Background, what is known

Toxoplasma gondii, infects approximately half of all persons with 16 million people infected congenitally. Congenital toxoplasmosis (CT) causes loss of life, sight, cognitive and motor function [1–5]. In 2013 the World Health Organization (WHO) estimated there are up to 190,100 new cases of CT and 1.20 million disability-adjusted life years each year globally [4–7]. Disease burden is particularly high in Latin America and certain populations in the US and elsewhere with high exposure. Almost all untreated congenitally infected persons develop manifestations [1–17]. In addition to considerable progress toward definitive cure and prevention of Toxoplasma infection with novel potential medicines and vaccines [6,7], a critical part of eliminating the disease and reducing suffering and disease burden of CT requires prompt recognition of seroconversion and expeditious, early treatment of the acutely infected pregnant women with available, effective medicines [3,6–9]. Screening monthly, beginning before or near conception to one month post-partum for development of antibody to the parasite in previously seronegative women can enable treatment to prevent trans-placental transmission of newly acquired maternal Toxoplasma infection or treat the fetus to prevent sequelae [3,7–17]. France, Austria, Slovenia, Colombia, Panama, Brazil, Argentina and Morocco have or are developing screening programs [8,10,11] but the United States does/has not [3,6,7,8,12–14]. Actual and artificially inflated costs to make profit are potential barriers [1,13–45], even though cost benefit analyses all have found substantial cost savings and benefits with routine testing [13–18]. Introduction of prenatal screening tests that fulfill the WHO REASSURED criteria (Real-time connectivity, Ease of use, Affordable, Sensitive, Specific, User-friendly, Rapid, Robust, Equipment-free, Deliverable) can improve benefit [13–18]. False positive results using currently available commercial test kits for anti-Toxoplasma IgM compound problems [19–22,38]. In 1998, the United States Food and Drug Administration (FDA) mandated that a positive result for acute infection (IgM) with a non-reference laboratory (NRL) test should be confirmed at the Palo Alto Medical Foundation /Remington Specialty Toxoplasma Serology Laboratory (PAMF-TSL) [22]. However, this can and often does lead to associated delays, and causes concern for patients and their physicians. Substantial costs (more than $800 per panel of tests, which may require repeat testing, in the USA with additional problems for this exceptional testing from insurance denials and capitation of obstetrical health care) have been arguments against screening programs [14]. Therefore, NRL tests with high specificity and low cost are needed. A recently developed lateral immunochromatography test with antigen on beads that captures IgG and /or IgM antibodies, Toxoplasma ICT IgG-IgM test (LDBIO Diagnostic, Lyon, France, hereafter called ICT) is a promising candidate NRL test that satisfies REASSURED criteria [23–5]. Nonetheless, positive results influencing clinical care still require clinical reference laboratory confirmation.

Gaps in knowledge

Gaps include whether while meeting World Health Organization REASSURED criteria the ICT could achieve the specific regulatory requirements for European CE Mark in Europe and USA dual 510K FDA clearance and Clinical Laboratory Improvements Amendments (CLIA) waiver. For the dual FDA clearance and CLIA waiver, this process required at least 240 matrix tests of whole blood and serum with half from seropositive persons. It was essential that 1/3 of the data be obtained from the USA from at least 3 sites and 3 testers with at least 5 seropositive persons for each tester. Limit of detection and Quick Information studies with 9 testers to determine that this test could be easily utilized with written instructions by untrained professionals (not lay users) were other gaps. Other unanswered questions were determining safety and efficacy in practice at the point of care and in the laboratory, patient and provider satisfaction, applicability in Middle-Low income countries, utility in epidemiologic studies, and whether and how this could be integrated into clinical practice easily in the flow of usual clinical care. Whether the ICT could function comparably to current commercial and reference laboratory tests for those with low, mid-range and high antibody levels, or whether other illnesses interfered with the test also were critical questions to consider.

Hypotheses

Hypotheses were that studying the ICT would provide data that meet all the specific requirements for CE mark and dual 510 K FDA clearance and CLIA waiver while demonstrating that the ICT would have: 1) Real-time connectivity, Ease of specimen collection, and be Affordable, Sensitive, Specific, User friendly, Rapid, Robust, Equipment free, and Deliverable to end users. In meeting these REASSURED criteria, results would enable screening for acquisition of Toxoplasma gondii infection during gestation in a manner faster and more accurately than other commercially available tests; This testing would be: 2) less costly than other commercial tests; 3) preferred by patients and clinicians to other test protocols; and 4) useful for diagnosing infection in North and Latin America, Europe, and North Africa.

Objectives

In 12 sequential studies and analyses we aimed to learn how well the ICT met WHO REASSURED criteria and guidelines for achieving FDA clearance and CLIA waiver in the USA and CE Mark in Europe. This work also was designed to determine whether it is feasible to use the ICT, maintaining its high performance, with users in multiple clinical circumstances and settings. We also intended to determine acceptability of the use of the ICT to patients, families, and front-line health care personnel. In so doing we intended to learn how the ICT also might be used optimally in the clinic and in the laboratory for rapid back up testing to address whether false positive results in standard commercial tests might have occurred. If these criteria were met, this test could help to enable serologic screening to detect, diagnose and thereby promptly treat to prevent this infection and its adverse sequelae providing an improved paradigm for care to prevent or ameliorate the manifestations of this infection. When we encountered difficulties in the first 6 tests with false positive IgM results with the commercial, predicate comparator compared with gold standard tests in our feasibility clinical trial, but not the ICT, we were faced with the unanticipated constraint of cost of positive confirmation of a number of tests at PAMF-TSL, and recognition that this type of cost from the frequent false positive IgM results could de-rail screening programs in the United States. Given the true negative ICT results when compared with gold standard tests in our setting, we queried whether the ICT could be part of a paradigm to rule-out false positive IgM results with NRL test, both at the point-of-care and in the hospital clinical laboratory. Further, we tested samples with the ICT that were suspected false positives from local laboratories that had been referred to reference laboratories.

Implications of unexpected findings leading to next steps

Our unanticipated findings of false positive NRL commercial predicate tests led us to identify a solution for our clinical problem: We placed these data in the context of practical clinical problems we encountered and collated our results with ongoing and reported similar studies to define whether this could be an approach helpful in addressing false positive predicate NRL test results. Solving this problem emphasized how the ICT can be used in screening programs to benefit pregnant women and their families, creating a new model to approach the problem of need for accurate screening and of false positive NRL tests. This highly accurate test may help enable screening for acquisition of T. gondii in gestation [23–39], as well as being useful in other settings, and thereby contribute to saving sight, cognition, motor function and lives and improve quality of life [1–10,12–15,17,24,35].

Ethics

The ongoing UCMC study, under the name “Prevention of CT: Feasibility of prenatal screening using point-of-care tests,” is conducted with ethical standards for human experimentation established in the Declaration of Helsinki. Research received approval from the University of Chicago Institutional Review Board (University of Chicago IRB Protocols 20–0442, 19–0505, 21–1446, 8793, 8794, 8795, 8796, 8797, 8798, 16708A and met the standards of the Health Insurance Portability and Accountability Act. Results were or will be reviewed by/discussed with the Data Safety Monitoring Board. Informed consents were obtained from subjects in accordance with the University of Chicago Institutional Review Board and the National Institutes of Health guidelines. No subjects are under age 18 years in the University of Chicago Clinical Trial. All participants provided informed, written consent prior to their study participation. All studies were performed in accordance with the Declaration of Helsinki.

Samples from the Lyon Reference laboratory of the University hospital were anonymized in this analysis. Testing in Colombia was performed in accordance with local Ethics Committees approvals and guidelines.

Stored sera from Cincinnati was approved by the IRB for future testing for a wide range of pathogens.

Overall goals

These methods were structured to meet our initial specific objectives to determine the performance of the ICT. When we found high performance in initial studies objectives were then to carry out studies to assess whether the test met WHO REASSURED criteria for a point of care test, to meet all CE Mark, FDA and CLIA testing requirements, and to determine feasibility of implementation of use of ICT in multiple “real-life” clinical settings.

Design

We performed a series of studies with a lateral immunochromatography test (ICT) we previously found met WHO REASSURED criteria with serum and whole blood from a finger prick. This new work performed herein included prospective real time studies of feasibility and acceptability in the USA, France, and Colombia. In so doing, we discovered paradigm shifting approaches and utility, proving our original hypotheses and extending beyond these hypotheses. These studies used methods that follow. Box 1 lists and provides a “roadmap” to and milestones for 12 studies/analyses in this decades-long work. An overview of methods follows:

Study 1: Feasibility clinical trial study

Overall.

The design of this Study 1 (Clinical Trials.gov number NCT04474132) and how it is related to earlier work is shown in Fig 1A–1E. To begin to implement a reasonable, feasible, low cost workflow for USA gestational screening programs, which has currently and historically been problematic (please see S1 Commentary showing those problems for care), a series of 12 studies described herein were performed. A formal clinical trial feasibility study at the University of Chicago Medical Center (UCMC) began in July 2020. The goal was to evaluate a sufficient number of verifiable ICT results to complete the FDA (510(k) clearance and CLIA regulations waiver process. This study involves comparing results of the ICT to an already cleared serum test, also called predicate test (Bio-Rad Bioplex 2200 Toxo Enzyme Immunoassay). Serologic samples for the UCMC feasibility study 1 (ongoing as of February 2024) were obtained from 41 pregnant women, 40 undergoing regular prenatal appointments at the UCMC (23 in first trimester, 12 in second trimester, four in third-trimester) and from seventeen non-pregnant volunteers. Each participant’s whole blood and sera were tested with ICT; Participants’ sera were also tested at the UCMC’s CLIA-approved Clinical Laboratory, which uses an automated Bio-Rad Bioplex 2200 Toxo Enzyme Immunoassay as its FDA-cleared standard predicate test to detect IgG and IgM anti-Toxoplasma antibodies. For this ICT feasibility clinical trial, there were three testers in 3 different sites in the University. All discrepant results between ICT and predicate were sent to Remington Specialty Laboratory- PAMF-TSF or Lyon Reference laboratory for confirmation immediately using a panel of tests described elsewhere [27,37]. Specifically, as the FDA guidelines specified, this was carried out to determine whether this new ICT could be used successfully within formal, usually structured, clinical care systems in a University outpatient obstetrical and in-field practice settings. It was compared with the simultaneously performed FDA cleared, standard, predicate test. The ICT with whole blood was read by three readers independently and with sera by two independent readers also at the same times. Three different healthcare providers performed the point of care (POC) test. Each provider tested a minimum of five sero-positive persons. Results were entered into the University hospital EPIC medical record, standard clinical laboratory, and Clinical Research Redcap systems of the University of Chicago Hospitals with ICT results also documented by smart phone photograph.

Box 1. Roadmap and Milestones Create a Flow Chart of Studies, Analyses, Content, Supporting Figures, Boxes, and Tables.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Design, “Roadmap” and Context for Study 1 Performed herein.

Photograph of hands and kit from [23] open access license. Map produced in Carta. From [8]. Open Access License.

https://doi.org/10.1371/journal.pntd.0011335.g001

This was tested to determine whether this novel, simple test could perform well even during a time with the constraints posed by the SARS CoV-2 pandemic. As the FDA clearance, CLIA waiver and Institutional Review Board (IRB) processes specified must occur, this was a formal Clinical Trial included in Clinical Trials.gov NCT04474132, that would mimic ways the test would be used to identify serologic positivity or seroconversion during pregnancy in the University of Chicago Medicine practice or in other clinical settings that would be suitable to apply for dual FDA clearance and CLIA waiver. This was to formally determine whether this test could perform well and be feasible and easy to use within a clinical trial in a standard obstetrical and other care and research settings in accordance with FDA and CLIA requirements. We were expected to use a standard, comparator, “predicate” test that the FDA had already cleared for use with serum in the USA to detect and confirm presence of antibodies to T. gondii. There were only three such cleared tests in the USA and we selected the test our clinical laboratory utilized regularly, herein designated “predicate comparator test” so it could exactly replicate the academic practice setting in accordance with FDA/CLIA guidelines. The test cost was lowered for the clinical trial from $650 for the IgG and IgM clinical laboratory test to $13 to facilitate the study. The cost of the predicate testing was paid for by the Seed Fund Award from the Susan and Richard Kiphart Family Foundation, the A.K. Thrasher Children’s Charity and other charitable sources of funds. US Reference laboratory confirmation of test results could not be subsidized and remained over $1600 for two consecutive tests with a full “adult panel” when a positive IgM test was identified. Back up testing was performed for two patients in the Remington Toxoplasma Specialty Reference Laboratory and also was performed in the Lyon reference laboratory without charge using Abbott ELISA IgM/IgG. In August 2022 the Abbott IgG reagents also became FDA cleared.

Study 2a: Samples from Lyon Reference Laboratory that had been referred when reported/referred by local laboratories with erroneous positive IgM

A set of 32 samples obtained at the Parasitology Laboratory of the University Hospital of Lyon, France (Institut des agents infectieux, Hôpital de la Croix-Rousse, Lyon, France) were selected for being reported as false positive with at least one first-line, NRL automated assay and confirmed to be negative by a panel of additional tests in the laboratory including Abbott Architect G, M, Vidas G and Bio-Rad Platelia IgM ELISAs. They were in some cases additionally tested at LDBIO Diagnostics using ICT and WB ToxoII IgG which is a gold standard in France [40–45].

Study 2b. Testing of other samples at Hôpital Bichat, Paris

A total of 558 US serum samples that had been tested in a one week time period at the University of Chicago Medicine Clinical Laboratory, then stored for one week, that would otherwise be discarded were tested at Hôpital Bichat, Paris (Bichat-Claude Bernard Hôpital, Laboratory of Parasitologie, Paris, France [30]). Another set of French samples also was tested at Hôpital Bichat in Paris. Results are being presented in detail in a separate report describing a variety of tests from Hôpital Bichat [30]. Twenty-two false positive NRL IgG and 6 false positive NRL IgM samples were tested in this part of the study. Follow-up without seroconversion and other reference tests were used to confirm that positive results were false positives.

Study 3: Testing of US Samples in a study to Determine Feasibility and Acceptability of fingerstick in monthly US gestational screening program 2017–2022

Study and Analysis 4. Comparison of pink bead used earlier and black bead used in later studies and herein (Study 4) and Collation of Earlier Testing, Bibliographical search, and development of novel paradigm (Analysis 4)

Sample preparation

Samples were prepared in accordance with FDA/CLIA requirements and guidance for instructional material for CLIA waiver for a point of care test (Recommendations for Clinical Laboratory Improvement Amendments of 1988 [CLIA] Waiver Applications for Manufacturers of In Vitro Diagnostic Devices ([version of January 30, 2008 –in force and updated in 2020]). A serum sample that had T.gondii antibody, that was positive at the limits of detection for the Toxoplasma ICT IgG-IgM test was used as positive sample. The sample was prepared according to CLSI EP12-A2 User Protocol for Evaluation of Qualitative Test Performance guideline, with the objective of obtaining a sample that would be positive around 95% of the time (defined as positive between 36 to 39 times out of 40 trials, as per guideline instruction). The sample used was a citrate-based plasma obtained from the Etablissement Français du Sang (the French national blood bank) with both IgG and IgM for Toxoplasmosis (38 UI/ml of IgG and 63.89 ratio for IgM according to Roche Toxoplasma kits) prior dilution into seronegative heparinized blood obtained from venipuncture of a known seronegative person the day prior to dilution. The selected dilution was the 90^th (1:89 ratio) as the test was read positive 37 times by an untrained user and 39 by an experienced user at that dilution (LDBIO Diagnostics). Assuming linearity of dosage for the Roche kit, the IgG titer of the final sample was therefore 0.42 UI/ml, and 0.71 ratio for IgM, both below the threshold of 3 and 1 for the Roche IgG and IgM kit, respectively.

Knowing that this limit of detection was determined to occur when this positive serum was diluted 1:89 with whole blood, serum or plasma from a seronegative donor, the study proceeded as follows: Prior to being tested by nine testers, the whole blood from another seronegative donor in Chicago was confirmed to be negative with ICT from fingerstick. Thirty ml of whole blood was drawn from this second seronegative donor by venipuncture and placed into three ten ml tubes with EDTA anticoagulant and gently mixed by repeatedly inverting the tubes. This blood was divided into two parts, one part was to be used to create the negative samples and the second part was to be used to create the “limits of detection positive” sample. Half the blood from the seronegative donor was diluted 89:1 with the previously tested positive serum and gently mixed to create the “positive at the limits of detection whole blood sample”. The ability of an experienced tester to distinguish the negative and positive samples was tested with the ICT and was confirmed. Two hundred microliters of the negative whole blood was placed into each of sixty-three 1.5 ml Eppendorf tubes used in this study for the negative samples. Two hundred microliters of the “positive at the limits of detection whole blood sample” was placed in each of sixty-three 1.5 ml Eppendorf tubes. The samples were assigned and labeled with random numbers between 1 and 14. The code was known only to the scientist who diluted the samples and prepared the labeled tubes and created the unknowns to be tested. A set of unknown blood samples labeled from #1 to #14 was prepared for each of the nine testers along with ICT cassettes labeled with the number between 1 and 14. Each set had the testers initials on each cassette. Each tester had the same samples labeled 1 to 14. The limits of detection, “Quick Instructions” (“QI”) study was then performed as follows: Nine testers were identified:

Testers

The testers were asked to read and sign the informed consent if they concurred. They were asked to read the information (Quick information, QI) document and complete the top of the data sheet documenting they had read the QI and indicate if they believed they understood its content. They also answered questions about their educational and other ICT experience. The cassettes with initials and numbers along with capillary tubes, laboratory coat, gloves, and data sheet were provided in a work space separate from other testers and with guidance from the written instructions. Informed consent, and signature into a study log were obtained.

Testers were three practicing physicians, three nurse/nurse practitioners, two medical students and one medical resident. They were not experienced with this type of ICT using whole blood or this cassette. None worked as a certified laboratory technician. They were selected to reflect categories of potential users of this test who were unfamiliar with and unskilled with this test. Each tester took the University of Chicago blood-borne pathogens and universal precautions training to work with whole blood, and their competence in understanding and using this material currently was formally documented for the study, as was IRB-required. The ability of three groups of testers with different clinical roles to read the instructions, to perform the test in accordance with the instruction, and to distinguish negative and positive results at the pre-established limits of detection were evaluated.

Test and analysis

The ability of three groups of testers with different clinical roles to read the instructions, to perform the test in accordance with the instruction, and to distinguish negative and positive results at the pre-established limits of detection were evaluated. This was done to determine whether the instructional material was adequate to teach unskilled users in three different practice type settings. Familiarity with universal precautions and biosafety instruction precautions was documented as described above. The fourteen samples, half positive and half negative were tested by each tester in a blinded manner. The tester and one other reader read the tests. The testers read the cassette at 20–30 minutes after sample followed by buffer were applied to the cassette, and recorded their interpretations on a data sheet. Photographs of the cassettes were taken using smart phones with photography in an area with some natural light. The photographs were provided to the tester to read in a blinded manner. One of the testers was a second reader for 7 of the photograph sets and a laboratory scientist was the photograph reader for the remaining two tester’s photographs including the tester who was the photographer and second reader for the first 7 testers/readers. A skilled physician scientist with familiarity with the test was a third independent photograph reader. The tester, another unskilled reader (other than the laboratory reader who was experienced), and experienced reader all independently read the photographs of the cassette in a blinded manner and their results were recorded. Data from the data sheet then were analyzed.

Study 1, Feasibility clinical trial study performed exactly as the test would be used in practice, 2020 to 2021, demonstrates feasibility, identifies false positive predicate test results and develops new paradigm to help to obviate that problem

A flow chart showing the design and historical context of this study is in Fig 1.

Individual results in the ICT and the predicate test in this ongoing clinical trial study in the context of this design with each of the testers represented by a different color or symbol are in Fig 2A and Table 1 and Table A in S1 Commentary. In addition to Study 1, Table 1 summarizes Studies 2, 3, and 4. Study 4 detail is also in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. High performance of ICT. A. Clinical Feasibility Trial.

Solid symbols represent positive predicate test, open circles represent negative predicate test. Small black dot indicates false positive predicate test. Solid colored symbols indicate a positive test. Open symbols indicate negative test result. Each of the three testers is indicated by a different color in this figure* False positive IgM predicate test for seropositive person. B. ICT is negative with 32 false positive standard tests in Lyon. C. ICT is negative with false positive standard tests in Paris, Blue IgG, green IgM. D. ADBIO, a USA test, suffers from false negative and positive IgM results. E. ICT detects acute infections with positive IgG and IgM in Colombia. F LOD Quick Instructions and Limit of Detection Study 6 Results. Box 3 contains Design and detail of methods and results. Quick instructions are able to teach 9 testers, 3 physicians, 3 medical students and 3 nurses to use test accurately with 7 samples that are negative and 7 samples at the limits of detection. Details of each study are also in Tables and Figures as follows in parentheses: Fig 2A (Study 1, Fig 1A–1E, Table 1 and A in S1 Commentary); B. (Study 2a, Table 1 and Table B in S1 Commentary); C. (Study 2b and Box 2); D. (Study 10, Table 5); E. (Study 7, Table 5); F. Box 3 and its legend provide detail for Fig 2F (Study 6). Limits of Detection and Quick Information, Box 3-Part A is the actual Quick Information (QI) provided, Photograph of hand and kit from [23] open access license. Box 3 Part B shows the study design, Box 3 Part C shows examples of test results for two testers, Box 3D shows the tabulated results of all the readers in real time and of the photographs). Box 3 Legend provides Study 6’s. structure and results. In this study we demonstrate the ease in learning to use a simple, carefully designed test description and that these specific instructions which were prepared in accordance with FDA and CLIA requirements and review (A), can function as an effective teaching tool (B, C). This study 6 is also pertinent to demonstrating repeatability and reproducibility of this test. In (B), whole blood was tested at the defined limits of detection of the sample, with 14 negative or positive samples. Positive or negative status is blinded for these 9 independent tester/readers plus additional blinded reader. For this study, these 9 testers were without other training with the ICT and worked in 3 different locations. (B, C). Performance was perfect for each of them after reading this page of instructions (B, C). This is consonant with extensive and rigorous testing of use of this ICT in a variety of USA settings, as well as in other countries in many settings with minimal instruction. Detail of objective, methods, data content of Study 6 also is in the accompanying Box. As outlined in the Box, Study 6 shown in 2F, demonstrates that it is easy for physicians, physicians in training, and nurses to learn to use a simple carefully designed description (Quick Information, QI) shown in (A). The QI is about use of the ICT. This study and its results demonstrate that this QI is an effective teaching tool. Also, we find that results with the ICT are robust, repeatable, reproducible, sensitive and specific in multiple settings. Box 2 and Tables 1, 2 and 3 and Box 3 and Perspectives section of Discussion, and Supplement (S1 Commentary) contain additional pertinent information and data.

https://doi.org/10.1371/journal.pntd.0011335.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Overview of Chicago formal feasibility implementation, Lyon referred false positives study, and Lyon Architect and Quindio Vidas G and M back up testing studies of Chicago monthly screening study, and comparison of red bead (pink line) and black bead (black line) test kits.

https://doi.org/10.1371/journal.pntd.0011335.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Study 4 shows validation of the black version of the kit in head-to-head comparisons with pink version studied earlier.

https://doi.org/10.1371/journal.pntd.0011335.t002

Between August 2020 and December 2021, we performed this prospective clinical feasibility trial, Study 1, in which 43 seronegative and 15 seropositive persons were tested with the ICT using whole blood and sera. The 58 sera were also tested with the predicate test used by The University of Chicago Medicine Clinical Serology laboratory, the Automated Bio-Rad assay. All negative results for the ICT and Bio-Rad predicate test were concordant. This is illustrated by the open symbols below the horizontal line in Fig 2A.

This is also shown in Table A in S1 Commentary by participant lines for each of the three testers (Maternal fetal medicine nurse, Obstetrician, Medicine Pediatric Infectious Diseases specialist, each indicated in Table A in S1 Commentary by separate colors matching the Study 1 design). Any ICT positive results resulted in testing of the sera in Reference Laboratories. Results obtained by all readers for the ICT were uniformly concordant with true positive results found in the Reference laboratories. The specificity, sensitivity, NPV, and PPV were 100%. AUC was ~1. Integration into workflow and acceptability were strong. However. a study related Bio-Rad Bioplex 2200 automated IgM EIA problem arose with this NRL predicate test.

Specifically, results of the ICT, as shown in Fig 2, always had at least two independent observers to insure objectivity in their interpretation as well as to avoid any possible bias. Readings were documented by photographs 20–30 minutes after applying the sample and the buffer immediately after the sample. RMc always read the results independently after initial readers and never influenced the readings of the initial readers. Results of all readers were uniformly concordant. Any positive results were re-tested in Reference Laboratories.

As specified by the FDA, and as shown in Fig 2A and Table A in S1 Commentary the three testers of different professional backgrounds who might be among the categories of users of the test included an obstetrician, an infectious diseases specialist, and an obstetrics maternal fetal medicine nurse. All who tested participants were blinded to the participants serologic status. As shown in Fig 2A and Table 2A, each tester tested at least 5 persons who turned out to be seropositive and at least 5 persons who turned out to be seronegative. The different sites for testing in Chicago shown in Fig 2A and Table 1A included triage area and nurses examining area, the obstetricians examining area/office and sometimes in outdoor field settings during the SARS CoV-2 pandemic. Feasibility, accuracy, and one hundred percent correlation between the whole blood point of care test, the ICT with serum performed by a laboratory member, and a predicate and when needed gold standard test were identified.

Practical problems with the predicate test and USA health care system that arose in this USA Clinical Trial feasibility Study 1 demonstrate a compelling need for novel, practical solutions to enable gestational screening to be implemented into the workflow of USA obstetrical practices

Within testing the initial 6 pregnant participants, we encountered false positive results for two participants in the IgM predicate test and three others later revealing a false positivity rate of 10%. Contacting Bio-Rad about their cut off value for a positive test, re-standardization of the automated machine, checking all reagents did not correct these false positive values above the cut off requiring reporting of a positive predicate test in this practical setting. Further, finding these frequent false positive predicate test results was disturbing for patient participants, providers, and investigators. Correcting the erroneous predicate test data in this Study 1 with follow up gold-standard testing was time consuming, costly, and would result in delays in care in a real-world setting. The prior 1998 FDA advisory was cautionary but the confounding false positives were initially unexpected. This was because we had earlier used US and French reference laboratory backup testing. We had incorrectly considered that the FDA clearance of the predicate test that we were required to use for this clinical trial would have addressed and solved the problem of false positives.

In this real-life practical model in Study 1 that the IRB and FDA required the resource Reference laboratories in the USA and France provided high quality results providing an excellent solution to the problem of the false positive results, we encountered but they had associated substantial delays, cost and inconvenience. In this Study 1 to define practical functioning of the ICT, this problem with false positives occurred while we as investigators promptly saw the negative ICT result in whole blood and sera testing. We knew the results in the context of our earlier data showed very high performance, sensitivity, and specificity of the ICT with Reference laboratory back up. Our earlier work had demonstrated all negative results were accurate with the ICT. NRL erroneous predicate test results considered to be positive occurred earlier when samples were referred to the USA reference laboratory (27). This occurred when 33 patients with 60 sera were tested in our earlier work [27]. As our Study 1 enrollment reached the first 6 participants with 2 false positives identified by the US and Lyon France Reference laboratories, we recognized that if we could not solve this problem of false positive predicate tests, this de-railed the research study and its longer-term goal of proper testing in systematic gestational screening programs for the USA, and as a model for other countries. For the USA, along with their present high costs in the USA, false positive tests also are substantially harmful.

New paradigm elucidated by this experience

Thus, as we tested the whole blood and sera with the ICT in parallel with testing the sera with the Bio-Rad test, we developed an easy, inexpensive, paradigm shifting approach to solve this problem of false positive tests for the USA. This approach showed the ICT could help to eliminate the problem of false positives both in the clinic and the clinical laboratory. This model (Fig 3) was to have a method for diagnosis with a test that meets WHO REASSURED criteria available promptly at the time the test was performed and to have a first backup of positive results in serum rapidly in the local laboratory. Our experience shown demonstrates that this ICT performs properly in clinical practice and field studies. We noted that it could be used correctly by previously untrained professional observers, meeting WHO REASSURED criteria. We found that this also could help to obviate the difficulties caused when a commercial predicate test has false positive result. This was while introducing a novel test that could be low cost and easy to use in the clinic. It became clear from the experience and data presented that this novel test and paradigm could be a useful new method for the clinical laboratory to identify true positives rapidly for this emergent problem/disease. This could help to determine whether there was need for further screening. It could help to clarify whether there was need for emergent care with life, sight, cognition saving medicine that should be initiated promptly while waiting for backup reference laboratory confirmation of a suspected true positive test (Fig 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3.

A. Graphic showing monthly test method. Use of ICT in NRL as considered originally in and modified from our earlier work (33). Help refers to seeking expert guidance and reference laboratory assistance. B. Role of ICT in this algorithm shown diagrammatically. C. Representative example of ICT detecting very early seroconversion using sera originally stored for another purpose tested retrospectively, showing contrast with false positives. First pregnancy by IVF. The testing was all done retrospectively after positive IgG, and IgM at ~12–14 weeks gestation was discovered, At this time the ICT was an experimental research laboratory test done prior to studies performed for consideration of FDA clearance, CLIA waiver process and not for clinical care. However, they demonstrated, in this unusual circumstance, sensitivity early in this true seroconversion. This was while the Sabin Feldman Dye test that detects IgG directed against T.gondii and other Palo Alto Specialty laboratory tests including the IgM ELISAs were negative (antedating the time that they later became positive documenting seroconversion). Details with dates were as follows: December 2018. Began planning for IVF-sera Toxoplasma gondii antibody negative (Neg). March 2019. IVF embryos harvested cryopreserved-sera Toxoplasma gondii antibody negative. March 2019. Patient traveled to New Zealand returned end of March to USA. April 6- Patient sera obtained pre-transfer of embryos. Palo Alto reported negative g and m but there was for the first-time background in IgM ELISA, (~ 0.2, cut off ~2 for an adult), ENZO lab at NYU hospital -commercial lab reported T.gondii specific IgM positive. Since the commercial IgM was not considered reliable because of known false positives, the serum sample was sent to Palo Alto where this was found to be negative(neg). April 8 -Serum samples sent to University of Chicago Research laboratory, were tested retrospectively at a later time. ICT was weakly, but clearly positive *; Note this was pre-implantation of the embryo and IVF was in March before infection. This test is binary and not quantitative. April 9- Embryo transferred to woman. April 16-Sera hormones showed embryo implanted. ICT from this time was weakly positive when tested later, and was a little stronger than in the first sample. Five observers confirmed these readings. This was all with the same lot of the ICT. April 22- ICT a little stronger positive. Dye test 1.64, M~9, Avidity low, AC/HS acute. Subsequent sera also positive with significant rise in dye test titer and acute patterns for other tests. Pregnant woman began spiramycin at ~12 weeks gestation when the first IgG dye test, and IgM ELISA were positive. Other sera collected and tested retrospectively. Amniocentesis at 18 weeks, tested at Remington Specialty Laboratory. PCR was negative. All obstetrical ultrasounds were normal. Infant was uninfected. Abbreviations: neg is negative, + is positive, nd is not done. SFDT is Sabin Feldman Dye Test. IVF, in vitro fertilization. “wild screening” is a colloquial term referring to non-systematic testing.

https://doi.org/10.1371/journal.pntd.0011335.g003

Study 2 with additional testing of erroneous false positive local predicate tests with ICT and gold-standard testing in reference laboratories demonstrates utility of the novel approach with ICT

Then, Lyon and Paris Reference laboratories’ identified erroneous false positive results reported for samples referred for testing from local private laboratories that used commercial tests. Thirty-two samples that were referred to the Lyon reference laboratory from September 2021 to February 2022 from private laboratories because of the detection of isolated IgM in the course of monthly prenatal retesting, which the main system used in France in this context (Fig 2B and Table 1 and Table B in S1 Commentary): The tests that had been used included Cobas Roche (n = 21), Abbott (Architect n = 1 or Alinity n = 7), Siemens (n = 3) (Fig 2B and Table 1 and Table B in S1 Commentary). None of the additional 32 samples gave positive results with ICT or in the reference laboratory with Abbott Architect despite the erroneous reports of positive IgM results (Table 1). Further, in Lyon France, none of the 32 false positive IgM tests with the predicate local laboratory tests used had false positive test results with the ICT or gold standard Western Blot (Table 1 and Table B in S1 Commentary). This was using the same referred serum that was tested and reported to be positive from the local private laboratory. These results are included in Table 1 and Table B in S1 Commentary, and Box 2 and Fig 2.

Box 2. All Studies with ICT.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Box 2. (continued). Legend. All studies with ICT:

^a n this column, the numbers in the green font in parentheses indicate the Study number in this manuscript. In this column, the numbers in the blue font in brackets indicate the Reference number in the literature already published.

^b In this column, the numbers in the black font indicate data for ICT with serum. In this column, the numbers in red font indicate data for ICT with whole blood from finger prick.

^c For calculations: TP = true positives, FP = False positives, FN = False Negatives, TN = True Negatives; Sensitivity = probability that a test will give a correct answer, TP/(TP+FN) x 100; Specificity = TN/(TN+FP) x100; Positive Predictive Value, PPV = TP/(TP+FP) x 100; Negative Predictive Value, NPV = TN/(TN+FN) x 100; AUC = (sensitivity + specificity) divided by 2 (where sensitivity and specificity range from 0–1); NA, data not available.

*In Colombia, in this acceptability study, in this published manuscript, the comparison was between result in whole blood obtained by finger prick ICT and serum tested with ELECSYS-CLIA performed in clinical practice by the clinics. Total:783 (366; 417) whole blood, compared to results from 413 (199, 214) sera tested for those women who returned for serum test. Sensitivity was 96.3% and specificity was 95.9% for ELECYS IgG in this subset of sera. Serum was not tested with ICT. IgMs in sera were also tested with ELECSYS-CLIA IgM performed in clinical practice. Three discordant results between sera and whole blood samples were also tested with VIDAS ELFA G and M. Two acutely infected pregnant women were identified and treated. Using these tests and prior testing results available from earlier testing in clinical practice and follow up of the pregnant women and infants, persistent IgM from infections pre-conception and natural IgM antibodies also were identified. These data are not included in the tabulated overall data analyses because of the differences in study purposes, approaches and methodology, and the absence of serum ICT. Studies in Cincinnati and Panama/USA also had differences in purpose, without any or full confirmation with another Reference test method or substantial methodologic differences. Thus, they also could not be included in the overall calculations. The bolded total numbers are from those samples included in the overall calculations for sensitivity, specificity, NPV, PPV and AUC in Box 2. The performance in the USA hospitals/Reference Laboratory, including field settings, were slightly higher (~100%, 100%, 100%, 100%, 1) and also in the two French Reference laboratories. * In the south of France tabulated with Study 1, concordance for pink line, black line and Abbott Architect IgG and IgM for 774 sera and additional comparison of testing using pink-line and black-line kits for the same additional 300 samples, without knowledge of standard lab serologic results, was >99% (please see foot note to Table 2 for 2x2 Table). Sera from 44 persons who were seronegative for T.gondii with standard tests and who had malaria, had no false positives test results with ICT.

The following lists tests utilized which have links to full details in Google. All other than LDBIO tests CEMark with applications pending FDA review are CLIA approved (PAMF) or FDA cleared other than Architect IgM: Palo Alto Medical Foundation Remington Specialty Laboratory Adult and Infant panels, PCR, and Avidity; ABBOTT Architect IgG, IgM; Biomerieux Agglutination (IgG only), TXC(G and/or M), BIORAD PLATELIA ELISA IgG, IgM: BIORAD 2200 Bioplex ToRC; ELECYS ROCHE IgG, IgM; LDBIO IgG/IgM ICT; LDBIO Western Blot II IgG, IgM; LIASON IgG, IgM; VIDAS IgG, IgM, ELFA; VIDAS IGG IGM ELISA; with the manufacturers performance evaluations included. The use of these tests with high correlation with ICT is shown by region and test used in that region in the inset into this footnote that follows. The hand-written inset was written at a meeting with the FDA providing the instructions for feasibility efficacy studies in different settings with numbers of participants required for 510K “clearance” and CLIA “waiver” to be achieved.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Box2_Figure.

https://doi.org/10.1371/journal.pntd.0011335.g004

Data summarized in Box 2 and Table 3 place the results in the Clinical Trial and monthly screening acceptance studies (Study 3), in the context of other ongoing studies and our earlier published work. Box 2 and Table 3 address details of studies in the USA and elsewhere. Box 2 and Table 3 also collate and address studies of false positive IgMs referred to reference laboratories in the USA and France. Box 2 collates all these studies including those of other countries as a summary of all available results. Sensitivity, specificity, confidence intervals and details of studies are in Box 2 and, Table 3 and Tables A, B, and C in S1 Commentary. Performance of the ICT is high, sensitivity ≥ 97.5%, specificity ≥ 99.4% (serum and/or blood).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. ICT in US and false positives with comparators.

https://doi.org/10.1371/journal.pntd.0011335.t003

Additional testing of a set of samples with Architect, Vidas (Table C in S1 Commentary) and other tests was performed in an additional matrix analysis in Study 3 and is pertinent to consideration of false positive test results. Back up testing of another set of sera from a monthly screening and acceptability of the monthly screening program was performed in the Lyon France and Quindio Armenia Colombia Reference laboratories (Table C in S1 Commentary). In these reference laboratory settings as well as in the Paris Hôpital Bichat reference laboratory [30] false positive test results were less problematic (Fig 2C and Table C in S1 Commentary and Box 2, [30]) than in the clinical trial. There were no false positives in the Quindio laboratory Vidas testing and only one patient with multiple consecutive IgG false positives in the Lyon Architect tests.

Study 4 Comparing kits with pink and black bead and Analysis 4 placing US ICT in the context of other ongoing studies, including Study 3 and previously published studies, demonstrates high performance

One thousand seventy-four serum samples were tested with the pink (red bead) line and black line test kits and Abbott Architect in Lyon (N = 200; Wallon, Chapey), Marseille (N = 200; Lollivier), Ste Etienne (N = 374; Flori, Mahinc), LDBIO (N = 300; Piarroux, Limonne) with >99% concordance. Any discrepancies were resolved with Western blot. (Links to line data). This is part of the formal CE Mark, French Reference Laboratories, and FDA/CLIA file submissions of line data that also were reviewed in Chicago with concordance similarly documented (Table 2).

Analysis 4 Bibliographical search confirms high performance and that data analysis herein includes all published studies

As the ICT is now commercially available following CE Mark approval in Europe, we also used a bibliographic search to attempt to identify results with which we might have been unfamiliar or with inferior performance of the ICT. Box 2 details all studies performed to date: Number of persons (N), Sensitivity (Se)/ Specificity (Sp), country of samples, N of false positive IgM, ICT results on false positive IgM and confidence intervals, testing for serum and whole blood are in this Table. To date all studies have involved the authors of this manuscript. There were no additional studies identified that have been reported to date. Table 3A and 3B highlight the data from the USA where multiple sequential studies consistently showed very high performance of the ICT with both sera and blood from fingerstick, and studies addressing M positivity.

Overall, a total of 4967 sera, 1685 positive and 3282 negative, and 1093 whole blood tests, 362 positive and 731 negative tests have been performed, including all published, ongoing studies and those herein. Additional testing in an acceptability study in Colombia added another 413 whole blood paired with serum samples with high performance not included in the total. 4967 total sera (1685 positive and 3282 negative) and 1093 total (362 positive and 731 negative) whole blood finger prick samples were tested in matrix analyses correct for calculating sensitivity, specificity, NPV, PPV and AUC. There was high sensitivity and specificity, 99.2%/99.0% for serum and 97.5%/99.7% for whole blood (Box 2). Overall NPV is 99.7% (serum), 98.78% (whole blood), PPV is 97.5% (serum), 99.44% (whole blood). AUC is 1 for USA and 0.99 for serum and 0.99 for whole blood for all the tests calculated worldwide.

Study 2 a and b and earlier overall performance of ICT with NRL tests demonstrates that false positive IgM also is high

Overall, including our own results herein, we found 137 samples with false positive IgM in at least one NRL technique also tested with ICT, among which 132 were found negative in ICT. The specificity of ICT for false positive IgM was 96.4%. Three samples were from IgG negative pregnant persons in Chicago. Two seropositive persons also had false positive NRL IgM, was not found in Reference laboratory IgM tests. In addition 22 false positive IgG and later 6 false positive IgM results were correctly identified by ICT at Bichat-Claude Bernard Hôpital, Laboratory of Parasitologie, Paris, France [30] (Fig 2C and Box 2) and 5 false positive IgG identified in the predicate Abbot Architect in the University of Chicagomedicine samples tested in Institut des agents infectieux, Hôpital de la Croix-Rousse, Lyon, France (Tables 1 and B in S1 Commentary, and Box 2).

Study 5 comparing time, cost of tests and approaches demonstrates ICT time/cost savings and aids in eliminating delays, Approach and an analysis of relative time and cost is in Table 4. The ICT is substantially time and cost saving,

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Approach using ICT to gestational screening is cost and time-saving.

https://doi.org/10.1371/journal.pntd.0011335.t004

Representative case summaries illustrate problems in care that false positive tests can cause and utility in identifying seroconversion in infection acquired prior to conception, called study 4b and “analysis 11” in methods)

Box A in S1 Commentary has brief summaries of some consequences of false positive and negative results in ongoing cases in USA clinical practices. These provide further evidence of problems that false positive test results cause, and additional practical operational difficulties encountered in the USA.

Fig 3 (Study 4b) shows this new approach diagrammatically and its exquisite sensitivity and therefore utility in identifying infection prior to conception. In Fig 3C the ICT detected seroconversion a day earlier than the Sabin Feldman Dye test and IgM ELISA in the USA Reference laboratory. There are a number of examples of patients who developed M alone then M and G [26]. In the Mahinc et al study ([26], Box 2 and Table 3) there were 50 serum samples from 24 women for whom there were 17 samples with IgM only and 33 samples with IgM and IgG; ICT was positive for all samples except one that had a borderline IgM ISAGA of 5 for a patient who later was found to have acute Toxoplasma infection. There were also another 144 acutely infected persons identified in the USA, France, Morocco and Colombia all identified as positive with the ICT [24,25,27,29,30] Box 2 and Table 3. It was unusual, however, to watch seroconversion with as much precision in narrow time intervals so early in infection as shown in Fig 3C.

Box A and Tables A and B and Figs A and B in S1 Commentary contrast the current status and consequences of congenital toxoplasmosis at earlier times in France and continuing to the present in the USA. This illustrates that the ICT and gold standard back up testing can help to solve a substantial health care problem. This is both in a historical context and at present, with potential spillover benefit for care for pregnant women and their families.

Study 6, testing ability of written instructional materials to facilitate clinical use of the ICT by healthcare practitioners not skilled in using the ICT in a limit of detection quick information study per FDA and CLIA instructions, demonstrates high performance

Moving toward implementation, ability of written instructional material to be used in clinical practice with samples at limits of detection for positive whole blood samples and negative whole blood samples was found to have perfect performance. This perfect performance after they read the Quick Information (QI) materials was for all the 9 “blinded” readers and testers (3 nurses, nurse practitioners, resident and 2 medical students, and 3 practicing physicians) who were not previously trained to use this ICT (Fig 2F and Box 3). It was for all users and all samples when the cassette was evaluated in real time and when the smart phone photographs were evaluated for all 14 unknowns at the limits of detection samples (Fig 2F and Box 3). There was reproducible, replicable, robust, perfect performance in Study 6, with additional un-trained users reading the Quick information page of instructions as their learning material. In this setting, the ICT also worked perfectly with whole blood at the limits of antibody detection by the 9 users reading test kits in real time and photographs of those tests subsequently.

Box 3. presents detail for CLIA required Study 6 that has overview of results summarized in brief in Fig 2F: Overview and detail of objective, methods, data content, and conclusions of Study 6 performed in accordance with FDA and CLIA regulations

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Box3_Figure.

https://doi.org/10.1371/journal.pntd.0011335.g005

Box 3. Continued.

Legend for Box 3.

The Objective of Study 6 was to determine whether multiple relevant users could effectively learn from a “Quick Information (QI)”, single page with two sides, in accordance with FDA/CLIA regulations.

A.QI, regulatory processes, and participant eligibility. QI shown in A is not only an instruction page about how to perform a finger-prick but part of a specific required experiment germane to Federal regulations. These regulations require a study with experimental testing at the “cut off” limit of detection using this brief stand-alone QI (A). The FDA and CLIA required creation of and testing this QI in 2020, and iteratively reviewed the versions of this QI until it complied with all their guidelines. This review culminated in creation of (A) that was ready to test whether it could be an effective stand-alone learning tool for use of the ICT. Then multiple levels of IRB and research office approvals of this page and study were required. To comply with IRB approval, testers were asked to complete a suite of safety courses and tests providing information about blood borne pathogens and Toxoplasma. Certifications of completion of this course and testing had to be received by the IRB before beginning. Volunteer participants in 3 relevant categories of care providers who might use an ICT in the future were sought by word of mouth. No compensation was provided. An informed consent was signed. A log page with specific information and questions about eligibility was completed. (B) i.

Preparation of samples: A serum from a French blood bank was identified after testing the initial serum to know its IU. Then, in France the “cut off limit of detection” dilution was defined. This was a dilution that had to be found when tested to be incorrect one of a hundred times it was tested when characterized at multiple dilutions. Dilutions were tested 100 times each by an experienced (RP) and untrained user. A dilution of 1:89 was defined as the “cut off” dilution to be used to create the positive whole blood samples for the testers. Then, this serum was sent to the United States where the test had to be performed on 14 samples by each of the relevant US care provider personnel in the following study: Whole blood samples, either negative or the negative containing a 1:89 dilution of the French serum were prepared at the “cut off” dilution (called “positive”). Then the samples were aliquoted into separate tubes. Then a random number table was used to assign a number from 1 to 14 to each tube of blood. There was one set of numbered tubes for each tester. Each numbered tube contained either a negative or positive sample. Nine identical sets of tubes with the assigned numbers were prepared with one set for each tester.

B,C. i. Testing. After signing an informed consent pre-approved by the IRB (ethics committee), answering a set of questions concerning their background and confirming eligibility, being given a data sheet approved by the IRB, FDA/CLIA to complete, signing a log book, the testers each received a container with materials to use in a designated work space. In this container they received appropriate PPE, a barrier for the work space, a container to dispose their work materials, their set of samples in a small test tube rack, their own data sheet, a pen to record their results, a QI page, cassettes, buffer, marked capillary tubes, and a timer. They were then each able to begin testing these samples in sites approved for this testing and with appropriate PPE for handling human blood samples. The testers did not receive instructions other than their QI instructions. Testers were separated in time and space from each other and could not talk with or watch each other. When they completed their readings and recording of their results, a photographer documented the appearance of the cassettes at 20–30 minutes after loading the cassette with sample according to instructions. In this test, per FDA/CLIA instructions, there was no fingerstick performed. The testing first included two testers who checked the negative and prepared positive initially. They found discrimination of negative and positive samples was readily accomplished. Then, the blinded 9 testers, testing the replicate samples, performed the testing. This was to address who can learn to perform this test easily, reliably and how well with the QI, questions about reproducibility, repeatability, testing of learning and an objective, rigorous proof of results.

B, C. ii. Evaluation of photographs: The same blinded reading process of photographs provided to each tester also took place. The only person who knew code for labeling of tubes was the scientist who prepared them. No tester knew the code or what tubes contained.

D. Data and analysis: Concordance of tube contents and tester readings of cassettes in real time and photographs was 100%. This provides evidence that multiple users of various relevant backgrounds can learn to use this test with samples with IU at the established limit of detection. This is the US CLIA FDA process, and without it, this test cannot be available in the US. This testing demonstrated that physicians in training, in practice, and nurses can learn to use the ICT in 3 physically separate settings from reading the QI. This study also demonstrates reproducibility and repeatability of testing, in the results at the point of test with the ICT. It is also included to show the relevant user learning and testing in this mandated study, in addition to the QI showing how to do the finger-prick. Additional information in the new “Perspectives” section also presents data from over 1500 fingerstick tests with close to 100% accuracy in many diverse settings by many testers in 4 countries with varied backgrounds. In addition to this testing in Study 6, we include information about the wide variety of testers, settings, locations of testing, levels of education of testers.

In Conclusion, we demonstrate that it is easy for physicians, physicians in training, and nurses to learn to use a simple carefully designed description of use of the ICT and that this QI is an effective teaching tool. Also, we find that results with the ICT are robust, repeatable, reproducible, sensitive and specific in multiple settings. Box 2, Perspectives section of Discussion, and Supplement (S1 Commentary) contain additional pertinent information and data.

Study 3 demonstrates feasibility and acceptibility of monthly gestational screening with ICT

Overview.

Early in the work with the ICT in 2017 to 2018 we performed this study to determine whether monthly gestational screening would be feasible in a research study setting (Fig 4 and Table 1 and Table C in S1 Commentary).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Feasibility and acceptability of monthly screening in U.S. Academic obstetric practice.

Abbreviations in A. “neg” indicates negative results in tests. Note congruence of ICT and Reference Laboratory test conclusions. ^* Elective termination secondary to fetal anomalies, ^** Spontaneous abortion, ^*** Elected to leave the study after first test time due to traditional beliefs regarding having blood drawn. “md” indicates a missed appointment, as patient did not attend her regularly scheduled appointment. Came at alternative time and we did not connect for testing and/or survey. “n/a” indicates not available. “L&D” indicates that the patient’s monthly test was missed due to concern for premature labor, which resulted in a visit to the emergency department and then labor and delivery. “E” indicates an equivocal result, in which a barely visible band appeared which was not reproducible upon photographing the test. Per manufacturer instructions, this test was interpreted as negative. +Survey sent later, not included on graph as was not at initially planned time of six-week postpartum visit^a. A. Times and results of monthly screening for each participant. B. Survey questions and Likert scale. n the survey, number responses were considered as follows: strongly agree was 5, 4 was agree, 3 somewhat agree, 2 was somewhat disagree, 1 was strongly disagree C. Results of satisfaction survey for 14 participants. Each respondent’s answer is represented by the circles in the figure. The mean is indicated by the horizontal line, with error bars indicating standard deviation. In response to questions 1 and 4, 13/14 respondents indicated that they “strongly agree” that they would pursue testing for T. gondii in future pregnancies with POC testing, and that knowledge of T. gondii is important for pregnant women. Results were more mixed if testing required venipuncture, as indicated by the responses to questions 2 and 3, but most agreed that the test was important, would have it again in a subsequent pregnancy, and would recommend this to family and friends. It was noteworthy that other family members such as fathers of the fetus asked to be tested, relevant to the possibility of retinal disease. Right panel showed results after the time of the 6 week post-partum visit. `We did not conduct a formal survey at the time; however, we began with two providers, and other providers in the practice asked to join with their patients. Providers continued in the further analysis of the ICT as it moved toward clearance and waiver.

https://doi.org/10.1371/journal.pntd.0011335.g006

Results of the testing did not enter standard medical records or the EPIC system at that time and the testing took place earlier than Study 1 but was completed after that study. This study was initiated before the clinical trial and led to the initial meeting with the FDA when a program officer from the Thrasher Foundation emphasized the importance of FDA clearance for the test to be useful to help patients in the USA. We also asked participants at its completion whether these participants felt it was important and comfortable to have knowledge about Toxoplasma in pregnancy and whether they would want serologic testing and/or the finger stick point of care test in subsequent pregnancies if it were approved in the USA. The intent was to determine whether screening might be acceptable in standard academic obstetrical USA practice: Some parts of the study, e.g., the questionnaire and additional backup testing were performed in 2020. Results for the initial tests for the participants’ visits were included in the earlier 2018 publication [23]. Thus, numbers included for this study in the cumulative total of tests were subtracted from tests shown in Fig 4 and for this entry in Box 2.

Patient characteristics and testing details

Patients were all identified at between 8–12 weeks gestation. Patients had a median age of 31 years (range: 24–40 years). Seven of the 22 participants were nulliparous, while the remainder had been pregnant once or twice before. None reported having been tested for T. gondii infection in the past. Participants were enrolled in the study between September and November, 2017. The study initially concluded in September, 2018 with the birth of the last participant’s child and 6-week post-partum visit. Because five mothers were missed by our study group at their 6-week postpartum visit, an anonymized questionnaire was provided in 2022 for those participants. This was considered separately in our analyses. Patients were tested at monthly intervals after their initial enrollment and tested until their 6-week postpartum visit. A small subset of patients (3/22) were withdrawn from the study: One individual underwent elective termination due to fetal anomalies. One participant suffered a spontaneous abortion. The third patient chose to withdraw from the study due to traditional beliefs about dangers associated with venipuncture. No patients (0/22) had evidence of prior infection with T. gondii upon their initial testing with the whole blood POC test, and none seroconverted during gestation. One participant had a faint band in the ICT suggesting the possibility of a positive test on one test, but this was only visible to the naked eye and could not be independently confirmed with photography. Per manufacturer instructions, this test was interpreted as negative. There was 100% concordance between testing interpretations of the POC test and confirmatory testing, including the ARCHITECT/Vidas/Western blot systems and the serum-based POC test variant, commercially available and now CE mark approved in France. The course of gestational screening for each participant is presented in Fig 4 and Table C in S1 Commentary.

Patient perceptions of POC gestational screening are favorable

Initial response of patients and their families to educational materials screening was positive. No patient declined to participate. Patient participants asked whether other pregnant and non-pregnant family members and friends could join. Some fathers asked to be tested to know their own serologic status and if they might be at risk of retinal disease if infected. At the end of the consecutive screening tests, we administered the patient preferences survey to an available subset of the cohort (14 in total) at their 6-week postnatal visit. Two of those participating women who were missed completed the questionnaire in 2022. The POC testing and screening for acquisition of T. gondii in gestation was well received by all participants. In the post testing survey 100% strongly agreed with the value of education and testing and viewed the point of care fingerstick testing favorably. 100% agreed they would have testing in subsequent pregnancies and advise a friend or family member to do so. Testing by venipuncture and send out testing was viewed less favorably in contrast (p<0.05 at 6 weeks). Specifically, on average, patients appeared to prefer finger-stick POC screening to venipuncture, (POC- 4.93 ± 0.27 vs venipuncture- 4.00 ± 1.04) (Fig 4B and 4C). More than 100% (14/14) of enrolled patients “strongly agreed” or “agreed” with Question 1 that they would like to have POC testing in a future pregnancy if it involved ICT by fingerstick if this did not require monthly venipuncture. In contrast for Question 2, only 9 of 14 (64%) “strongly agreed” or “agreed” with screening with venipuncture (Chi squared equals 6.087 with 1 degree of freedom. The two-tailed P value equals 0.0136). 100% of patients “strongly agreed” that knowledge of T. gondii infection and ways to avoid exposure were important for pregnant women. The two participants who responded in 2022 also agreed that screening was worthwhile but only would want to do this by fingerstick and not venipuncture, a similar pattern. (Fig 4C right panel). In summary, in this post testing survey at 6 weeks post-partum or several years later, 100% (16 of 16) of participants “strongly agreed” or “agreed” with Question 1, having and recommending other pregnant women have finger stick ICT monthly during gestation. In contrast, for Question 2, 56% (9 of 16) of survey respondents “strongly agreed” or “agreed” with having venipuncture and send out testing in subsequent pregnancies and recommending this for other pregnant women (Chi squared equals 8.960 with 1 degree of freedom. The two-tailed P value equals 0.0028). All agreed that education and testing are valuable for obstetrical patients with their providers.

Provider perceptions of POC gestational screening are favorable

There was not a formal questionnaire for providers. Rather, level of interest and enthusiasm was reflected by the following: All providers remained involved in the study with their patients. Additional providers in the practice noting and learning about the ongoing study with the initial providers asked to join with their patients. Those still practicing at the University of Chicago at a later time collaborated in the subsequent clinical trial Study 1 presented herein. These objective measures demonstrate provider satisfaction. All providers indicated verbally that they found the finger-stick testing and monthly screening a constructive addition to their practice. The rapidity of obtaining the results was viewed positively.

In Studies 4b, 7 and 8, ICT detects early seroconversion and distinguishes additional seropositive and seronegative samples in USA, France and Colombia

We noted, as shown in Fig 3, the ability of the ICT to detect very early seroconversion in a study of sera obtained at narrow intervals to monitor hormone levels during in vitro fertilization (IVF), that happened to occur during very early seroconversion. This was a USA patient whose in vitro fertilization had occurred 6 months prior to implantation of their embryo at a time that she was seronegative. In the interval between in vitro fertilization and implantation she had traveled to New Zealand where she likely acquired T.gondii infection in the weeks before implantation as shown in Fig 3C.

Our earlier studies (Box 2 and Table 3) have all identified perfect performance in detecting sera from those with acute infection in the USA. It was, therefore, of interest to determine whether the same would be found in sera from patients with acute infection with the genetically different parasites found in Colombia. Fig 2E and Table 5 A-B shows perfect ability, sensitivity, and specificity of the ICT to also identify acute infection (IgG, IgM) in Colombia (N = 22) and those who are seronegative (N = 12), (p<0.0001)(sensitivity 100% and specificity 100%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Results of the studies in Colombia with LDbio, Vidas G and M, and AdBio.

https://doi.org/10.1371/journal.pntd.0011335.t005

This brought the total to 144, as above. Further, serologic status was correctly identified for additional NCCCTS participants tested with finger-stick whole blood and serum between March and December 2018 herein (N = 20 positive chronically infected persons [times after infection years were known to be greater than 17 years for all except 3 persons], and 5 negative). We found perfect correlation of testing of whole blood obtained by fingerstick and serum testing in the United States, herein, and almost perfect correlation in Morocco this ICT using whole blood accurately distinguishes seronegative and seropositive status as occurs in seroconversion. In study 6 in which we tested whole blood (that contained serum that originally had 38 UI/ml of IgG and 63.89 ratio for IgM according to Roche Toxoplasma kits, diluted 1:89 in whole blood from a seronegative donor) at the limits of detection in the “QI study”, there was high accuracy in distinguishing positive and negative samples (Fig 2F). There were N = 63 negative and 63 positive, making a total of 126 tests of samples performed. There was 100% accuracy both with the cassette and with photographs read by the tester and two additional readers. All readings were congruent and consistent. All were blinded for 9 testers with 3 testers in each of three settings (physician office, nurse health care setting, and laboratory conference room), and with testers differing professional backgrounds (3 nursing, 3 medicine in training, 3 licensed physicians in practice previously unskilled in use of such a test) (Fig 2F and Box 3).

In Study 9, use of ICT for Cincinnati epidemiology study between 2017 and 2019 demonstrates that ICT is an efficient way to perform such studies and that prevalence is low in Cincinnati

Of the 265 mothers tested, 8 (3%) had a positive IgG for Toxoplasma infection. None of these had a positive IgM. Variables of interest were available for 264 of the mothers including residential address (longitude and latitude), age, education, race, income and pet ownership as part of the original cohort study. There were no significant associations of testing positive for Toxoplasma infection and any of these variables (Fig 5).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Location of seropositive persons in Cincinnati and associated demographic factors such as socioeconomic status, maximum educational level achieved, pet ownership, and ethnicity.

Sera were collected between 2017 and 2019. The low prevalence of seropositivity did not allow testing for clustering by any known risk factors for infection, including proximity to watersheds associated with sewage water run-off. None of the sociodemographic parameters, neighborhood deprivation, nor residential latitude and longitude measures achieved statistical significance. The measures of neighborhood deprivation are indicated by the color of the symbols. This map is generated from OpenStreetMap(https://www.openstreetmap.org/search?query=cincinnati#map=12/39.1506/-84.5007). OpenStreetMap data is available under the Open Database License, compatible with CC BY 4.0 license.

https://doi.org/10.1371/journal.pntd.0011335.g007

In Study 10, results with ICT AdBio test (Onsite POC) that detects anti-T. gondii IgM and IgG separately has both substantial false negatives and false positive IgG (9%) and false negative IgM (18.5% true positives detected)

We had hoped that a test developed in the USA called ADBio that is purported to distinguish IgG and IgM specific for Toxoplasma might be useful in a field setting. This test had a high proportion of False negative and substantial number of False positive results (Fig 2D and Table 5). A total of 147 samples were tested using the Onsite POC test (Fig 2D and Table 5) in Quindio, Colombia. By using Toxoplasma-IgG detection by VIDAS as reference test (Fig 2D and Table 5), the sensitivity of Adbio for Toxoplasma-IgG detection was 100% (95% confidence interval [CI], 94.4%–100%), and specificity for Toxoplasma-IgG detection was 90.8% (95% confidence interval [CI], 80.3%–96.2%). Of note, in 6 of 65 (9%) samples, the Onsite POC kit tested positive for IgG but were negative by VIDAS IgG. In 5 of these sera the bands observed in AdBio were visible. One serum with positive result for IgG in the AdBio but negative in the IgG VIDAS, was obtained from a boy with confirmed congenital toxoplasmosis who was being treated when the sample was obtained. For diagnostic accuracy of IgM detection, sensitivity for Toxoplasma-IgM detection was 18.5% (95% confidence interval [CI], 7.0%–38.7%; 5 of 27), and specificity for Toxoplasma-IgM detection was 100% (95% confidence interval [CI], 96.1%–100%; 120 of 120). Summary of the diagnostic properties found in this group of sera are shown in Table 5, including NPV and PPV values.

Implications of main clinical findings In the 12 studies

Interpretation in view of the literature, and additional strengths and limitations

Implications for clinical laboratory practice

The ICT also should be very useful in clinical laboratories testing with sera with a potential false positive IgM result without IgG as described herein. It could function as a second-line test to confirm or find IgM specific for T. gondii is not present before sending the sample to a reference center, while continuing to follow the patient while awaiting Reference laboratory results. This is a major advance as this will save time and reduce the need for gold standard tests. It can help reduce concern for patients and physicians. When the ICT test is used initially with whole blood the only predicate test for confirmation needed will be if the whole blood test is positive. ICT not only helped to obviate the problems with false positives but also can result in detection of true positives and very early seroconversion as described herein and also recently acquired infections described elsewhere [9,23,24,26,27,30].

Implications for Public Health for Large Scale Gestational Screening Programs to reduce burden of congenital toxoplasmosis, to reduce the burden of eye disease from post-natal infections, and in a variety of other settings

We placed this work in the context of ongoing problems for healthcare (Figs A and B in S1 Commentary) and potential for direct and spillover benefit for the care of pregnant women and their families (Table A in S1 Commentary, [14]). We also placed these studies 1 to 12 herein in a historical context building parts of a toolbox working toward a role of screening using WHO REASSURED criteria compatible tests in the elimination of congenital toxoplasmosis (Figs A and B in S1 Commentary).

As considered above for individual clinical circumstances and in the local clinical laboratory there are similar additional public health considerations in the same categories. There also are a variety of other clinical and epidemiologic circumstances where knowing T. gondii serologic antibody status can be of considerable clinical and public health utility and benefit [1,2,6,7,27,32–9,46–50]. Very high-quality, low-cost screening tests such as the ICT can improve infectious diseases care in gestation, eliminating perinatal infections. There are a variety of public health threats to pregnant women from wildtype Toxoplasma as well as parasites that can be infected with single stranded RNA viruses in fresh and sea water, venison, worsening with heavy rainfall, endemic and epidemic. Value extends beyond care for toxoplasmosis to other aspects of health care for pregnant women and other clinical and research settings. Large scale screening programs reduce the burden of congenital toxoplasmosis for individual suffering and societal costs.

Implications and perspectives for further research

Congenital toxoplasmosis is a treatable and preventable disease, and physicians and other obstetrical providers now have the tools, in-hand, to improve outcomes and reduce patient and familial suffering. This screening, the standard of care in other countries, is now increasingly feasible in countries like the United States, where the primary argument against screening has been its economic burden. In the development of this test and other high-functioning point-of-care tests, there is potential for transformation in the provision of obstetrical care to improve maternal-child health. These benefits are amplified in subpopulation demographics in the USA [28] and regions of the world where the burden of disease is even higher. Examples of this occur in the Lancaster Amish population in the USA [8,12,32,46], parts of Florida, are likely in other US subpopulations [28], and occur in Central [36] and South America [47], and parts of Africa [29].

Use of the ICT for the Cincinnati maternal cohort study found ICT to be efficient (Study 9). Due to the small proportion who were seropositive, we were unable to test for any clustering by known risk factors for exposure: none of the individual socio-economic or location factors in a regression analysis achieved statistical significance. A larger overall sample size will be needed to evaluate risk factors in this population. Reasons for relatively low prevalence in Cincinnati in this cohort remain to be discovered. Even with the low prevalence found, it is likely still that gestational screening would be worthwhile.

Our recent study in Colombia also demonstrated high acceptability of a single use of the POC on a large scale of 783 women and 30 providers [47]. Although Toxoplasma infections occur in all demographics it was a particular problem in those who had lower education and socioeconomic status [31,47]. To understand risk factors during gestation and to develop programs to prevent such infection will require monthly screening in areas of high to low prevalence.

The implementation of this study in the clinical trial and the QI limit of detection study demonstrated that it should be easy to introduce this test into obstetrical or other practice taking little extra time or causing inconvenience. For example, when patients are evaluated for vital signs, blood pressure, glucose including by fingerstick, by a medical assistant or nurse, the test can be performed and the cassette can also be brought to the obstetrician or other health care practitioner for additional reading and entry into the medical record. Photography, using a smart phone for documentation, could easily be included for when the care provider requires assistance with interpretation and for additional documentation for the patient and incorporation into the medical record.

In France screening was mandated by law. In Austria those screened received additional health care benefits. In Colombia it was introduced through practice societies. In the USA those in advisory positions recommended that education, easy feasibility, low cost would result in those who would benefit choosing to have testing incorporated in medical practice and USA patient culture at many levels by personal preference. The acceptability study demonstrated that informed patients would want this and obstetricians could use this comfortably and without inconvenience in their practice. It could easily be introduced into family practice and adolescent pediatric care to identify seropositive patients at risk of this most common retina disease due to infection and loss of sight. Such screening in adolescence could also provide pre-pregnancy testing for young women to allow knowledge of who is seronegative and should be screened during pregnancy. Pre-marital/conception screening, as initially occurred in France, could also be helpful as families plan to have children. As Toxoplasma has been transmitted by organ donation and white blood cell transfusion and by sperm in domestic non-human animals, can relapse with immune suppression, and may be causative for epilepsy and some neurodegeneration, there are a number of other medical settings where knowledge of Toxoplasma serologic status may be useful.

A future possibility to obviate the potential limitation of use of saliva which does not perform well with ICT includes a test that might be used with saliva or serum or whole blood in conjunction with the ICT if developed further in the future. This nanogold NIRMIDAS test was studied with saliva, serum, and whole blood. It was found to have high sensitivity and specificity and dye test precision for the detection of IgG, and for IgM [37]. We had suggested earlier this might be an ideal test to use before conception or if cost were constrained and it was feasible for initial testing in gestation. Although finger stick for glucose is standard, easy, and familiar in obstetrical practices, obtaining saliva may be viewed as less difficult than whole blood. Thus, some might view testing using saliva a potential advantage. However, the nanogold for any sample, so far, would require transport to where the machine is, associated delays to reach a clinical laboratory, and electricity and a sophisticated machine for testing. Recently manufacture of this nanogold test was discontinued. NIRMIDAS has also used a gold bead ICT for SARS CoV-2, but nothing like this has been produced for Toxoplasma to date. The diagnosis and management of Toxoplasma infection requires a knowledgeable health care provider urgently making home testing of saliva less advantageous.

Another consideration for the future includes the evaluation of the potential for cost savings to individuals, health care providers and systems as well as limiting suffering for individual patients and their families. As a foundation for such analyses in a variety of settings, to understand if the ICT should be widely used in gestational screening programs, studies of cost efficacy were performed [14]. Initially, we, with economists, created a mathematical model which can be used for such analyses in various settings. We first demonstrated potential cost benefit using health care data in the US [14], which found a financial cost saving incentive for screening programs sponsored by government health care systems as took place in France. With this first adaptable mathematical model [14], we applied it to actual costs in Austria [15] and then in France [16,17] with economists Drs. Stillwaggon and Sawers. Cost/benefit was shown elegantly by Stillwaggon, Sawers, Prusa, and Kasper to occur even with low prevalence as in Austria. Gestational screening was found to be 14- fold cost saving for the health care system, as well as being good for people, saving life, sight, cognition and motor function in these countries [16,17]. The important study of Mandelbrot et al [3] demonstrated the benefit for improved gestation outcomes from prompt diagnosis and treatment, as a major consideration for a screening program. A well-functioning point of care test meeting WHO REASSURED criteria, adds time and cost savings as shown by our analyses herein. In the future this could be an additional cost savings and benefit consideration for gestational screening and treatment as well as in a variety of health care systems and in epidemiologic study.

In future studies, and perhaps ultimately in the future in clinical practice, the ICT also might be useful to demonstrate efficacy of candidate anti-parasitic agents [51]. This could be useful in establishing definitive cure of infection. This would be by demonstrating elimination of presence of any serum antibody to T. gondii with elimination of the three clinical forms of T. gondii (tachyzoites, bradyzoites in cysts, and “persister” organisms). This might be used to establish definitive cure in those treated in clinical trials, and with the possible goal of ICT use, ultimately, in clinical practice in this manner. It might be useful to establish vaccine efficacy [52]. The work herein provides part of the foundation for such studies. The ICT has also been utilized in studies of seroprevalence in wild-life and in characterizing epidemics taking place in zoos in primates and marsupials [53,54]. These latter studies are being performed in conjunction with studies of a new vaccine for zoo animals. In the future the ICT could be an efficient, low cost tool for one health and other epidemiologic studies of prevalence, and in epidemic as well as endemic settings.

Implications of additional considerations, and perspectives, concerning this novel test, paradigm, implementation, and practical use of this approach

The following information addresses some of these special considerations: For context, we emphasize that our initial specific overall objectives of this study were to determine the performance of the ICT which we found to be sensitive and specific as described herein. When we found high performance in initial studies, our objectives were to carry out a series of well-controlled studies to assess whether the test met WHO REASSURED criteria for a point of care test, could meet all FDA and CLIA testing requirements to enable its use in the United States and elsewhere, and to determine feasibility of implementation and acceptability of use of the ICT in multiple real-life clinical settings and in an epidemiologic study. These objectives were achieved successfully as outlined above. Nonetheless, even such a high performing test was not developed and does not function well in a vacuum. Rather, this is part of a tool box beginning with knowledge of the diseases Toxoplasma causes and their consequences, and thereby whether, how, when and where to use the test. Certain special considerations concerning ease of use, learning to use the test, who could use it successfully, feasibility, ease of implementation, reproducibility, repeatability, and acceptability became evident in our studies. Herein, in all our studies in Chicago, per IRB regulation, 100% of our results were confirmed, and in practical use any positive result was also confirmed. Reference laboratory back up must be performed for any positive ICT result during gestation, especially seroconversion. We do not advocate home testing by patients, but believe testing should be performed with availability of a well-educated, knowledgeable physicians’/care givers’ advice because of the substantial impact of this congenital infection on outcomes lifelong.

The FDA/CLIA regulations required multiple settings and multiple testers in these separate stepwise studies. The variety of settings for point of care finger prick for whole blood, as shown in the individual patient/participant tables varied and the results were robust in all. Among all our US studies herein, testers included those with a wide variety of relevant backgrounds and a wide variety of relevant settings listed in Box 2 legend. All US tests were read by the testers and additional medical and laboratory personnel, and readings were documented by photograph. All readings in the US were congruent among multiple readers who worked independently and who did not influence each-others’ readings of the test kit or photographs of the test kits. Readings were consistently congruent by varied testers in different settings. In the US, every person who had a whole blood reading had serum results. The earlier pink line serum test was found to be congruent with the black line whole blood and serum tests herein. Photographic documentation was performed with all US study readings in Chicago, and at Stanford earlier, with multiple congruent readings both for whole blood and for serum in Chicago. As we did in these studies, we suggest that the results influencing clinical care, for example seroconversion or a positive test during gestation, be confirmed in a reference laboratory, and also be documented by smart phone or other photograph.

Concerning education, the test kit brochure has a 24/7 telephone number where photos can be reviewed for assistance. We have created a You tube movie linked to reference [23], and an instructional test set of photos we will place on our website. We have created a suite of educational materials for various audiences (high schoolers, when considering pregnancy, during pregnancy, medical students, residents, fellows, practicing physicians, in text book chapters) [6,7]. Educational materials have improved knowledge in all those settings [32–5]. Concerning educational materials, families of affected children (some of whom are now adults) want to introduce information at multiple levels. This was to try to build a robust system of informed persons and call systems when physicians, other care givers, or patients need advice. In the US, in June 2023, eight states (Hawaii, Arkansas, Nebraska, Kansas, Kentucky, Pennsylvania, Minnesota, Wisconsin and Delaware) have made it the law to report toxoplasmosis. They are creating a knowledgeable cohort of CDC epidemiologists in those state programs. They have created uniform reporting criteria which were approved by the full Council of CDC epidemiologists nationally and in all 50 states and territories in June 2023. To help to enable screening in those settings as noted above, we have made educational materials freely available [32–5]. We view the ICT as part of a “toolbox” to implement high quality care to prevent toxoplasmosis, where education and expertise in interpretation and advice are critical. Tables 1, 2, 3A and 3B and Box 2 and Tables A, B and C in S1 Commentary also illustrate feasibility, and successful implementation of testing. Although ICT result informs initial approach to care, a positive result should be confirmed and in certain clinical circumstances such as during gestation with differentiation of acute from chronic infection is essential. Exceptions for need for confirmation include epidemiologic prevalence studies where the test is not influencing direct patient care.

The high performance of the ICT and development of this new paradigm for care (as shown in Fig 3 and Table 6) can help to improve outcomes for congenital toxoplasmosis substantially and contribute to its elimination. As noted above, this neglected disease causes a significant health burden with 190,000 infants or more infected each year causing 1.2 million disability adjusted life years. Obstetricians, nurse midwives, family practitioners, obstetrical nurses, and other obstetrical providers, neonatologists, pediatricians, infectious diseases specialists, policy makers, public health officials, and others are uniquely positioned to intervene to prevent this disease and improve the health of both mother and child. POC test based monthly gestational screening of seronegative patients for T. gondii infection provides a valuable tool in the obstetric armamentarium to ensure maternal-child wellness and lower its health-care burden around the world. Inclusion of POC-ICT into Toxoplasma diagnosis brings considerable promise.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Summary of Role of ICT and Novel Paradigm in Approach to Screening and Actions Taken.

https://doi.org/10.1371/journal.pntd.0011335.t006

This new paradigm is to have a test that meets WHO REASSURED standards available promptly at the time the test is performed and to have a first backup of positive results in serum rapidly in the local laboratory in the context of knowledgeable medical care and reference laboratory backup testing. Table 6 contains a summary of approaches to testing in screening programs and some of the ways in which results for this novel paradigm are actionable for care providers. This includes and is especially focused for the pregnant woman and her fetus. Those who are seronegative are tested monthly until 6 weeks post-partum. Pre-conception testing is obtained when possible. Testing for IgG and IgM is obtained for those who are seropositive, and for anyone who has IgM an avidity and AC/HS test is potentially helpful to date infection. The ICT is also potentially useful in a variety of other clinical settings. When such tests have undergone appropriate evaluation by the FDA and CLIA, as they have undergone in CE-mark evaluation and approval in Europe, this testing can enable substantial improvements in management of risks associated with exposure to T. gondii.