Human toxocariasis is associated with the migration and persistence of larvae of nematodes of the genus Toxocara, primarily T. canis and T. cati, in tissues. Clinical manifestations depending on the localization of the larvae and the host's response include visceral, ocular, and neurological lesions; a mild course of the infection is also possible. Since the parasite does not reach sexual maturity in the human body, detection of eggs in feces is not the method to confirm human toxocariasis. That is why serological methods are central for assessment of patients and studying the population's contact with the pathogen [1, 2].

The toxocariasis epidemiological significance has been confirmed by the meta-analysis of a total of 250 studies conducted in 71 countries: the global seropositivity estimate was 19.0%, with significant variations between regions and the groups surveyed [3]. The data for Kazakhstan also suggest the need for regional studies. In the earlier study conducted in Uralsk we assessed seropositivity and avidity of IgG in adults and infection rate in dogs; we also distinguished between the presence of antibodies in a human and indicators of a patent intestinal infection in the definitive host [4]. These findings characterize a different territory and do not allow for the direct assessment of the avidity profile of adults in Almaty.

The use of the Toxocara larval excretory/secretory antigens for the enzyme-linked immunosorbent assay (ELISA) was substantiated in the earlier studies of serological diagnosis [5]. However, the IgG positivity does not determine the duration and activity of infection. Immunoblot and recombinant antigens can increase the antibody detection specificity, but address a different task than assessing the infection stage [6]. In one study involving the long-term follow-up of children after treatment, the decrease in IgG titers was recorded significantly later than that in titers of immunoglobulins E and A (IgE and IgA), starting from the 4^th year of  follow-up [7]. These data suggest t that antibodies persist for a long time; however, the timeframe determined for children cannot be directly extrapolated to adults. Determination of immunoglobulin M (IgM) is also limited by cross-reactivity with Toxocara antigens in patients with other helminth infections and healthy individuals [8].

Avidity characterizes the overall strength of the multivalent interaction between antibodies and an antigen, and differs from the affinity of an individual antigen-binding site. In the modified ELISA, it is assessed based on the signal persistence following the exposure to the chaotropic agent weakening noncovalent interactions. Such an approach involving the use of the dissociating reagent has been reported [9]. The resulting index depends on the antigen, serum dilution, and processing regime; therefore, its limits should be evaluated for a specific test system [10].

Studies of toxocariasis confirm potential benefits of determining avidity but do not offer a unified scale for all tests. Some authors used a cutoff of around 40%, others compared relative avidity with the estimated duration of infection, while others proposed separate thresholds for supporting or ruling out active toxocariasis [11–13]. Persistence of low and borderline values for years reported in one of the papers limits interpretation of low avidity as a sign of recent infection [14]. Thus, further characterization of antibody binding is possible, but its clinical significance requires independent verification.

It is important to compare the avidity index (AI) with the strength of positive IgG antibody results when performing serological assessment of adults. Samples in which the optical density (OD) after treatment with the denaturing solution (DS) exceeds the OD of the paired well without DS should be considered separately. The cause of such a signal increase cannot be determined based on the calculated index only.

The study aimed to characterize the distribution of the avidity index of IgG against T. canis in seropositive adults assessed in Almaty in 2026 and evaluate its association with the strength of positive ELISA results and the specifics of OD changes following the denaturing treatment. The study aimed at applying the avidity analysis to the serological sample and assessing the limits of interpretation of the results obtained.

METHODS

Study design

The cross-sectional serological study of adult volunteers assessed in Almaty, Kazakhstan, in April-May 2026 was conducted. Venous blood sampling was performed by the medical staff of the Fenix Med LLP (Almaty, Kazakhstan), laboratory testing was  performed at the Scientific Research and Production Center “MVA GROUP” LLP (Almaty Region, Kazakhstan). A consecutive convenience sample was compiled: all the individuals, who met the criteria, visited the sampling point, and agreed to participate, were enrolled to reach the target size of 500 people. The sample was not considered as random or representative of the entire adult population of Almaty.

Inclusion criteria: age ≥18 years, living in Almaty, written informed consent, completed questionnaire, venous blood sample suitable for analysis, no use of anthelmintic drugs during the preceding three months. Exclusion criteria: age under 18 years, inability or refusal to provide the informed consent, contraindications to venipuncture, withdrawal of consent, incomplete or incorrectly filled questionnaire. No specific exclusion criteria were applied regarding the history of other helminth infections, allergic or autoimmune diseases, immunodeficiency states, or pregnancy, since testing was not the clinical diagnostic testing. Venous blood (5–7 mL) was collected in vacuum test tubes with the coagulation activator; after 20–30 min of coagulation samples were centrifuged at 1500–2000 × g for 10 min. Blood serum was transported at 2–8 °C, aliquoted, and stored at –20 °C. Samples with the pronounced hemolysis, lipemia, apparent contamination, or insufficient volume were not analyzed; repeated freezing and thawing were avoided whenever possible.

The questionnaire included the following items: age, gender, place of residence, contact with animals (including dogs and cats), facts of having a country house, vegetable garden, or single-detached house, sources of drinking water, consumption of raw or undercooked meat, consumption of insufficiently washed vegetables and fruits, as well as fishing and hunting. Age groups 18–29, 30–44, 45–59, 60–74, and ≥75 years were formed for the descriptive and secondary comparison. Clinical symptoms, eosinophilia, total IgE, concomitant disorders and pregnancy were not used as the criteria for clinical classification of avidity results. Associations between AI and questionnaire factors were not the main objective of the study.

The avidity index (AI) of IgG-positive samples was the main indicator; the IgG detection rate was the characteristic of the surveyed group. Avidity testing was planned for all IgG-positive sera; for the AI analysis, samples were included only if these provided a sufficient volume of material, yielded two valid paired measurements, and showed a non-zero OD value in the assay variant without denaturing treatment. The proportion of seropositive subjects was calculated based on the total number of individuals assessed, and the AI distribution was calculated based on the number of positive samples that had passed the paired testing quality control.

Determination of IgG against Toxocara canis

Antibodies were determined by the indirect solid-phase

ELISA using the experimental Toxocara-IgG-Human-ELISA– MVA GROUP kit (MVA GROUP, Kazakhstan). The basic test configuration corresponded to the previously described experimental platform [4]. The T. canis native antigenic extract was fractionated by gel filtration. The 20–22 kDa working antigen fraction was controlled by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (SDS-PAGE) using the Bio-Rad Laboratories equipment (Hercules, CA, USA), comparing the protein mobility with standard molecular weight markers. The fraction was immobilized on high-binding 96-well polystyrene plates at a load of 100 µg of antigen per plate; no separate blocking stage was used.

Sera were tested in duplicates with the final dilution 1 : 5:  20 µL of serum were added to 80 µL of dilution buffer. After the 30-min incubation at 37 °C, the plate was washed five times with phosphate-buffered saline with Tween-20. The conjugate of the anti-human IgG antibody with horseradish peroxidase diluted 1 : 30,000 was added and incubated for 30 min at 37 °C. After five washes, the 3,3′,5,5′-tetramethylbenzidine (TMB) substrate was added, staining was performed for 15 min at 20–25 °C in the dark, and the reaction was terminated with the 0.9 M sulfuric acid. OD was measured at 450 nm with the reference wavelength of 630 nm using the BIOBASE microplate photometer (BIOBASE Group, Jinan, China); an incubator and a plate washer from the same manufacturer were used for incubation and washing. 

Each assay run included positive and negative controls and the wells containing reagents without any serum for background control. The assay run was accepted with the mean OD of ≤ 0.20 for the negative control and ≥ 0.60 for the positive control. The positivity threshold was equal to the mean OD of the negative control plus 0.200; the sample was considered positive if its OD was not lower than this threshold. The 1 : 5 dilution and the additive value of 0.200 selected during the preliminary optimization phase based on the separation of signals from positive and negative sera, the background OD, and reproducibility served as internal operating parameters for the experimental system, rather than as universal diagnostic thresholds. In this sample, diagnostic sensitivity and specificity were not assessed independently; the ELISA results were interpreted as serological findings rather than as a confirmed clinical diagnosis.

The laboratory registry categories "weakly positive," "positive," and "strongly positive" were preserved for the intergroup analysis. These characterized the positive ELISA result strength and did not represent a measured antibody concentration. The analysis of these ordinal categories did not replace the IgG quantification.

Avidity determination and reproducibility control

Avidity was assessed in a paired format, with the final serum dilution of 1 : 100 in both variants. For that 10 µL of serum pre-diluted 1 : 10 were added to 90 µL of dilution buffer. Each assay variant was performed in duplicates. The 1 : 100 dilution was used in the avidity modification only; it was selected in preliminary assay runs to reduce the risk of signal saturation and obtaining comparable OD values  for the treated/untreated well pair. Therefore, the numerator and denominator for the AI were obtained at the same dilution of 1 : 100, separately from the screening ELISA 1 : 5.

After the 30-min incubation at 37 °C and three washes, 100 µL of denaturing solution (DS) containing 6.0 M urea was added to one set of wells. The paired variant was not subjected to additional denaturing treatment. Both variants were kept for 10 min at 20–25 °C. Then two washes, incubation with the conjugate, and further detection stages were implemented under the same conditions. “No DS” refers to the lack of further treatment with urea at this stage.

Operating conditions were determined using control sera with expected low and high avidity, 10 samples per group. Control sera were aliquoted, stored at –20 °C, and used, minimizing the repeated freeze–thaw cycles. We compared 4.0, 6.0 and 8.0 M urea at an exposure time of 10 min and 20–25 °C. For 6.0 M urea, additional comparisons were made for 5, 10, and 15 min at 20–25 °C, and for 10 min at 37 °C. The separation of AI distributions, signal retention in the high-avidity group, and reproducibility were assessed. These controls that characterized binding stability in the assay were not used as groups with the independently determined infection duration.

Reproducibility was further characterized at three control avidity levels. The intra-assay variability was assessed based on replicate measurements within a single assay run, while interassay variability was assessed based on independent assay runs. CV was recalculated using the mean values  and standard deviations provided in the laboratory summary: CV (%) = SD / mean × 100. These indicators characterize the measurement repeatability, but do not determine the test clinical diagnostic accuracy.

Index calculation and interpretation principles

The avidity index was calculated using the following formula:

AI (%) = (OD_DS / OD_noDS) × 100,

where OD_DS was optical density in the well containing the denaturing solution, OD_noDS was optical density in the paired well containing no denaturing solution.

The OD values recorded in the laboratory table were used for calculation; no additional background correction was applied during the statistical recalculation. A separate lower threshold for the OD with no DS was not applied for inclusion in the AI calculation. For all positive samples, the availability of both measurements and a non-zero denominator was verified; IgG-negative samples were not included in the avidity analysis.

The primary analysis included the continuous AI distribution. Non-overlapping intervals were used to provide the clear description of that: 0–<20%, 20–<40%, 40–<60%, 60–<80%, 80–100%, >100–140%, and >140%. The first two intervals were combined when discussing the values < 40%. These intervals were used to describe the data and were not considered validated criteria for an acute process, reinfection, or recovery.

Values >  100% were kept in their original form and considered separately. Operationally, these meant that OD_DS > OD_noDS. To assess the impact of such observations, additionally the median and the interquartile range were calculated for the samples with AI ≤ 100%.

Statistical analysis

Proportions were presented as a number of observations and a percentage; 95% confidence intervals (CI) were calculated using the Wilson’s method. The median (Me), the 1^st and 3^rd quartiles (Q₁ and Q₃; the 25^th and 75^th percentiles, respectively), as well as the minimum and maximum were provided for AI; in addition, the mean and standard deviation (SD, standard deviation) were used. Me [Q₁–Q₃] represented the median and the interquartile range. AI distributions across the three ELISA categories were compared using the Kruskal–Wallis test; H is the statistic for this test. The comparison was general; pairwise comparisons and linear correlation analysis were not performed. Demographic and personal data were presented in the descriptive format as absolute and relative frequencies; the relationship between those and AI was not assessed. All tests were two-sided, the statistical significance threshold was 0.05. Calculation was performed in Python 3.12.14 with the NumPy 2.3.5 and SciPy 1.17.0 libraries.

RESULTS

Studied group and completeness of avidity analysis

A total of 500 were included in the analysis: 329 females (65.8%) and 171 males (34.2%). The average age was 41.3 years, SD 15.5 years; the range was 18–82 years. IgG against the T. canis antigen were detected in 93 subjects: 18.6% (95% CI 15.4–22.2%). All 93 positive samples had paired OD values and were included in the avidity analysis; there were no samples excluded from this subsample due to the absence of one of the measurements. The OD with no DS was 0.407–2.234, there were no zero denominators.

The data on the number of individuals who were invited, but refused or were excluded prior to the final sample formation were not available for this analysis. The gender and age distribution within the seropositive subsample and the relationship between the AI and the questionnaire factors were not assessed.

Selection of denaturation conditions and measurement variability

When selecting the conditions, 4.0 M urea did not provide sufficient separation between the control groups, whereas 8.0 M urea significantly reduced the AI even in high-avidity sera.

The 6.0 M urea with the exposure for 10 min at 20–25 °C was selected for further analysis. Under this regime, the AI ranges for the two control groups did not overlap; increasing the processing time or temperature reduced the values in the high-avidity group (tab. 1).

The intra-assay CV values were 4.9–8.3%, and the inter-assay values were 7.4–13.0% (tab. 2). The greatest relative variability between assay runs was reported for the low-avidity control.

Avidity index distribution

The median AI was 79.2% [52.7–90.7], the range was 8.9– 230.2%. The mean value was 77.9%, SD was 32.7 pp. The most frequent individual interval was the interval of 80–100%, but it did not cover the majority of observations. The values < 40% were rare; a significant portion of the distribution consisted of samples with the AI values 40–<80% (tab. 3).

Avidity at various strength of ELISA positive results

There were differences in the AI distribution between the three ELISA categories (H = 20.97; 2 degrees of freedom; p < 0.001). The median AI was lower in the strongly positive group, while the interquartile ranges for the weakly positive and positive groups overlapped substantially (tab. 4). The overall comparison does not establish statistical significance for each pair of groups. The results obtained have not shown a consistent increase in the AI in cases with the more strongly positive ELISA results.

Avidity index values over 100%

In 12 samples (12.9%; 95% CI 7.5–21.2%), OD with DS was higher, than OD with no DS; AI was 108.9–230.2%. The OD increase was within 0.116–0.919, the median difference was 0.311. In this group, OD with no DS was 0.518–1.996; therefore, all excesses cannot be explained by the small denominator. Paired measurements are presented in the figure. When considering 81 samples with the AI ≤ 100% separately, the median was 71.8% [50.4–88.6]; therefore, the inclusion of values >  100% had a noticeable effect on the final distribution characteristics.

DISCUSSION

The findings show the IgG avidity heterogeneity in seropositive adults in Almaty. The ELISA positive result strength was not associated with the consistent AI increase. Determining avidity therefore complements the characterization of antibodies, but it does not determine active toxocariasis. The long-term IgG persistence after treatment has been reported for children [7]; this paper supports a cautious interpretation of positive results, but the timing of antibody changes in adults requires separate assessment.

Clinical studies associate high avidity with a prolonged course of infection [11, 12, 15], although diagnostic thresholds and the information value of tests vary. According to some data, for the AI < 32%, sensitivity of 64.3% and specificity of 96.9% were obtained with regard to active toxocariasis; high AI was also found in a patient with active infection [13]. Other authors observed low or borderline values for 2–3 years [14]. There are also reports regarding the influence of the clinical composition of the groups and the selected AI thresholds [16, 17]. That is why the share of samples with the AI < 40% in our sample should not be equated with the share of recently infected individuals.

The dynamic changes in avidity depend on the studied model: in mice, the AI increased in the course of infection [18, 19], while in rabbits high avidity was reported as early as in the early seroconversion period [20]. Mostly high avidity has been reported in pregnant women and individuals over the age of 50 [21, 22]. It is necessary to consider the sample composition and the antigen, when comparing the results. The diagnostic significance of selecting low-molecular-weight fractions was demonstrated in an early study on immunoblotting [23], and subsequent assessment of recombinant proteins confirmed the influence of antigen composition on the detection of antibodies [24].

The lower median AI in the strongly positive group cannot be explained solely by the immune response maturity. OD with no DS is part of the formula denominator, and the antibody concentration and measurement non-linearity can affect the calculated ratio [25]. The avidity ELISA reproducibility was assessed after determining the dilution that yielded the OD of 0.5–1.3 [26]. This range is not universal for other antigens, but it does indicate the baseline signal control value. As for our system, it is necessary to check stability of the intergroup differences at various serum dilutions.

The AI values > 100% reflect the detected OD increase after processing and do not prove enhanced binding strength. These cannot be considered as confirmed measurement errors in advance. Thus, in the study focused on the type 1 herpes simplex virus (HSV-1) system, the results depended on the method of urea treatment; the need for the avidity-based ELISA standardization is also noted in another methodological paper [10, 27]. Early studies focused on the detection of circulating antigens and immune complexes in toxocariasis [28, 29] describe a possible biological context but do not confirm the cause of the OD increase in our system. Repeated paired measurements and a separate assessment of the effects of background, dilution, and denaturing treatment are necessary to determine it.

Our intra-assay CV values are comparable with the CV of 8% reported for the avidity-based ELISA for the SARS-CoV-2 S-protein involving treatment with 7 M urea [30]. There is a report of the CV <10% for avidity measurements within a preselected OD range [26]. These are methodological guidelines, not proof of test equivalence: other antigens, processing protocols, and avidity calculation methods were used. In our study, the low-avidity control inter-assay variability was higher. The resulting distribution characterizes the antibody response of the adults surveyed, but does not determine the clinical stage of infection. Further assessment should include repeated paired measurements, verification of the values >100% and comparison with clinical data. It is still important to confirm positive ELISA results by a separate method, since determination of avidity does not exclude cross-reactivity [6, 14].

CONCLUSIONS

Among 500 adults assessed in Almaty in 2026 IgG against the T. canis antigen were found in 93 people (18.6%). Avidity analysis of 93 samples with available paired OD values demonstrated a wide range of AI and substantial heterogeneity among seropositive samples. The interval of 80–100% was the most frequent individual interval, but it did not account for the majority of observations. The stronger positive response in ELISA was not accompanied by the AI increase. The reported difference requires further investigation, considering the fact that the index calculated depends on the input signal. The results > 100% reported for 12.9% of test samples should be considered separately and retested; the underlying cause could not be determined in this study. The use of avidity-based ELISA with the 20–22 kDa fraction expands the characteristics of the humoral response to Toxocara. The agreed-upon analysis conditions and transparent presentation of raw measurements are necessary for comparison of studies. The use of AI to determine the duration or activity of toxocariasis requires independent clinical confirmation.