-fusion antigen (A) all antigens were obtained from external providers and covalently coupled to polystyrene beads. B: Recombinantly expressed commercial HBs antigen expressed from Saccharomyces cerevisiae. In panels C and D, covalently coupled HBs VLPs were tested on three reference panels (Va-c) from different reference laboratories. Reference panel Va was tested based on the same reference assay as Vb, albeit in different laboratories. Reference panels Vb and Vc were tested by different reference assays, but in the same laboratory.

VLP: virus-like particle.

S.c. Saccharomyces cerevisiae.


S3 Fig. Graphical illustration of different criteria for overall HCV seropositivity based on antigens NS3 and Core.

Red lines: optimum cut-offs for overall HCV seropositivity defined by NS3 AND/OR Core seropositivity

Green lines: optimum cut-offs for overall HCV seropositivity defined by NS3 AND Core seropositivity.

Refstat: reference status.


S4 Fig. Correlation of measured HTLV-1 reactivities with HIV-1 reactivities based on homologous antigens gag, env, tax (tat) and rex (rev) on RP III.

No correlation could be detected. rgag = 0.005, renv = 0.09, rtax = 0.07, rrex = -0.01.


S5 Fig. Measured antibody responses against p22 and sag1 were plotted against each other and grouped by reference assay serostatus (blue triangle for positive, black box for negative).

In red, optimized cut-offs are shown.



The authors would like to thank Prof. Dr. Claus-Thomas Bock, Prof. Dr. Uwe Groß and Dr. Werner Bohne for providing DNA templates for antigen development. Furthermore, we thank Dr. Lena Kranz, Dr. Martina Willhauck-Fleckenstein, Dr. Bolorma Dondog, Dr. Kristina Prager, Claudia Brandel, Ute Koch and Monika Oppenländer for their input and excellent technical support and the UKB Infectious Disease Expert Panel: Prof. Allison Aiello, Prof. Judy Breuer, Dr. Michael Hill, Prof. Dr. Charles Bangham, Prof. Dr. Ray Borrow, Dr. Tim Brooks, Prof. Dr. Sir Rory Collins, Prof. Dr. Sir Brian Greenwood, Dr. Dominic Kelly, Dr. Vivek Naranbhai, Dr. Julian Parkhill, Dr. Richard Pebody, Prof. Dr. Tim Peto, Prof. Dr. Andrew Pollard, Prof. Dr. Robin Weiss, Dr. Denise Whitby, Dr. David Wyllie and Prof. Dr. Gary Clifford. We also thank Dr Pawan Dulal and Dr Florian Brod for their assistance in characterising the viral-like particle structure of the HBsAg.


  1. 1. Waterboer T, Sehr P, Michael KM, Franceschi S, Nieland JD, Joos TO, et al. Multiplex human papillomavirus serology based on in situ-purified glutathione s-transferase fusion proteins. Clinical chemistry. 2005 Oct;51:1845–1853. pmid:16099939
  2. 2. Michel A, Waterboer T, Kist M, Pawlita M. Helicobacter pylori multiplex serology. Helicobacter. 2009 Dec;14:525–535. pmid:19889070
  3. 3. Gossai A, Waterboer T, Nelson HH, Michel A, Willhauck-Fleckenstein M, Farzan SF, et al. Seroepidemiology of Human Polyomaviruses in a US Population. American journal of epidemiology. 2016 Jan;183:61–69. pmid:26667254
  4. 4. Karachaliou M, Waterboer T, Casabonne D, Chalkiadaki G, Roumeliotaki T, Michel A, et al. The Natural History of Human Polyomaviruses and Herpesviruses in Early Life–The Rhea Birth Cohort in Greece. American journal of epidemiology. 2016 Apr;183:671–679. pmid:26968942
  5. 5. King AM, Lefkowitz E, Adams MJ, Carstens EB. Virus taxonomy: ninth report of the International Committee on Taxonomy of Viruses. Elsevier; 2011.
  6. 6. Burns GS, Thompson AJ. Viral hepatitis B: clinical and epidemiological characteristics. Cold Spring Harbor perspectives in medicine. 2014 Oct;4:a024935. pmid:25359547
  7. 7. Taherkhani R, Farshadpour F. Global elimination of hepatitis C virus infection: Progresses and the remaining challenges. World journal of hepatology. 2017 Nov;9:1239–1252. pmid:29312527
  8. 8. Edmunds WJ, Medley GF, Nokes DJ, Hall AJ, Whittle HC. The influence of age on the development of the hepatitis B carrier state. Proceedings Biological sciences. 1993 Aug;253:197–201. pmid:8397416
  9. 9. McMahon BJ, Alward WL, Hall DB, Heyward WL, Bender TR, Francis DP, et al. Acute hepatitis B virus infection: relation of age to the clinical expression of disease and subsequent development of the carrier state. The Journal of infectious diseases. 1985 Apr;151:599–603. pmid:3973412
  10. 10. Poynard T, Yuen MF, Ratziu V, Lai CL. Viral hepatitis C. Lancet (London, England). 2003 Dec;362:2095–2100.
  11. 11. Villano SA, Vlahov D, Nelson KE, Cohn S, Thomas DL. Persistence of viremia and the importance of long-term follow-up after acute hepatitis C infection. Hepatology (Baltimore, Md). 1999 Mar;29:908–914.
  12. 12. Alter MJ. Community acquired viral hepatitis B and C in the United States. Gut. 1993;34:S17–S19. pmid:8314488
  13. 13. IARC. IARC Monograph Volume 100B; 2012. Available from:
  14. 14. Bouvard V, Baan R, Straif K, Grosse Y, Secretan B, El Ghissassi F, et al. A review of human carcinogens–Part B: biological agents. The Lancet Oncology. 2009 Apr;10:321–322. pmid:19350698
  15. 15. Schweitzer A, Horn J, Mikolajczyk RT, Krause G, Ott JJ. Estimations of worldwide prevalence of chronic hepatitis B virus infection: a systematic review of data published between 1965 and 2013. Lancet (London, England). 2015 Oct;386:1546–1555.
  16. 16. Maucort-Boulch D, de Martel C, Franceschi S, Plummer M. Fraction and incidence of liver cancer attributable to hepatitis B and C viruses worldwide. International journal of cancer. 2018 Jan;.
  17. 17. Gower E, Estes C, Blach S, Razavi-Shearer K, Razavi H. Global epidemiology and genotype distribution of the hepatitis C virus infection. Journal of hepatology. 2014 Nov;61:S45–S57. pmid:25086286
  18. 18. England PH. Hepatitis C in the UK 2017 report;. Available from:
  19. 19. Backus LI, Belperio PS, Shahoumian TA, Mole LA. Direct-acting antiviral sustained virologic response: Impact on mortality in patients without advanced liver disease. Hepatology (Baltimore, Md). 2018 Jan;.
  20. 20. of Experts for HBV PG, Flisiak R, Halota W, Jaroszewicz J, Juszczyk J, Ma Å, kowski P, et al. Recommendations for the treatment of hepatitis B in 2017. Clinical and experimental hepatology. 2017 Jun;3:35–46. pmid:28856289
  21. 21. Walker CM. Designing an HCV vaccine: a unique convergence of prevention and therapy? Current opinion in virology. 2017 Apr;23:113–119. pmid:28550816
  22. 22. Szmuness W, Stevens CE, Harley EJ, Zang EA, Oleszko WR, William DC, et al. Hepatitis B vaccine: demonstration of efficacy in a controlled clinical trial in a high-risk population in the United States. The New England journal of medicine. 1980 Oct;303:833–841. pmid:6997738
  23. 23. Davison SA, Strasser SI. Ordering and interpreting hepatitis B serology. BMJ (Clinical research ed). 2014 Apr;348:g2522.
  24. 24. Willems L, Hasegawa H, Accolla R, Bangham C, Bazarbachi A, Bertazzoni U, et al. Reducing the global burden of HTLV-1 infection: An agenda for research and action. Antiviral research. 2017 Jan;137:41–48. pmid:27840202
  25. 25. Osame M, Usuku K, Izumo S, Ijichi N, Amitani H, Igata A, et al. HTLV-I associated myelopathy, a new clinical entity. Lancet (London, England). 1986 May;1:1031–1032.
  26. 26. Uchiyama T. Human T cell leukemia virus type I (HTLV-I) and human diseases. Annual review of immunology. 1997;15:15–37. pmid:9143680
  27. 27. Kaplan JE, Osame M, Kubota H, Igata A, Nishitani H, Maeda Y, et al. The risk of development of HTLV-I-associated myelopathy/tropical spastic paraparesis among persons infected with HTLV-I. Journal of acquired immune deficiency syndromes. 1990;3:1096–1101. pmid:2213510
  28. 28. Tosswill JH, Taylor GP, Tedder RS, Mortimer PP. HTLV-I/II associated disease in England and Wales, 1993–7: retrospective review of serology requests. BMJ (Clinical research ed). 2000 Mar;320:611–612.
  29. 29. Chemoh W, Sawangjaroen N, Nissapatorn V, Suwanrath C, Chandeying V, Hortiwakul T, et al. Toxoplasma gondii infection: What is the real situation? Experimental parasitology. 2013 Dec;135:685–689. pmid:24140597
  30. 30. Bowie WR, King AS, Werker DH, Isaac-Renton JL, Bell A, Eng SB, et al. Outbreak of toxoplasmosis associated with municipal drinking water. The BC Toxoplasma Investigation Team. Lancet (London, England). 1997 Jul;350:173–177.
  31. 31. Rostami A, Karanis P, Fallahi S. Advances in serological, imaging techniques and molecular diagnosis of Toxoplasma gondii infection. Infection. 2018 Jan;.
  32. 32. Torrey EF, Yolken RH. Toxoplasma gondii and schizophrenia. Emerging infectious diseases. 2003 Nov;9:1375–1380. pmid:14725265
  33. 33. Martinez VO, de Mendonça Lima FW, de Carvalho CF, Menezes-Filho JA. Toxoplasma gondii infection and behavioral outcomes in humans: a systematic review. Parasitology research. 2018 Aug;.
  34. 34. Halonen SK, Weiss LM. Toxoplasmosis. Handbook of clinical neurology. 2013;114:125–145. pmid:23829904
  35. 35. Centers for Disease Control and Prevention. Parasites—Toxoplasmosis (Toxoplasma infection); Resources for Health Professionals;. Available from:
  36. 36. Pappas G, Roussos N, Falagas ME. Toxoplasmosis snapshots: global status of Toxoplasma gondii seroprevalence and implications for pregnancy and congenital toxoplasmosis. International journal for parasitology. 2009 Oct;39:1385–1394. pmid:19433092
  37. 37. Sehr P, Zumbach K, Pawlita M. A generic capture ELISA for recombinant proteins fused to glutathione S-transferase: validation for HPV serology. Journal of immunological methods. 2001 Jul;253:153–162. pmid:11384677
  38. 38. Dondog B, Schnitzler P, Michael KM, Clifford G, Franceschi S, Pawlita M, et al. Hepatitis C Virus Seroprevalence in Mongolian Women Assessed by a Novel Multiplex Antibody Detection Assay. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2015 Sep;24:1360–1365.
  39. 39. Bock CT, Malek NP, Tillmann HL, Manns MP, Trautwein C. The enhancer I core region contributes to the replication level of hepatitis B virus in vivo and in vitro. Journal of virology. 2000 Mar;74:2193–2202. pmid:10666249
  40. 40. Wu C, Zhang X, Tian Y, Song J, Yang D, Roggendorf M, et al. Biological significance of amino acid substitutions in hepatitis B surface antigen (HBsAg) for glycosylation, secretion, antigenicity and immunogenicity of HBsAg and hepatitis B virus replication. The Journal of general virology. 2010 Feb;91:483–492. pmid:19812261
  41. 41. Rigsby P, Rijpkema S, Guy EC, Francis J, Das RG. Evaluation of a candidate international standard preparation for human anti-Toxoplasma immunoglobulin G. Journal of clinical microbiology. 2004 Nov;42:5133–5138. pmid:15528706
  42. 42. Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistic. Family medicine. 2005 May;37:360–363. pmid:15883903
  43. 43. Revelle W. psych: Procedures for Psychological, Psychometric, and Personality Research. Evanston, Illinois; 2018. R package version 1.8.4.; 2018. Available from:
  44. 44. Koo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. Journal of chiropractic medicine. 2016 Jun;15:155–163. pmid:27330520
  45. 45. Department of Health & Human Services CfDC, Prevention DoVH. Interpretation of Hepatitis B Serologic Test Results; 2005. Available from:
  46. 46. Rodríguez-López M, Riezu-Boj JI, Ruiz M, Berasain C, Civeira MP, Prieto J, et al. Immunogenicity of variable regions of hepatitis C virus proteins: selection and modification of peptide epitopes to assess hepatitis C virus genotypes by ELISA. The Journal of general virology. 1999 Mar;80 (Pt 3):727–738.
  47. 47. Suemori K, Fujiwara H, Ochi T, Ogawa T, Matsuoka M, Matsumoto T, et al. HBZ is an immunogenic protein, but not a target antigen for human T-cell leukemia virus type 1-specific cytotoxic T lymphocytes. The Journal of general virology. 2009 Aug;90:1806–1811. pmid:19423550
  48. 48. Kuo CWS, Mirsaliotis A, Brighty DW. Antibodies to the envelope glycoprotein of human T cell leukemia virus type 1 robustly activate cell-mediated cytotoxic responses and directly neutralize viral infectivity at multiple steps of the entry process. Journal of immunology (Baltimore, Md: 1950). 2011 Jul;187:361–371.
  49. 49. Dekaban GA, King EE, Arp J, Palker TJ, Rice GP. Comparative analysis of the antibody response to the HTLV-I gag and env proteins in HTLV-I asymptomatic carriers and HAM/TSP patients: an isotype and subclass analysis. Scandinavian journal of immunology. 1994 Aug;40:171–180. pmid:8047838
  50. 50. Chen YM, Chen SH, Fu CY, Chen JY, Osame M. Antibody reactivities to tumor-suppressor protein p53 and HTLV-I Tof, Rex and Tax in HTLV-I-infected people with differing clinical status. International journal of cancer. 1997 Apr;71:196–202. pmid:9139842
  51. 51. Siachoque H, Guzman F, Burgos J, Patarroyo ME, Gomez Marin JE. Toxoplasma gondii: immunogenicity and protection by P30 peptides in a murine model. Experimental parasitology. 2006 Sep;114:62–65. pmid:16545806
  52. 52. Fang R, Feng H, Nie H, Wang L, Tu P, Song Q, et al. Construction and immunogenicity of pseudotype baculovirus expressing Toxoplasma gondii SAG1 protein in BALB/c mice model. Vaccine. 2010 Feb;28:1803–1807. pmid:20018269
  53. 53. Ganem D, Prince AM. Hepatitis B virus infection–natural history and clinical consequences. The New England journal of medicine. 2004 Mar;350:1118–1129. pmid:15014185
  54. 54. ABBOTT Diagnostics Division, Anti-HCV; 2009. Available from:
  55. 55. Diasorin. Murex anti-HCV;. Available from:
  56. 56. Division AD. HCV version 3.0;. Available from:
  57. 57. Couvreur G, Sadak A, Fortier B, Dubremetz JF. Surface antigens of Toxoplasma gondii. Parasitology. 1988 Aug;97 (Pt 1):1–10.
  58. 58. Lau YL, Fong MY. Toxoplasma gondii: serological characterization and immunogenicity of recombinant surface antigen 2 (SAG2) expressed in the yeast Pichia pastoris. Experimental parasitology. 2008 Jul;119:373–378. pmid:18457835
  59. 59. Bassig BA, Willhauck-Fleckenstein M, Shu XO, Koh WP, Gao YT, Purdue MP, et al. Serologic markers of viral infection and risk of non-Hodgkin lymphoma: A pooled study of three prospective cohorts in China and Singapore. International journal of cancer. 2018 Mar;.
  60. 60. Clifford GM, Waterboer T, Dondog B, Qiao YL, Kordzaia D, Hammouda D, et al. Hepatitis C virus seroprevalence in the general female population of 9 countries in Europe, Asia and Africa. Infectious agents and cancer. 2017;12:9. pmid:28168002
  61. 61. Burnett-Hartman AN, Newcomb PA, Schwartz SM, Bostick RM, Pawlita M, Waterboer T, et al. No association between antibodies to sexually transmitted infections and colorectal hyperplastic polyps in men: Minnesota Cancer Prevention Research Unit Polyp Study. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2012 Sep;21:1599–1601.