Contribution of serology in congenital toxoplasmosis diagnosis: results from a 10-year French retrospective study

ABSTRACT This study aimed to evaluate different serological strategies for the postnatal diagnosis of congenital toxoplasmosis (CT) and establish a biological algorithm for CT diagnosis. The study analyzed serological data of immunoglobulins M, A, and G (IgM, IgA, IgG) performed by immunoenzymatic and compared immunological profile (CIP) assays in 668 newborns with CT diagnosis across four testing periods: P1 (D0– D10), P2 (D11–D35), P3 (D36–D45), and P4 (>D45). Forty-nine percent of the 668 CT cases were diagnosed during P1 and 34%, 4%, and 12% during P2, P3, and P4, respectively. CIP assays detected neosynthetized IgMs/IgGs in 98% of CT cases diagnosed during P1, while IgMs and IgAs were detected in 90% and 57% of CT cases diagnosed during P2 and in 88% and 67% of diagnoses made during P3, respectively. Detection of neosynthesized IgMs/IgGs, IgMs, and IgAs by immunoassay contributed to CT diagnosis in 81%, 77%, and 60% of cases, respectively. In total, 46% of serum samples were positive for all three parameters, 27% for two, and 27% for one of the three. The study recommends using the CIP assay as standard during P1 for CT diagnosis and IgM and IgA immunoassays after P1. A clinical and biological follow-up in a specialized center with a close collaboration between biologists and clinicians is highly recommended to increase the chances of early diagnosis. Overall, this study provides useful information for the development of a biological algorithm for CT diagnosis, which can aid in early detection and appropriate treatment of this disease.

Our objective in this study was to evaluate the efficacy and contribution of the different serological assays to CT diagnosis during the postnatal period when the kinetics alone of IgG does not allow to confirm the diagnosis of CT and to propose standardiza tion of the management of newborns with suspected CT.

Serological parameters studied
The efficacy of IgMs/IgAs detection assays and CIP techniques was analyzed based on the sample that led to the CT diagnosis, depending on the time period in which the analysis was performed.Four periods were defined: the P1 period (10 first days of life of the newborn), the P2 period (11th-35th day), the P3 period (36th-45th day), and the P4 period (46th-90th day).

Exclusion criteria
Analysis reports were excluded from the study when the following information was missing: age of the newborn at sampling time, type of serological assay, or in the case of incomplete serological results.Cases diagnosed antenatally were excluded.Analysis reports in which the neonatal molecular biological diagnosis was positive were also excluded if no sera were collected at the same time as initiated treatments could have interfered with the serological results (11,12).Finally, the reports of newborns whose CT diagnosis had been established solely based on IgG kinetics using immunoenzymatic techniques were also excluded because they did not comply with the four periods defined.Finally, sera sampled after CT diagnoses were excluded from the analysis of the contribution of each parameter (positive rates and Venn analysis) regardless of the periods as antiparasitic treatment could interfere with the following serological results.

Study population
A total of 2,203 CT records were collected during 10 years from 27 centers (all experts in toxoplasmosis diagnosis), 1,022 of which were complete.Among these, 354 were excluded.Overall, 668 records of CT from 23 centers were analyzed, consisting of 2,525 serum samples (Fig. 1).

Serological diagnosis of CT
The serological work-up included IgM/IgA detection by immunoanalysis and investiga tion of any neosynthesized IgMs/IgGs using CIP assay.Toxoplasma testing was performed by all centers for IgMs and by 74% of the centers for IgAs.IgM and IgA were measured on the first three samples in 98% and 88% of cases, respectively.Reagents validated for newborns were used in 87% and 70% of assays, respectively.
Our analysis of the centers' different practices demonstrated that 87% of laborato ries performed CIP assays to detect neosynthesized IgMs and/or IgGs.The test was performed on the first serum sample (S1) in only 62% (414/668) of cases.Depending on the center, we noted four strategies for CIP tests realization: (i) one test performed at birth; (ii) one performed during P1 with a control test performed during P2 when necessary; (iii) one performed at birth with control tests performed during P2 and P3; (iv) only one CIP test performed during P1 or P2.

Practices analysis
On average, four serum samples (range: 1-21) were drawn from each of the 668 newborns during their follow-up.The first (S1) and second samples (S2) were taken during P1 in 87% and 61% of cases, respectively.Of note, 77% of S1 samples were collected within the first 3 days of life.The median follow-up time between S1 and S2 was 14 days compared to 30 days between S2 and S3.We also noticed one or more control serum samples were collected in 73% of patients once the CT diagnosis was established.
Finally, 6% (40/668) of CT diagnoses were established based on IgMs and/or IgAs detection during P1 with no serological confirmation of the positive result thereafter.

Follow-up frequency
Our study enabled us to show the wide variety in how different expert centers operate in terms of follow-up frequency during a newborn's first 3 months of life.Follow-up was most often structured based on the protocols established in each center.A majority of newborns (87%) were provided with at least one serological screening work-up during P1, and half of CT cases were diagnosed during this time.This early diagnosis is important because it allows for rapid treatment implementation.The sample frequency varied quite widely across the centers.Our study revealed that CT diagnosis was established during the first month of life for the majority of infants (83%), thus emphasiz ing how important it is to implement close neonatal and postnatal follow-ups.We also noted that 12% of CT cases were diagnosed late, i.e., during P4.For all of these cases, newborns had been lost to follow-up after being discharged and did not benefit from early serological work-up, hence the late diagnosis.By performing an initial sampling during the P1 period and a work-up during the following month, a majority of CT cases can, thus, be diagnosed.Furthermore, late diagnoses can be reduced by strengthening cohesion between laboratories, obstetric-gynecological and pediatric departments, and implementing postnatal diagnosis protocols.

Test efficacy and CT diagnostic contribution
For the centers where the CIP technique is available, this test is systematically carried out during the first month of life.However, a wide variety of strategies have been observed regarding the infant's age at the time of the first test and the frequency with which this test was performed.
CIP test greatly contributed diagnosis, providing positive results in 81% of cases and representing the only positive parameter in 19% of cases.Its value was found to be especially high during P1, with a positivity rate of 98%.When comparing infected and not infected population, previous study reported positivity rates from 33% to 79% (10,(13)(14)(15).During this period, it was the only parameter able to confirm CT diagnosis by revealing neosynthesized IgMs/IgGs (13)(14)(15).The positive immunoenzymatic tests from this period, for example, did not enable immunoglobulin originating from the mother to be distinguished from those synthesized by the newborn.
All the laboratory centers measured IgM levels and, in 87% of cases, they did so by using a technique validated for use in newborns under 1-year old (Toxo-ISAGA, BioMérieux and PlateliaTM Toxo, BioRad) (10).Performances for new CT diagnosis were identical when comparing P1 and P2 (88% and 91%, respectively).It was, on average, comparable to the sensitivity rates reported in previous studies that com pared eperformances on infected and not infected populations (76.9%-100%) (10,16,17,18,19).Moreover, 74% of the laboratories measured IgA levels, and in 70% of cases, they used a technique that had been validated for use in newborns.Positivity rates were almost identical when comparing P1 and P2 (69% and 80%, respectively), yet poorer than the sensitivity of the IgM tests (20).Nevertheless, during P1, newborn serum can be contaminated with the mother's serum, which could lead to false positive tests.Thus, IgM and IgA tests must be interpreted with precaution during this period.were less sensitive and contributed less to CT diagnosis than IgM and CIP tests, with rates of diagnosis contribution of 60%, 77%, and 81%, respectively (irrespective of time period).However, testing for IgAs was found to solely be positive in 2% of the samples for which the other two techniques (IgM tests and CIP) produced negative results.IgA testing can be a good complement to IgM and CIP tests, notably in cases of negative results from other work-ups (21).
We excluded 18 CT cases from the analysis due to a diagnosis made solely by IgG kinetics.For 8 of them, the serological follow-up was correct, but the tests were negative highlighting the risk of false negatives for these techniques.However, only 1-2 tests were performed on these cases, emphasizing the complementarity of these tests.
This study evaluated the contribution of the different serological parameters for CT diagnosis on infants with no positive antenatal screening.It is the first to include such a large number of CT cases allowing, thus, the determination of the most suitable paired serological parameters according to the post-natal periods.We could not evaluate formerly sensitivity and specificity as only infants with a CT diagnosis were included, yet we observed high positivity rates of each parameter.These good performances could be explained by a short exposure to maternal treatment as most infants included were contaminated during the third trimester of pregnancy.So, these results could, therefore, be particularly interesting in suspected CT when prenatal diagnosis was not performed.

Conclusion
This study enabled the CNR expert group to propose recommendations for postnatal CT diagnosis in case of negative or not performed prenatal diagnosis (Fig. 5).
Between day 0 and day 10, the CIP assay to detect IgM/IgG neosynthesis contributed the most to the CT diagnosis (98%) and should be standard for CT diagnosis.After day 10, testing for IgM and IgA is recommended and may be associated with CIP tests to detect the neosynthesis of IgMs/IgGs as well as testing for IgGs, which must be monitored monthly.Given the complexity of serological diagnosis, the expert group recommends clinical and biological follow-ups in a reference center, with collaboration between biologists, obstetricians, and pediatricians.This is of utmost importance to set up a close follow-up of the newborn whose mother seroconverted during pregnancy in order to detect and treat as soon as possible the CT.

FIG 5
FIG 5 Recommandations from National Reference Center to CT serological diagnosis on an infant with negative or no prenatal diagnosis.*Risk of contamination by maternal immunoglobulins.**IgG must be measured in parallel on successive sera with the same technique.