Discrepancy between haemagglutination and radioimmunological techniques for measurement of serum thyroglobulin autoantibodies

Laboratory Thyroid Tests: A Historical Perspective

Background: This review presents a timeline showing how technical advances made over the last seven decades have impacted the development of laboratory thyroid tests. Summary: Thyroid tests have evolved from time-consuming manual procedures using isotopically labeled iodine as signals (¹³¹I and later ¹²⁵I) performed in nuclear medicine laboratories, to automated nonisotopic tests performed on multianalyte instruments in routine clinical chemistry laboratories. The development of isotopic radioimmunoassay techniques around 1960, followed by the advent of monoclonal antibody technology in the mid-1970s, led to the development of a nonisotopic immunometric assay methodology that forms the backbone of present-day thyroid testing. This review discusses the development of methods for measuring total thyroxine and triiodothyronine, direct and indirect free thyroid hormone measurements and estimates (free thyroxine and free triiodothyronine), thyrotropin (TSH), thyroid autoantibodies (thyroperoxidase, thyroglobulin [Tg] and TSH receptor autoantibodies), and Tg protein. Despite progressive improvements made in sensitivity and specificity, current thyroid tests remain limited by between-method differences in the numeric values they report, as well as nonspecific interferences with test reagents and interferences from analyte autoantibodies. Conclusions: Thyroid disease affects ∼10% of the U.S. population and is mostly managed on an outpatient basis, generating 60% of endocrine laboratory tests. In future, it is hoped that interferences will be eliminated, and the standardization/harmonization of tests will facilitate the establishment of universal test reference ranges.

Development and Validation of a New Low-Cost Enzyme-Linked Immunoassay for Serum and Dried Blood Spot Thyroglobulin

BACKGROUND

Thyroglobulin (Tg), a biomarker of iodine nutrition, can be measured on dried blood spots (DBS), which simplifies collection and transport in surveys. The World Health Organization recommends DBS-Tg for monitoring iodine status in children. It could also be a useful iodine biomarker during pregnancy. However, the Tg antibody (Ab) used in earlier DBS-Tg assays is no longer commercially available. The aims of the present study were: (i) to develop a new low-cost serum and DBS-Tg sandwich enzyme-linked immunosorbent assay for assessment of Tg in population studies; (ii) to check the stability of DBS-Tg during long-term storage; and (iii) to assess within-subject variability in DBS-Tg.

METHODS

Serum and DBS samples were measured from healthy pregnant women (n = 424) with the new assays, as well as the Immulite 2000 (Siemens), including TgAb positive (n = 150) and TgAb negative (n = 274) women. DBS-Tg stability was tested over 15 weeks of storage at -20 °C. Within-subject variability was evaluated over four weeks in four healthy adults.

RESULTS

Intra-assay and interassay variability was 4.4-7.3% and 10.1-12.9% for the new serum Tg assay, and 7.6-12.3% and 7.6-16.5% for the DBS-Tg assay. Correlation between the two serum methods was high (r = 0.68, p < 0.01). Assay performance in all women and those TgAb negative was comparable. Correlation between the new serum Tg assay and the DBS-Tg assay was high (r = 0.78, p < 0.01), and agreement expressed as a function of the average Tg concentration for the two methods (X) was 0.59X -4.59 μg/L. DBS-Tg was stable for 15 weeks stored at -20 °C. Within-subject variability in DBS-Tg was 21.1%. Reagents and antibodies costs for the new serum and DBS assays are ∼ US$1.

CONCLUSIONS

These new low-cost serum and DBS-Tg assays perform well over a wide range of Tg concentrations, and the field-friendly DBS assay may be particularly useful in population studies of iodine nutrition.

Implications of thyroglobulin antibody positivity in patients with differentiated thyroid cancer: a clinical position statement

BACKGROUND

Even though the presence of antithyroglobulin antibodies (TgAbs) represents a significant problem in the follow-up of patients with differentiated thyroid cancer (DTC), the current guidelines on the management of DTC that have been published in recent years contain no text concerning the methods to be used for detecting such antibody-related interference in thyroglobulin (Tg) measurement or how to manage TgAb-positive patients in whom Tg cannot be used reliably as a tumor marker.

AIM

An international group of experts from the European Thyroid Association Cancer Research Network who are involved in the care of DTC patients met twice to form a consensus opinion on how to proceed with treatment and follow-up in TgAb-positive DTC patients based on the available evidence in the literature. Here we will report on the consensus opinions that were reached regarding technical and clinical issues.

RESULTS

This clinical opinion article provides an overview of the available evidence and the resulting consensus recommendations. The current literature does not provide sufficient data for giving evidence-based answers to many questions arising in the care of TgAb-positive DTC patients. Where insufficient evidence was available, a thorough discussion by a group of physician-scientists, all of whom have a distinguished track record in thyroid cancer care, was held to arrive at a consensus expert opinion. The questions and answers discussed were then summarized into an algorithm for the management of TgAb-positive patients.

CONCLUSION

We were able to define 26 consensus expert recommendations and a resulting algorithm for the care of TgAb-positive DTC patients.

Is TPO detectable in the circulation?

Recent reports have suggested that thyroid peroxidase (TPO) can be detected in the circulation of normal subjects and of patients with Graves' disease and we have attempted to confirm and extend these observations. A TPO radioimmunoassay with a sensitivity of 1 ng/mL was used to measure TPO in the sera from 20 normal subjects and 21 patients with Graves' disease. In addition, TPO was measured in serum samples from six normal subjects before and after oral TRH. We were unable to detect TPO in 46 out of the 47 sera studied (normals and autoimmune thyroid disease). In the one remaining serum (from a normal subject), low levels of TPO were apparently detected, but we demonstrated that this result was due to assay interference from TPO autoantibodies. Overall our studies suggest that (1) thyroid peroxidase is not detectable in normal subjects nor in TPO autoantibody negative patients with Graves' disease; (2) endogenous TPO autoantibodies can interfere in the TPO radioimmunoassay leading to false positive results; and (3) an acute increase of TSH in normal subjects does not result in TPO release into the circulation.

Assessment of antithyroglobulin and antimicrosomal autoantibodies in patients with autoimmune thyroid disease: comparison of haemagglutination assay, enzyme-linked immunoassay and radioligand assay

Four different assay systems for detection of antithyroglobulin (T-Ab) and thyroid antimicrosomal autoantibodies (M-Ab) were evaluated: two passive haemagglutination assays (PHA), an enzyme-linked immunoassay (ELISA) and a radioligand assay (RLA). Antibody levels measured with these methods correlated well (T-Ab: r = 0.72 to 0.88; M-Ab: r = 0.63 to 0.84; p less than 0.0001). However, when the results of the measured samples were classified as normal, slightly elevated and pathological, only 40-50% of the samples showed congruous results in all tests; 60-70% agreed in PHA and ELISA, whereas 80 to 90% corresponded in the two PHAs. RLA and ELISA gave more frequently positive results for T-Ab and negative results for M-Ab than the PHAs. Despite the lower sensitivity of the quantitative methods for M-Ab detection, they depicted more readily small changes after thyroxine treatment than the PHAs. We suggest that differences in autoantibody levels found with different methods may be due to autoantibody heterogeneity.

Sensitive assay for thyroglobulin autoantibodies in serum employing polyethylene glycol for precipitation

A sensitive assay for thyroglobulin autoantibodies (Tg-ab) is described. The assay is based on prolonged incubation of [125I]Tg with serum followed by precipitation of antibody bound [125I]Tg by polyethylene glycol (PEG). The PEG concentration was chosen to give a low precipitation of free [125I]Tg while effectively precipitating Tg-ab-bound [125I]Tg. The lowest reference employed contained 19.5 MRC U/l Tg-ab. This was always different from an incubate without Tg-ab. Tg standard added to samples interfered in the assay giving spuriously low Tg-ab values as in other Tg-ab assays. Tg-ab added to samples was recovered quantitatively. The interassay coefficient of variation was 19.8% and the intra-assay coefficient of variation 11.3%. Only eight of 66 serum samples with various amounts of Tg-ab, measured with the new assay, had levels of Tg-ab detectable by the passive haemagglutination test.

A high prevalence of thyroglobulin autoantibodies in adults with and without thyroid disease as measured with a sensitive solid-phase immunosorbent radioassay

Sera from 228 patients with thyroid disease and 140 healthy subjects without clinical or biochemical evidence of thyroid disease, were tested using a sensitive solid-phase immunosorbent radioassay (RIA) and a passive hemagglutination test (TRC test) for thyroglobulin antibodies (Tg-ab). With the RIA technique, Tg-ab was found in 27% of the controls (36% of the women and 15% of the men), whereas only 0.7% of them were Tg-ab positive with the TRC test. All individuals with primary hypothyroidism were Tg-ab positive with the RIA, compared with only 56% with the TRC test. Tg-ab (RIA) were found in 43/53 (81%) of the patients with toxic diffuse goiter, and in 30-40% of the patients with toxic nodular goiter, toxic adenoma, atoxic goiter, and thyroid carcinoma, the TRC test being positive in 10-17% of these patients. The high prevalence of Tg-ab in the healthy population suggests that subclinical thyroiditis is more frequent than has been assumed from antibody measurements made with less sensitive methods, and is in agreement with the prevalences reported from autopsy studies.

Enzyme-linked immunosorbent assay for determination of thyroglobulin autoantibodies. Comparison with RIA and haemagglutination

An enzyme immunoassay (ELISA) has been developed for the detection of human autoantibodies to thyroglobulin (TgAb). The ELISA assay was compared to the previous routine method, the tanned red cell (TRC) method and a radioimmunoassay (RIA) technique for thyroglobulin autoantibodies. Sera from 163 healthy non-hospitalized persons, 31 patients with Hashimoto's thyroiditis, 13 patients with Graves disease, 136 patients with thyroid cancer and 365 sera consecutively received for thyroid autoantibody screening were examined for TgAb in the three methods. The nosological sensitivity for ELISA methods was comparable to that of RIA and higher than that of TRC. The ELISA method was accurate, precise, objective, inexpensive and well suited for quantitative routine determination of TgAb.

A new sensitive immunosorbent radioassay for the detection of circulating antibodies to polypeptide hormones and proteins

A solid-phase immunosorbent radioassay for the detection of circulating antibodies to protein hormones is described. The assay is based on the binding of the homologous 125I-labelled antigen to the antibodies which are then bound to anti-IgG antibodies covalently coupled to Sepharose. It can easily be applied as a complement to any radioimmunoassay for the detection of circulating antibodies to the ligand measured. The assay system avoids falsely elevated values due to interference of high serum concentrations of the antigen. The assay was applied to measure antibodies to FSH, LH, TSH, GH, prolactin, insulin and thyroglobulin (Tg). Among patients with chronic thyroiditis Tg antibodies were found in 100% of the sera. In diffuse toxic goitre 73% of the patients had detectable Tg antibodies. Insulin antibodies were present in 82% of the sera from patients with insulin treated diabetes. No antibodies were found against the other protein hormones tested.