Sage Journals: Discover world-class research

Abstract

Thyroglobulin (Tg) is a tumour marker for differentiated thyroid cancer. Interpretation requires a knowledge of the current thyrotropin (TSH) concentration as secretion is TSH-dependent. While a raised serum Tg may be indicative of residual or recurrent thyroid cancer, trauma to the thyroid (e.g. surgical, biopsy or due to radioiodine treatment) also causes an increase. Tg may be measured when TSH is suppressed and also following recombinant TSH (rhTSH) stimulation. Interpretation of results in pregnancy and in children is discussed. Assay bias and interference by endogenous Tg antibodies (Abs) are the main confounders in the interpretation of results. Although there is an international standard for Tg, there are large differences in results and yet there are few assay-specific clinical decision limits. Patients should therefore be monitored with the same assay. Endogenous TgAbs may cause false-negative interference in immunometric assays and may cause false-positive results in radioimmunoassay. Although the measurement of TgAbs has been advocated for predicting interference, it is now clear that interference can still occur when TgAbs have not been detected, the effect being TgAb-assay-specific. Approaches to identifying those samples where there may be interference are discussed. The laboratory should have a protocol for the investigation of possible interferences and data on the bias of the Tg assay that they use. An appreciation of the clinical uses of the service is required as an understanding by endocrinologists, oncologists and endocrine surgeons of the analytical limitations of the service.

Introduction

Thyroid cancer, although rare, is the commonest endocrine cancer with over 2000 people diagnosed with the disease in 2007 in the UK and a men:women ratio of 1:3.¹ Patients typically present with thyroid nodules, with or without goitre, and the clinical need is to identify those nodules which represent thyroid cancer, of which differentiated thyroid cancer (DTC) is the commonest form. Prevalence rates for thyroid nodules are dependent on mode of detection (palpation versus ultrasonography), population, gender and age. In the survey of thyroid disease in the community of Whickham, the prevalence of thyroid nodules was found to be 3.2% against a background prevalence of goitre of 6.9% and 8.6% (palpable and visible and palpable but not visible, respectively).² In general, treatment consists of total thyroidectomy, radioiodine ablation of the thyroid remnant and treatment with thyroxine (T4) in doses sufficient to suppress serum thyrotrophin (TSH). Prognosis is good with a survival of 90% at 10 years. The American, British and European Thyroid Associations have provided guidelines on the treatment of thyroid cancer.^3–6

Thyroglobulin (Tg) is a large molecular weight glycoprotein (approximate molecular weight 660 kDa) found in thyroid cells and the follicular lumen, and is involved in the synthesis of thyroid hormones. In humans, the gene is found on chromosome 8 and transcription is controlled by a number of thyroid-specific transcription factors. Tg mRNA is very heterogeneous and, once translated, the protein undergoes glycosylation. The extent of glycosylation and iodination depends on the source of the protein, e.g. from normal thyroid, tumour tissue or the peripheral circulation. It is released into the peripheral circulation during normal hormone synthesis and after trauma to the thyroid. As the standard approach to treatment of DTC is thyroidectomy followed by radioactive ablation, successful treatment is reflected by an undetectable serum Tg. Thus, measurement of serum Tg can be used as a tumour marker for monitoring treatment of DTC and to detect disease recurrence, with detectable/rising serum Tg potentially indicating residual or recurrent disease.⁷ This contrasts with a subject with an intact thyroid who will have serum Tg in the range of 3–40 μg/L.⁸ It cannot be overemphasized that reference ranges and values for measured Tg are assay-dependent.

Serum Tg is currently measured by immunoassay and there are a number of analytical challenges, namely the requirement for stability of the assay over decades given the need for long-term clinical follow-up, differences in assay bias and interference in the immunoassays by endogenous Tg antibodies (Abs).^9,10 Concern remains that endocrinologists, endocrine surgeons and oncologists may not be aware of the analytical limitations of immunoassay for Tg and that laboratories may not be aware of the varied clinical situations in which results are used. The aim of this paper is to consider the use of serum Tg as a long-term marker of DTC and the pitfalls of interpretation of results by a multidisciplinary team.

Routine review of the patient with DTC

Serum Tg is one element of the assessment of the patient and interpretation of the result requires a full knowledge of, for example, clinical presentation, type of thyroid surgery, thyroid hormone replacement therapy and results of imaging. It is essential that serum TSH is measured at the same time as serum Tg as secretion of Tg is TSH-dependent. It is unlikely that the clinical biochemistry laboratory will always have sufficient information to ‘interpret’ a Tg result at the stage of result authorization, but knowledge of histological type of DTC, surgery, ablation and TSH will allow assessment of whether the Tg result is analytically and clinically plausible. It is inappropriate to quote a reference range for serum Tg determined from a population of subjects with intact thyroids.

Serum Tg may be requested at the following stages of patient care and these will be discussed:

At presentation with thyroid nodule or nodular goiter;

Prethyroidectomy/preradioiodine ablation;

A few weeks post-ablation of the thyroid remnant;

On T4 replacement therapy;

Off T4 replacement therapy;

Following stimulation with recombinant TSH (rhTSH; Thyrogen^®).

The thyroid nodule and nodular goitre

Thyroid nodules may be the presenting feature found on palpation of the thyroid or may be discovered incidentally on imaging of the head and neck. Depending on clinical presentation and findings, history and family history, it has been recommended⁴ that only nodules >1 cm require evaluation as these may represent clinically significant cancers. Serum TSH should be estimated. A subnormal serum TSH with a normal or raised serum free T4, i.e. the biochemistry of overt or subclinical hyperthyroidism, is rarely associated with thyroid cancer. Measurement of serum Tg is not indicated as concentrations of Tg are elevated in most thyroid diseases and coexisting lymphocytic thyroiditis or Graves’ disease may be more prevalent in patients with DTC prior to surgery,^11,12 thus the test lacks specificity.¹³

Diagnostic ultrasound (US) establishes the size and US nature of the nodule and may be accompanied by fine needle aspiration (FNA) biopsy. Results of FNA cytology (FNAC) can then be considered according to defined diagnostic categories. These can be summarized as follows: Thy1 – non-diagnostic; Thy2 – non-neoplastic; Thy3 – follicular lesion/suspected follicular neoplasm; Thy4 – suspicious of malignancy; and Thy5 – diagnostic of malignancy.⁵

DTC is the commonest thyroid cancer comprising papillary cancer at 85% of cases, follicular 10% and Hürtle cell or oxyphil tumours 3%. Certain histological subtypes of papillary thyroid cancer may have a worse prognosis.

Metastatic disease – cervical lymphadenopathy

The neck represents the commonest site for DTC metastases and the presence of cervical metastases will determine the extent of primary surgery or the need for second surgery. Preoperative imaging may also identify cervical lymphadenopathy and indicate the possible presence of metastases. A number of studies have considered the measurement of Tg in aspirates from cervical nodes as indicating metastatic disease. There are a number of potential analytical pit-falls with this approach. There is reported variability in how the sample is collected, including dilution effects when collecting needle ‘washouts’ and possible contamination with serum and thyroid cells. Laboratories need to establish protocols for the exclusion of the high-dose hook effect and investigate the possible matrix effects of the needle washout fluid in their assay. Logistical difficulties in collecting needle washouts into serum-based diluents have been reported.¹⁴ As an alternative, sodium chloride (9 g/L) has been used to collect the needle washout sample, despite the reported over-recovery of exogenous Tg of 100–140%.¹⁴ Other studies have shown a wide range of results in such samples. Boi et al.¹⁵ defined Tg cut-off values of 36 μg/L if the patient had an intact thyroid and 1.7 μg/L if there was no thyroid, to indicate that the lymph node that had been aspirated contained metastatic disease. An earlier study also showed overlap between patient groups (pre- and post-thyroidectomy). In both patient groups, there was a significant elevation in Tg in the samples from patients with metastatic disease compared with those with negative lymph nodes (post-thyroidectomy patients; range of aspirate Tg < 0.25–76,000 versus <0.25–5.0 μg/L and pre-thyroidectomy patients; range of aspirate Tg < 0.25–89,000 versus <0.25–20 μg/L for metastatic versus negative lymph node, respectively).¹⁶

There are few data on the comparative bias of different Tg assays in aspirates,¹⁷ and further work is required to determine the analytical limitations of analysing such samples particularly with regard to assay specific cut-offs. It is clear that cytology should always be performed on the aspirates as Tg will be undetectable in undifferentiated metastases.

Metastatic disease – unknown primary

As for many tumour markers, the measurement of serum Tg is not diagnostic. The common sites for metastatic disease for DTC, other than cervical lymph nodes, are the lungs and bone. The laboratory may receive requests for serum Tg in patients with such metastases with a primary of unknown origin. Although very high serum Tg concentrations may suggest metastatic thyroid cancer, moderately elevated or normal concentrations do not rule it out. In an early study,¹⁸ 18 of 40 patients with metastatic thyroid cancer had a serum Tg greater than 400 μg/L compared with two out of 55 with benign nodular disease. Cut-off values are not agreed and other co-existent thyroid diseases may cause similar elevations.¹⁹

Preoperative/preradioiodine ablation Tg

The measurement of Tg preoperatively/preradioablation is not recommended as there are few data to suggest that this influences further treatment/actions.^3–6 Given that the patient has an intact thyroid it is not possible, with current assays, to determine the source of serum Tg, i.e. whether from normal thyroid tissue, tumour or metastases.

Early measurement of Tg post-thyroid remnant ablation

Patients with known metastases, extra-thyroidal tumour extension, primary tumours larger than 4 cm or other high-risk features will generally be offered radioiodine ablation. This may be carried out after thyroid hormone withdrawal or following administration of rhTSH. Initial assessment of the success of ablation is usually performed at six months. There is nothing to be gained by measurement of serum Tg less than six weeks post-ablation (or thyroidectomy) as any measured Tg may merely reflect the recent injury to the thyroid and subsequent clearance of released Tg. Assessment at six months may also include a ¹³¹I diagnostic whole body scan (WBS) to determine the success of ablation and extent of any residual disease in the neck or elsewhere.

Long-term follow-up

The frequency of follow-up after surgery and thyroid remnant ablation will depend on histological type and other risk factors. Initial review may be at six-month intervals and then yearly and is likely to be lifelong.⁵ The laboratory should therefore endeavour to provide Tg results from the same assay over a period of decades for a given patient. Requests for serum Tg may be made when the patient is (1) taking T4; (2) off T4/T3, i.e. biochemically hypothyroid; or (3) after administration of rhTSH. It is important that serum Tg has a high negative predictive value, not only to reassure the patient of the absence of disease but also to optimize the frequency of follow-up. Review of the patient may include both serum Tg and a ¹³¹I-WBS.

Of the possible outcomes of the above approach, an undetectable serum Tg with no radioiodine uptake into the thyroid bed or elsewhere, and a detectable serum Tg with detectable isotope uptake, would be considered to be in agreement, indicating successful or unsuccessful thyroid ablation, respectively. The difficulties in interpretation may arise where there is a detectable serum Tg but no radioiodine uptake or an undetectable serum Tg with significant isotope uptake.

Serum Tg may be a less sensitive marker in patients with less differentiated tumours, with some tumours losing the ability to trap radioiodine and secrete Tg. The possibility of a falsely low serum Tg due to interference by endogenous TgAbs in the immunoassay should be excluded.

Where a serum Tg is detectable but with no radioiodine uptake on diagnostic scan, a false-positive interference in the Tg assay, for example reflecting heterophilic antibodies in an immunometric assay (IMA) or anti-TgAbs in a competitive radioimmunoassay (RIA), should be excluded (see ‘Is the result correct?’ and following sections).

Recombinant thyrotropin

Protocol

Recombinant human TSH (Thyrogen^®; Genzyme Corp, Cambridge, MA USA) is licensed for use with or without radioiodine imaging and with measurement of serum Tg for the detection of thyroid remnant and thyroid cancer in patients post-thyroidectomy.²⁰ It is also licensed to increase the uptake of radioiodine for the ablation of remnant thyroid tissue post-thyroidectomy. The dose can be administered as two 0.9 mg intramuscular injections 24 h apart. With a single injection, the serum TSH reaches a peak 6–8 h post-injection of 280 mIU/L, falling to 140 mIU/L at 24 h and 25 mIU/L at 48 h.²¹ Use of rhTSH is contraindicated in patients with hypersensitivity to human or bovine TSH and its use should be avoided in pregnancy and patients who are breast feeding.

Interpretation

Early reports of the use of rhTSH-stimulated serum Tg as a marker of residual or recurrent disease were based on whether the Tg was undetectable or detectable. These studies utilized an assay with a quoted functional sensitivity of 0.5 μg/L and also a coefficient of variation of 24.8% at a Tg of 0.44 μg/L.^22,23 A number of studies have reported on the sensitivity and specificity of the test though these are difficult to compare as some have assessed performance for the detection of metastatic disease, some for residual disease and yet others for both.²⁴ With serum Tg cut-off concentrations of between 1.0 and 3.0 μg/L, sensitivity varied from 56% to 100% and specificity from 82% to 100%. This may be due to the different populations studied, for example, the proportion of high- versus low-risk patients and whether studied at first or subsequent follow-up. In addition, some studies have excluded those patients who were found to have endogenous antibodies to Tg, which may cause interference in the Tg assay, thus influencing the estimates of sensitivity and specificity.

Subsequently, the cut-off value of serum Tg that indicates residual or recurrent disease in the context of rhTSH stimulation has been widely considered to be a value of 2.0 μg/L.²⁵ It is worth noting the origins and limitations of this value. Table 1 gives details of reported cut-off values with details of the analytical methods and standardization used. The cut-off of 2.0 μg/L has been used in clinical practice but with no allowance for assay bias. Kloos and Mazzaferri³¹ noted that in some studies of rhTSH, using a cut-off for serum Tg of 2 μg/L, the positive predictive value was only around 50%. They concluded from their follow-up study, using a chemiluminescent IMA, that a single rhTSH Tg greater than 2 μg/L predicted persistent tumour but that no value entirely excluded future recurrence. They suggested that a single rhTSH-stimulated value for serum Tg of less than 0.5 μg/L (in the absence of endogenous TgAbs) gives an approximate 98% likelihood that the patient is free of tumour.

Table 1

Cut-off values for serum Tg used to assess the performance of rhTSH stimulation in detecting metastatic and residual disease (adapted from Robbins and Robbins²⁴)

Tg cut-off (μg/L)	Assay	Standard
2.0²⁶	Dynotest-Tg S IMA with polyclonal and monoclonal antibodies (Brahms Diagnostics)	Calibrated to 0.5 of the CRM 457 value
2.0 and 5.0²²	In-house RIA	CRM 457
1.0 (increment) 3.0 (threshold)²⁷	(Sanofi Diagnostics Pasteur)	Not stated

2.0²⁸	Optiquant IMA	Not stated
	(Kronus)
1.0²⁹	IMA (DPC)	CRM457
2.0³⁰	ICMA (Nichols Institute Diagnostics)	Not stated
2.0²⁵	Not stated	Not stated

Brahms Diagnostics, Berlin, Germany; Sanofi Diagnostics Pasteur, Mions, France; Kronus, Star, ID, USA; DPC (now Siemens Healthcare Diagnostics), Los Angeles, CA, USA; Nichols Institute Diagnostics, San Juan Capistrano, CA, USA

Tg, thyroglobulin; rhTSH, recombinant thyrotropin; IMA, immunometric assay; RIA, radioimmunoassay; ICMA, immunochemiluminometric assay

Since then, it has been suggested that patients can be monitored for persistent or recurrent disease without the need for TSH stimulation if a Tg assay with a functional sensitivity of at least 0.1 μg/L is used, though there might be a concomitant decrease in specificity of the test.⁴ Smallridge et al.,³² using the Beckman Access IMA (Beckman Coulter, Brea, CA, USA), found that only a small number of patients (2.5% of their cohort) with an unstimulated Tg of <0.1 μg/L had a stimulated Tg greater than 2 μg/L and imaging evidence of local recurrence or distant metastases. They concluded that patients can be monitored with a serum Tg while on T4 suppressive therapy with periodic neck US and that a rising Tg or the finding of abnormal lymph nodes should prompt further investigation. The authors did note that in their practice, there was a requirement for extensive evaluation whenever there was a change in reagent lot (this comprised a 40-patient sample comparison in duplicate, linear regression analysis of the data showing no significant change in the intercept and no change in slope of the regression line greater than 10%) to ensure that there was no clinically significant effect at low Tg concentrations of the change in reagents. Other studies using the same assay have established different cut-off values. Malandrino et al.,³³ studying patients on T4 suppression, found that a cut-off of 0.15 μg/L for serum Tg gave the best specificity and sensitivity for detecting disease recurrence, and others³⁴ have used a value of 0.27 μg/L.

The difficulty in establishing cut-off values for serum Tg for routine clinical practice is illustrated by a study³⁵ in which serum Tg was analysed by both Immulite IMA (Siemens Healthcare Diagnostics, Los Angeles, CA, USA; quoted functional sensitivity of 0.9 μg/L) and the Beckman Access IMA (experimentally determined functional sensitivity of 0.1 μg/L) on a proportion of the study samples. The authors present absolute values of serum Tg for 35 patients while on T4 suppression and post-rhTSH. Analysis of this published data, excluding data points below the relevant assay's lower limit of reporting, shows that the results from the Beckman assay are significantly lower (Wilcoxon test, P < 0.001) than that of the Immulite, despite traceability of both assays to the Certified Reference Material (CRM) 457. This contrasts with the data presented by Schlumberger et al.³⁶ who showed good agreement of these two assays when the CRM 457 was added to a basal sample. These differences in assay bias for patient samples but not ‘spiked’ samples might be due to differences in the molecular forms of Tg recognized by the difference assays and make comparison of functional sensitivities and cut-off values problematic.

Comparison of studies investigating the use of serum Tg while on T4 suppression with Tg post-rhTSH are further complicated by differences in the patient populations studied with respect to risk of recurrence, the method for excluding false-negatives due to TgAb interference and at what stage of the follow-up protocol neck ultrasonography is used. The implications of misclassification of patients as disease-free or with recurrent disease requiring further radioiodine treatment are significant and the guidelines emphasize the significance of a rising Tg and further investigations.

Whether a cut-off value of serum Tg has been determined for TSH-stimulated or TSH-suppressed patients, the main difficulty in extrapolating cut-off values determined for one study population to another is primarily the significant effect of assay bias, particularly at low concentrations of Tg in serum. In addition, of the published studies, many of the assays are no longer used or insufficient method detail is given, making assessment of their bias difficult. Although many assays are stated to be traceable to the CRM 457,^37,38 a wide scatter of results is obtained as noted above. When the CRM 457 was prepared from the normal thyroid tissue of nine patients undergoing thyroidectomy for a single adenoma, it was noted on electrophoresis that the preparation, as well as containing the intact Tg molecule, also contained smaller molecular weight forms. With the availability of recombinant human Tg, it is hoped that its use as a standard will result in a reduction in inter-assay variability.³⁹

In the meantime, the problems of assay variability are illustrated by External Quality Assessment Scheme data. For example, data from Distribution 47 UK National External Quality Assessment Service (UKNEQAS)⁴⁰ showed one sample with a range of measured Tg of 0.6–10.26 μg/L (assay mean for nine different IMAs) and for TgAbs of 1.4–217 U/mL (assay mean of eight assays), confirming four-fold differences in results previously reported.⁸ These issues are beginning to be recognized in the clinical literature.⁴¹

Biological variability

There is a significant literature on the biological variability of many analytes of clinical interest and recognition that the within-subject variability may differ in patients with disease as compared with healthy controls.⁴² There are few data for serum Tg – within-subject variability in a group of healthy women has been reported to be 8.5% and 16.2% for TgAbs and Tg, respectively.⁴³ Thus there is little information to allow formal evaluation of critical differences or reference change values for serum Tg. However, a recent retrospective analysis of patients with treated papillary thyroid cancer (post-thyroidectomy and TSH suppression) has suggested that the serum Tg doubling time may be helpful in risk stratification.⁴⁴ Given that low concentrations of serum Tg measured when TSH is suppressed can be due to persistent disease and normal thyroid remnant, this approach requires further investigation. The reported values for the half-life of Tg vary from 6 to 96 h, which may reflect both heterogeneity of Tg structure and assay specificity.⁴⁵ Thus, interpretation of results of samples collected less than six weeks post-surgery or radioiodine ablation is problematic.

In pregnancy

The care of the pregnant patient with thyroid disease is particularly demanding given the implications for both patient and fetus.^46,47 The DTC patient may first present during pregnancy or have a known diagnosis requiring follow-up. Detailed advice is given in the Endocrine Society Guidelines covering the need for FNA cytology, timing of possible surgery and the absolute necessity of avoiding the use of radioactive isotopes,⁴⁸ and has recently been discussed.⁴⁹

Of note for the laboratory is the need for more frequent monitoring of thyroid function and review of the dose of T4 in the patient with treated DTC. Dose requirements for T4 are increased in pregnancy and change throughout pregnancy, the major clinical need being to avoid under-replacement and potential effects on the fetus. Laboratories should be encouraged to quote method and trimester-specific reference ranges for TSH and free T4 and should recognize the need for more frequent monitoring, for example at 4–6 weeks if T4 dosage adjustment is made.⁵⁰

There have been few studies of serum Tg in pregnancy, with conflicting results in terms of the statistical significance of trimester-specific changes in women with an intact thyroid⁴⁹ (Table 2). Any changes in serum Tg may reflect changes in TSH alone. A stimulated Tg is not possible as both T4 withdrawal and rhTSH stimulation are contraindicated in pregnancy. Serum Tg may be monitored in the pregnant DTC patient, with a significant rise prompting detailed neck palpation and neck US.⁵⁴

Table 2

Serum thyroglobulin (μg/L) in pregnancy in subjects with an intact thyroid

Population	First trimester	Second trimester	Third trimester	Method
Mean (range)	8.4 (1.3–18.0)	9.2 (1.7–25.6)	10.1 early (1.8–22.8)	IMA⁵¹
			12.1 late (5.3–25.2)
Mean (μg/L) (standard error of the mean)	15.48 (1.96)	14.92 (2.05)	18.55 (2.8)	Immulite⁵²
Median (interquartile range), Swedish	15.5 (8–24)	10.5 (7–19)	18.0 (13–25)	In-house RIA⁵³
Median (interquartile range), Sudanese	27.5 (12–40)	25.0 (15–43)	30.0 (15–67)	In-house RIA⁵³

RIA, radioimmunoassay; IMA, immunometric assay

In children

Although thyroid cancer is not common in children, the prevalence of thyroid nodules found on clinical examination has been reported to be 1–1.5%, with a greater risk compared with adults for the presence of an underlying diagnosis of thyroid cancer if a nodule is present. In addition, thyroid cancer is reported to be a common secondary cancer in children treated with radiation therapy for another malignancy and it represents 0.5–3% of all childhood malignancies. DTC is less frequently inherited than medullary thyroid carcinoma but may be associated with the autosomal dominant conditions of Carney's complex, familial adenomatous polyposis and Cowden disease. Younger children are reported to have greater disease burden at presentation of DTC, with additional risk factors being identified such as male sex, large primary tumour size and extrathyroidal extension. The natural history of the disease in children is less well understood than in adults but the treatment pathway will be similar following FNAC: total thyroidectomy and TSH suppression with T4. However, the use of radioactive iodine remnant ablation in children is controversial.⁵⁵

The monitoring of treatment is based on the measurement of serum Tg, with similar problems of interpretation as those of adults. In a cross-sectional study, serum Tg was measured in children with an intact thyroid aged 7–18 years. There was no significant difference in serum Tg found between the genders (mean serum Tg was 28.0 μg/L [girls] and 25.3 μg/L [boys]), but a negative correlation with age was demonstrated. The authors concluded that there was a decrease in serum Tg with chronological age in children and adolescents which might be explained by changes in serum TSH.⁵⁶ As for adults, it is inappropriate to use a reference range determined in euthyroid subjects for the interpretation of results on children treated with DTC.

Other indications for the measurement of serum Tg

There are two other circumstances where serum Tg may be measured outside the context of treated thyroid cancer, namely in the investigation of congenital hypothyroidism and factitious thyrotoxicosis.

In the neonate, serum Tg has been reported to rise in the first six hours postnatally and for the first postnatal week, with concentrations falling thereafter in both term and preterm neonates. Reported concentrations vary between studies, with mean concentrations during the first week of life ranging from 41.9 to 180 μg/L and may be affected by study group characteristics such as iodine status and serum TSH concentrations.^57–60

The UK guidelines on screening for congenital hypothyroidism⁶¹ indicate that, if radiological imaging indicates that there is no thyroid present, measurement of serum Tg may be desirable, to exclude false-negative scans due to iodine uptake defects such as that associated with mutations in the sodium iodide symporter (NIS) gene.

In patients with congenital hypothyroidism, it might be expected that in the absence of a thyroid gland (athyreosis or thyroid aplasia), serum Tg would be undetectable or low, while for other causes with a confirmed raised TSH (such as transient hypothyroidism), serum Tg would be elevated.⁶² In dyshormonogenesis, concentrations of serum Tg can be low if there is a defect in Tg synthesis or intracellular processing or high if the defect is in other pathways. In an early study of congenital hypothyroidism, no significant differences in serum Tg were found between neonates with thyroid aplasia confirmed by thyroid scintigraphy (range = 15–290 μg/L) and a control group of euthyroid neonates (range = 34–700 μg/L).⁶³ In the neonates with congenital hypothyroidism, irrespective of cause, the range was <2.0–1100 μg/L. A more recent study⁶⁴ found a range of serum Tg of <1.0–18.7 μg/L in athyreosis while those with an ectopic gland had a range of 4.5–123 μg/L. Given this overlap in serum Tg, scintigraphic imaging and high-resolution colour Doppler US may be considered alternatives.⁶⁵

In the case of factitious thyrotoxicosis with a suppressed serum TSH, serum Tg will also be suppressed (having excluded analytical interference causing a false-negative result).⁶⁶ In contrast, in thyrotoxicosis, serum Tg will be increased despite a suppressed TSH due to release from the diseased thyroid.

Is the result correct?

Given the possibility of interference by endogenous antibodies in the Tg immunoassay, the laboratory plays a key role in developing strategies to identify and minimize the clinical impact of any such interference. This requires close collaboration with clinical staff and reporting such uncertainties is not straight forward. Different approaches will be considered such as the measurement of TgAb concentrations and Tg RIA-IMA discordance.

Endogenous antibodies to Tg

Endogenous antibodies to Tg may interfere in immunoassays for Tg. If there is antibody interference, it is thought generally to result in positive interference in competitive immunoassays and negative interference in IMAs, though this is not always the case.⁶⁷

The analytical limitations of TgAb assays themselves have been reported with respect to thyroid autoimmunity,⁶⁸ but are not as well appreciated in the area of thyroid oncology. Agglutination assays lack the required sensitivity and should not be used. However, there is significant between-method variability in results with the immunoassays for TgAbs.^68,69 Different immunoassay formats are used, either competitive or reagent excess, with all based on the binding of the endogenous TgAbs to either solid phase Tg or labelled Tg. These details of reagent type and preparation are likely to be critical given the variable epitope recognition of TgAbs. Although there is an International Reference Preparation available (IRP 65/093), variability in results may represent differences in antibody specificity and epitope expression in different disease states. In patients with DTC, antibodies with both restricted and broad specificities have been described, and interference in a Tg IMA has been shown to be related more to the number of epitopes recognized than to the pattern of epitope recognition.⁷⁰ In addition, high concentrations of serum Tg may result in under-recovery of TgAbs when TgAbs are measured by immunoassay.⁷¹

Further confusion may result from the use of different concentrations or cut-off values for defining ‘positive’ and ‘negative’ TgAb results. Such definitions are used either to define positive/negative for the detection of disease recurrence (for thyroid autoimmunity or for DTC) or positive/negative in terms of predicting interference in Tg assays. Even in defining ‘positive/negative’ for autoimmune thyroid disease, manufacturers’ cut-off values were found to vary between 80 and 325 kU/L⁶⁸ and for defining recurrence of DTC to vary between 27.8 and 100 U/mL.^72–76

Increasingly the limitations of TgAb assays in identifying those samples showing interference in Tg IMAs are being recognized. In part, these limitations may be due to the use of manufacturers’ cut-off values for positive/negative derived for the investigation of thyroid autoimmunity. Comparison of four automated TgAb assays and receiver operator curve analysis enabled Algeciras-Schimnich et al.⁷⁷ to derive cut-off values which gave comparable performance for identifying false-negative Tg IMA results. These were: Beckman Access <4 IU/mL, Siemens Centaur (Siemen Diagnostics, Los Angeles, CA, USA) 44 IU/L and Roche (West Sussex, UK) 22 IU/L, though a comparable cut-off could not be derived for a fourth assay. Of course, these cut-off values may be specific for interference in their Tg assay alone. Spencer et al.⁷⁸ recommended the use of the TgAb assay analytical sensitivity, rather than the reference range. Even so, they reported that for two of the assays investigated, interfering antibodies were not identified in 20–30% of cases. A similar lack of concordance has been reported for TgAb assays currently in use in the UK,⁷⁹ although the concordance achieved using the manufacturers’ reference data was 74% and could be increased to 90% by adjusting the cut-off values. Thus, the re-assurance of a negative TgAb and thus exclusion of a false-negative serum Tg has to be uncertain. Similarly, some laboratories have taken the approach of not reporting Tg results in patients who are TgAb ‘positive’ though this may give false reassurance when TgAbs are not detected and the possibility of assay interference remains.

TgAbs can also be used as a marker of disease recurrence.^4,6,75 The previously reported prevalence of TgAbs in DTC patients of 10–25% is higher than that of the general population. TgAbs decline after surgery with approximately half of patients in remission becoming TgAb ‘negative’ at 1.5 years,⁸⁰ though the decline may continue for up to three years post-surgery.⁸¹ Interpretation may be complicated by the presence of thyroiditis at the time of diagnosis and thyroid remnant ablation. Thus, a persistently detected or rising concentration of TgAbs may be more indicative of remaining disease or of recurrence than a single result. What is clear from a consideration of the analytical limitations of TgAb assays is that results from different assays cannot be used interchangeably.

Recovery of added Tg

Under-recovery of added Tg to a patient sample may indicate interference by TgAbs. Recovery has been reported to be dependent on the type and concentration of added Tg and on incubation conditions⁸ and thus has not been universally recommended. However, given the limitations now noted for TgAb assays themselves, measurement of recovery may aid interpretation in difficult cases.⁸²

IMA-RIA discordance

Given that TgAb interference, if present, is generally reported as giving rise to false-positive results by RIAs, and false-negative result in IMAs, a demonstrated discordance in results would indicate the possibility of assay interference in either or both assays. Any definition of discordance needs to take into account the bias of both assays in the absence of interference and importantly that the bias may change with changes in reagent lot. Of particular clinical significance is the discordance of IMA-RIA results at low concentrations. A falsely undetectable Tg by IMAs may result in a delay in detection of disease recurrence. A false detectable Tg by an RIA may prompt further unnecessary investigation. Thus, knowing the effect of assay bias at low concentrations is critical in deciding whether there is IMA-RIA discordance.

To address this, Spencer et al.⁷⁸ have estimated the ratio of serum Tg measured by IMA (Beckman Access) to that measured by an in-house RIA in 24 samples that were negative for TgAbs by four methods (RSR [Cardiff, Wales], Roche, Immulite and Access), i.e. samples where TgAb interference was unlikely. The mean IMA:RIA ratio was 90% (95% confidence interval = 75–110%). Thus, discordance could be defined if the ratio fell outside that range.

An alternative is to take the regression line for the comparison of the two methods with TgAb-negative samples and calculate the standard error of that line. Discordance would be indicated if the results fell outside of those limits. However, the regression line is likely to be unduly influenced by high Tg concentrations and may be less useful at low concentrations where the intercept of the regression line might have a significant impact.

The analytical or functional sensitivity of an assay is commonly used as a basis for clinical cut-off values, whether as quoted by the manufacturer or determined by experiment. This forms the basis of classification of a result as detectable/undetectable and of IMA-RIA discordance. As noted by Ross et al.,⁸³ bias (and changes in bias) near the zero point for automated Tg IMAs may be ‘hidden’ as instrument settings do not permit a measured signal to be reported below a certain level. They used an alternative approach of determining upper reference limits in a group of DTC patients who had undetectable TgAbs and undetectable Tg when TSH was suppressed during T4 therapy. In order to assess positive and negative bias near zero concentrations, they used an in-house computer programme for calibration which allowed the calculation of ‘negative’ concentrations. Thus, negative bias could be determined. They used a one-sided 99.9% confidence interval for the Tg-negative serum samples to determine an upper reference limit for each of six IMAs. This approach may minimize the effects of assay bias at low concentrations and allow more accurate determination of IMA-RIA discordance.

The question remains as to how these approaches to the investigation of antibody interference can be used in routine laboratory practice. A pragmatic approach is to indicate the potential error in either or both assays and continue to follow up with both assays and with TgAb monitoring for a change or trend. In some countries, there may be limited availability of an RIA. Differences in calibration, bias and functional sensitivity may make the use of IMA-RIA discordance more difficult at low Tg concentrations.

Heterophilic antibodies

Heterophilic antibody (HAB) interference in Tg IMAs generally results in false elevations. A prevalence of 3% was found in patients with an initial serum Tg ≥ 1 μg/L using the Beckman Access or Access 2 and the authors recommended the routine use of HAB blocking tubes to overcome this effect.⁸⁴ This type of interference and its prevalence is likely to be assay-dependent, having been reported for the Brahms Tg-plus immunoradiometric assay,⁸⁵ Immulite assay⁸⁶ and the CisBio (Bedford, MA, USA) and Access Tg IMA.⁸⁷ Thus, if a laboratory using an IMA obtains detectable serum Tg results that are inconsistent with the clinical picture or unexpected changes with time, linearity checks and/or treatment with HAB blocking tubes should be used.

The problems of TgAb and HAB interference and the care required in interpretation are exemplified in a recent study⁸² of 288 consecutive DTC patients, 47 of whom were found to have an undetectable serum Tg by IMAs but had residual ¹³¹I uptake on post-ablation WBS. To investigate the possibility of antibody interference, the authors measured TgAbs by two different assays, performed Tg recovery experiments, pretreated samples with HAB blocking tubes and measured Tg by two other IMAs. Of the 47 patients, 10 were found to have TgAbs and one HAB causing a negative interference. Of the remaining 36 patients, 18 were found to have a detectable Tg by a second IMA and 18 a detectable Tg by a third IMA. Thus, after extensive laboratory investigation, of the original 47 patients with an initial undetectable serum Tg with residual ¹³¹I uptake, only nine were confirmed to have an undetectable serum Tg with no evidence of assay interference. The authors noted that their investigations were extensive and expensive such that the treatment of such difficult DTC patients should be transferred to specialist referral centres.

Given the limitations of both Tg and TgAb assays and the costs of laboratory investigations, it is likely that no single analytical strategy will necessarily identify all samples exhibiting Tg assay interference and consideration of other imaging techniques and the broader clinical picture is essential. As a minimum, it is suggested that all reports should comment on the possibility of assay interference. There are then a number of options: (1) measurement of TgAbs; (2) measurement of Tg by another IMA if the first assay is an IMA; (3) measurement of Tg by an RIA if the first assay is an IMA; (4) measurement of Tg by an IMA if the first assay is an RIA and a false-positive is suspected; (5) measurement of TgAbs by another method; (6) recovery of added Tg experiments; (7) investigation of possible HAB interference; and (8) assessment of linearity on dilution. The optimal strategy is likely to be dependent on the assays available and the feasibility and costs involved.

Conclusions

Serum Tg is a tumour marker used for monitoring treatment of DTC. Measurement is of little value pre-thyroidectomy and samples should not be collected until at least six weeks post-thyroidectomy or radioiodine ablation. Interpretation of serum Tg results requires knowledge of the extent of thyroid surgery and concurrent serum TSH concentrations. Results should not be related to a reference range obtained from euthyroid subjects with an intact thyroid.

There are important analytical considerations, namely of assay bias and interference by endogenous TgAbs, which clinical staff should be made aware of. Most assays are reported to be calibrated against the international standard CRM 457; yet, there is a significant lack of agreement between assay results. Patients should be monitored using the same assays for serum Tg. Between-assay variability of TgAbs determines that the same TgAb assay should be used.

There are limitations in classifying samples as TgAb-positive or -negative (with respect to predicting interference in the Tg assay) and a numerical result should be reported, though interference in the Tg assay cannot be ruled out on the basis of TgAb status alone. Analysis by another method (IMA or RIA) and investigation for heterophilic antibodies are alternative approaches. Each laboratory should establish their own protocol for the investigation of possible interference.

DECLARATIONS

Competing interests: None.

Funding: No funding was received for this review.

Ethical approval: Not required.

Guarantor: PC.

Contributorship: Both authors contributed to all stages of the preparation of the review.

References

Cancer Research UK. Cancer Statistics 2007. See http://infocancerresearchuk.org/cancerstats/types/thyroid/incidence/ (last checked 27 July 2010)

Tunbridge

, Evered

, Hall

, The spectrum of thyroid disease in the community: the Whickham survey. Clin Endocrinol 1977;7:481–93

Cooper

, Doherty

, Haugen

, Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 2006;16:109–42

Cooper

, Doherty

, Haugen

, Revised American Thyroid Association Guidelines for management of thyroid nodules and differentiated thyroid cancer: progress on the road from consensus to evidence-based practice. Thyroid 2009;19:1167–214

Perros

, Clarke

SEM

, Franklyn

, In: Perros

, ed. British Thyroid Association and Royal College of Physicians Guidelines for the Management of Thyroid Cancer. 2nd edn. 2007. See http://www.british-thyroid-association.org/Guidelines/ (last checked 7 February 2012)

Pacini

, Schlumberger

, Dralle

, Elisei

, Smit

JWA

, Wiersinga

and the European Thyroid Cancer Taskforce. European consensus for the management of patients with differentiated thyroid carcinoma of the follicular epithelium. Eur J Endocrinol 2006;154:787–803

Black

, Sheppard

, Hoffenberg

. Serial serum thyroglobulin measurements in the management of differentiated thyroid carcinoma. Clin Endocrinol (Oxf) 1987;27:115–20

Spencer

, Takeuchi

, Kazarosyan

, Serum thyroglobulin autoantibodies: prevalence, influence on serum thyroglobulin measurement and prognostic significance in patients with differentiated thyroid cancer. J Clin Endocrinol Metab 1998;83:1121–7

Clark

. Laboratory services for thyroglobulin and implications for monitoring of differentiated thyroid cancer. J Clin Pathol 2009;62: 402–6

10.

Spencer

, Lopresti

. Measuring thyroglobulin and thyroglobulin autoantibody in patients with differentiated thyroid cancer. Nat Clin Pract Endocrinol Metab 2008;4:223–33

11.

Kebebew

, Treseler

, Ituarte

, Clark

. Coexisting chronic lymphocytic thyroiditis and papillary thyroid cancer revisited. World J Surg 2001;25:632–7

12.

Gerenova

, Buysschaert

, de Burbure

, Daumerie

. Prevalence of thyroid cancer in Graves’ disease: a retrospective study of a cohort of 103 patients treated surgically. Eur J Intern Med 2003;14:321–5

13.

Pacini

, Pinchera

, Giani

, Grasso

, Doveri

, Baschieri

. Serum thyroglobulin in thyroid carcinoma and other thyroid disorders. J Endocrinol Invest 1980;3:283–92

14.

Snozek

CLH

, Chambers

, Reading

, Serum thyroglobulin, high-resolution ultrasound and lymph node thyroglobulin in diagnosis of differentiated thyroid carcinoma nodal metastases. J Clin Endocrinol Metab 2007;92:4278–81

15.

Boi

, Baghino

, Alzeni

, Lai

, Faa

, Mariotti

. The diagnostic value for differentiated thyroid carcinoma metastases of thyroglobulin (Tg) measurement in washout fluid from fine-needle aspiration biopsy of neck lymph nodes is maintained in the presence of circulating anti-Tg antibodies. J Clin Endocrinol Metab 2006;91:1364–9

16.

Frasoldati

, Toschi

, Zini

, Role of thyroglobulin measurement in fine-needle aspiration biopsies of cervical lymph nodes in patients with differentiated thyroid cancer. Thyroid 1999;9:105–11

17.

Borel

A-L

, Boizel

, Faure

, Significance of low levels of thyroglobulin in fine needle aspirates from cervical lymph nodes of patients with a history of differentiated thyroid cancer. Eur J Endocrinol 2008;158:691–8

18.

Edmonds

, Willis

. Serum thyroglobulin in the investigation of patients presenting with metastases. Br J Radiol 1988;61:317–9

19.

Torrigiani

, Doniach

, Roitt

. Serum thyroglobulin levels in healthy subjects and in patients with thyroid disease. J Clin Endocrinol 1968;29:305–14

20.

BNF 61 Thyrotropin Alfa. See www.bnf.org (last checked 18 July 2011)

21.

Torres

, Ramirez

, Simkin

, Braverman

, Emerson

. Effect of various doses of recombinant human thyrotropin on the thyroid radioactive iodine uptake and serum levels of thyroid hormones and thyroglobulin in normal subjects. J Clin Endocrinol Metab 2001;86:1660–4

22.

Haugen

, Pacini

, Reiners

, A comparison of recombinant human thyrotropin and thyroid hormone withdrawal for the detection of thyroid remnant or cancer. J Clin Endocrinol Metab 1999;84:3877–85

23.

Spencer

, Wang

. Thyroglobulin measurements: techniques, clinical benefits and pitfalls. Endocrinol Metab Clin North Am 1995;24:841–63

24.

Robbins

, Robbins

. Recombinant human thyrotropin and thyroid cancer management. J Clin Endocrinol Metab 2003;88:1933–8

25.

Pacini

, Castagna

. Diagnostic and therapeutic use of recombinant human TSH (rhTSH) in differentiated thyroid cancer. Best Pract Res Clin Endocrinol Metab 2008;22:1009–21

26.

Robbins

, Chon

, Fleisher

Is the serum thyroglobulin response to recombinant human thyrotropin sufficient, by itself, to monitor for residual thyroid carcinoma?

J Clin Endocrinol Metab 2002;87:3242–7

27.

Mariani

, Ferdeghini

, Augeri

, Clinical experience with recombinant human thyrotrophin (rhTSH) in the management of patients with differentiated thyroid cancer. Cancer Biother Radiopharmaceut 2000;15:211–7

28.

Robbins

, Tuttle

, Sharaf

, Preparation by recombinant human thyrotropin or thyroid hormone withdrawal are comparable for the detection of residual differentiated thyroid carcinoma. J Clin Endocrinol Metab 2001;86:619–25

29.

Pacini

, Molinaro

, Lippi

, Prediction of disease status by recombinant human TSH-stimulated serum Tg in the postsurgical follow-up of differentiated thyroid carcinoma. J Clin Endocrinol Metab 2001;86:5686–90

30.

Mazzaferri

, Kloos

. Is diagnostic Iodine-131 scanning with recombinant human TSH useful in the follow-up of differentiated thyroid cancer after thyroids ablation. J Clin Endocrinol Metab 2002;87:1490–8

31.

Kloos

, Mazzaferri

. A single recombinant human thyrotropin-stimulated serum thyroglobulin measurement predicts differentiated thyroid carcinoma metastases three to five years later. J Clin Endocrinol Metab 2005;90:5047–57

32.

Smallridge

, Meek

, Morgan

Monitoring thyroglobulin in a sensitive immunoassay has comparable sensitivity to recombinant human TSH-stimulated thyroglobulin in follow-up of thyroid cancer patients

. J Clin Endocrinol Metab 2007;92:82–7

33.

Malandrino

, Latina

, Marescalco

, Risk-adapted management of differentiated thyroid cancer assessed by a sensitive measurement of basal serum thyroglobulin. J Clin Endocrinol Metab 2011;96:1703–9

34.

Brassard

, Borget

, Edet-Sanson

, Long-term follow-up of patients with papillary and follicular thyroid cancer: a prospective study on 715 patients. J Clin Endocrinol Metab 2011;96:1352–9

35.

Iervasi

, Iervasi

, Ferdeghini

Clinical relevance of highly sensitive Tg assay in monitoring patients treated for differentiated thyroid cancer

. Clin Endocrinol (Oxf) 2007;67:434–41

36.

Schlumberger

, Hitzel

, Toubert

, Comparison of seven serum thyroglobulin assays in the follow-up of papillary and follicular thyroid cancer patients. J Clin Endocrinol Metab 2007;92:2487–95

37.

Feldt-Rasmussen

, Profilis

, Colinet

, Human thyroglobulin reference material (CRM 457). 1st Part: assessment of homogeneity, stability and immunoreactivity. Ann Biol Clin (Paris) 1996;54:337–42

38.

Feldt-Rasmussen

, Profilis

, Colinet

, Human thyroglobulin reference material (CRM 457). 2nd Part: Physicochemical characterization and certification. Ann Biol Clin (Paris) 1996;54:343–8

39.

Powell

, Hudson

, Lambirth

, Recombinant expression of homodimeric 660 kDa human thyroglobulin in soybean seeds: an alternative source of human thyroglobulin. Plant Cell Rep 2011;30:1327–38

40.

MacKenzie

. UKNEQAS Thyroglobulin Surveys. March 2011

41.

Mazzaferri

. Will highly sensitive thyroglobulin assays change the management of thyroid cancer? Clin Endocrinol (Oxf) 2007;67:321–3

42.

Dittadi

, Peloso

, Gion

. Within-subject biological variation in disease: the case of tumour markers. Ann Clin Biochem 2008;45:226–8

43.

Jensen

, Petersen

, Blaabjerg

, Biological variation of thyroid autoantibodies and thyroglobulin. Clin Chem Lab Med 2007;45:1058–64

44.

Miyauchi

, Kudo

, Miya

, Prognostic impact of serum thyroglobulin doubling-time under thyrotropin suppression in patients with papillary thyroid carcinoma who underwent total thyroidectomy. Thyroid 2011;21:707–16

45.

Hocevar

, Auersperg

, Stanovnik

. The dynamics of serum thyroglobulin elimination from the body after thyroid surgery. Eur J Surg Oncol 1997;23:208–10

46.

Oduncu

, Kimmig

, Hepp

, Emmerich

. Cancer in pregnancy: maternal-fetal conflict. J Cancer Res Clin Oncol 2003;129:133–46

47.

Holt

. Care of the pregnant thyroid cancer patient. Curr Opin Oncol 2010;22:1–5

48.

Abalovich

, Amino

, Barbour

, Management of thyroid dysfunction during pregnancy and postpartum: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 2007;92 (Suppl.):S1–47

49.

Mazzaferri

. Approach to the pregnant patient with thyroid cancer. J Clin Endocrinol Metab 2011;96:265–72

50.

Mandel

, Spencer

, Hollowell

. Are detection and treatment of thyroid insufficiency in pregnancy feasible? Thyroid 2005;15:44–53

51.

Nakamura

, Sakata

, Komaki

, Serum thyroglobulin concentrations in normal pregnancy. Endocrinol Jpn 1984;31:675–9

52.

Soldin

, Tractenberg

, Hollowell

, Jonklaas

, Janicic

, Soldin

. Trimester-specific changes in maternal thyroid hormone, thyrotropin and thyroglobulin concentrations during gestation: trends and associations across trimesters in iodine sufficiency. Thyroid 2004;14:1084–90

53.

Eltom

, Elnagar

, Elbagir

, Gebre-Medhin

. Thyroglobulin in serum as an indicator of iodine status during pregnancy. Scand J Clin Lab Invest 2000;60:1–7

54.

Rosen

, Korman

, Walfish

. Thyroid nodular disease in pregnancy: current diagnosis and management. Clin Obstet Gynaecol 1997;40:81–9

55.

Josefson

, Zimmerman

. Thyroid nodules and cancers in children . Pediatr Endocrinol Res 2008;6:14–23

56.

Penny

, Spencer

, Frasier

, Nicoloff

. Thyroid-stimulating hormone and thyroglobulin levels decrease with chronological age in children and adolescents. J Clin Endocrinol Metab 1983;56:177–9

57.

Pezzino

, Filetti

, Belfiore

, Proto

, Donzelli

, Vigneri

. Serum thyroglobulin levels in the newborn. J Clin Endocrinol Metab 1981;52:364–6

58.

Eltom

, Eltom

, Idris

, Gebre-Medhin

. Thyroid function in the newborn in relation to maternal thyroid status during labour in a mild iodine deficiency endemic area in Sudan. Clinical Endocrinol 2001;55:485–90

59.

Ares

, Escobar-Morreale

, Queri

, Neonatal hypothyroxinemia: effects of iodine intake and premature birth. J Clin Endocrinol Metab 1997;82:1704–12

60.

Kok

, Tegelaers

WHH

, Vijlder

JJM

. Serum thyroglobulin in preterm infants with and without respiratory distress syndrome. II A longitudinal study during the first three weeks of life. Pediatr Res 1986;20:1001–3

61.

UK National Newborn Screening Programme Centre. Newborn Blood Spot Screening in the UK. Policies and Standards. London: Department of Health; 2005

62.

Grűters

, Krude

, Biebermann

. Molecular genetic defects in congenital hypothyroidism. Eur J Endocrinol 2004;151:U39–U44

63.

Ilicki

, Ericsson

U-B

, Larsson

, The value of neonatal serum thyroglobulin determinations in the follow-up of patients with congenital hypothyroidism. Acta Paediatr Scand 1990;79:769–75

64.

Beltrão

, Juliano

, Chammas

, Etiology of congenital hypothyroidism using thyroglobulin and ultrasound combination. Endocrine J 2010;57:587–93

65.

Clerc

, Monpeyssen

, Chevalier

, Scintigraphic imaging of paediatric thyroid dysfunction. Hormone Res 2008;70:1–13

66.

Mariotti

, Martino

, Cupini

, Low serum thyroglobulin as a clue to the diagnosis of thyrotoxicosis factitia. N Engl J Med 1982;307:410–2

67.

Schneider

, Pervos

. Radioimmunoassay of human thyroglobulin:effect of antithyroglobulin autoantibodies. J Clin Endocrinol Metab 1978;47:126–37

68.

Tozzoli

, Bizzaro

, Tonutti

, Immunoassay of anti-thyroid autoantibodies: high analytical variability in second generation methods. Clin Chem Lab Med 2002;40:568–73

69.

Spencer

, Bergoglio

, Kazarosyan

, Fatemi

, LoPresti

. Clinical impact of thyroglobulin (Tg) and Tg autoantibody method differences on the management of patients with differentiated thyroid carcinomas. J Clin Endocrinol Metab 2005;90:5566–75

70.

Okosieme

, Evans

, Moss

, Thyroglobulin antibodies in serum of patients with differentiated thyroid cancer: relationship between epitope specificities and thyroglobulin recovery. Clin Chem 2005;51:729–34

71.

Gao

, Yuan

, Yu

, Lu

. Mutual interference between serum thyroglobulin and antithyroglobulin antibody in an automated chemiluminescent immunoassay. Clin Biochem 2007;40:735–8

72.

Aras

, Gültekin

, Küçük

. The additive clinical value of combined thyroglobulin and antithyroglobulin antibody measurements to define persistent and recurrent disease in patients with differentiated thyroid carcinoma. Nucl Med Commun 2008;29:880–4

73.

Seo

, Lee

, Ahn

, Lee

. Recurrence detection in differentiated thyroid cancer patients with elevated serum level of antithyroglobulin antibody: special emphasis on using (18)F-FDG PET/CT. Clin Endocrinol (Oxf) 2010;72:558–63

74.

Chung

, Park

, Kim

, Clinical significance of elevated level of serum antithyroglobulin antibody in patients with differentiated thyroid cancer after thyroid ablation. Clin Endocrinol 2002;57:215–21

75.

Kim

, Yoon

, Kim

, Change of serum antithyroglobulin antibody levels is useful for prediction of clinical recurrence in thyroglobulin-negative patients with differentiated thyroid carcinoma. J Clin Endocrinol Metab 2008;93:4683–9

76.

Rubello

, Casara

, Girelli

, Piccolo

, Busnardo

. Clinical meaning of circulating antithyroglobulin antibodies in differentiated thyroid cancer: A prospective study. J Nucl Med 1992;33:1478–80

77.

Algeciras-Schimnich

, Lasho

, Ness

, The Roche Elecsys and Siemens-Centaur thyroglobulin autoantibody assays show comparable clinical performance to the recently unavailable Beckman-Coulter Access thyroglobulin autoantibody assay in identifying samples with potentially false-low thyroglobulin measurements due to thyroglobulin autoantibody interference. Thyroid 2011;21:813–4

78.

Spencer

, Petrovic

, Fatemi

. Current thyroglobulin autoantibody (TgAb) assays often fail to detect interfering TgAb that can result in the reporting of falsely low/undetectable serum Tg IMA values for patients with differentiated thyroid cancer. J Clin Endocrinol Metab 2011;96:1283–91

79.

Taylor

, Parkington

, Bradbury

, Simpson

, Jefferies

, Halsall

. Concordance between serum thyroglobulin antibody assays. Ann Clin Biochem 2011;48:367–9

80.

Gorges

, Maniecki

, Jentzen

, Development and clinical impact of thyroglobulin antibodies in patients with differentiated thyroid carcinoma during the first 3 years after thyroidectomy. Eur J Endocrinol 2005;153:49–55

81.

Chiovato

, Latrofa

, Braverman

, Disappearance of humoral thyroid autoimmunity after complete removal of thyroid antigens. Ann Intern Med 2003;139:346–51

82.

Giovanella

, Suriano

, Ceriani

, Undetectable thyroglobulin in patients with differentiated thyroid carcinoma and residual radioiodine uptake on a postablation whole-body scan. Clin Nucl Med 2011;36:109–12

83.

Ross

, Netea-Maier

, Schakenraad

, Bravenboer

, Hermus

ARMM

, Sweep

FCGJ

. Assay bias may invalidate decision limits and affect comparability of serum thyroglobulin assay methods: an approach to reduce interpretation differences. Clin Chim Acta 2008;394:104–9

84.

Preissner

, O'Kane

, Singh

, Morris

, Grebe

SKG

. Phantoms in the assay tube: heterophile antibody interferences in serum thyroglobulin assays. J Clin Endocrinol Metab 2003;88:3069–74

85.

Verburg

, Wäschle

, Reiners

, Giovanella

, Lentjes

EGWM

. Heterophile antibodies rarely influence the measurement of thyroglobulin and thyroglobulin antibodies in differentiated thyroid cancer . Hormone Metab Res 2010;42:736–9

86.

Giovanella

, Ghelfo

. Undetectable serum thyroglobulin due to negative interference of heterophile antibodies in relapsing thyroid carcinoma. Clin Chem 2007;53:1871–2

87.

Massart

, Corcuff

J-B

, Bordenave

. False-positive results corrected by the use of heterophilic antibody-blocking reagent in thyroglobulin immunoassays. Clin Chim Acta 2008;388:211–3

Can we interpret serum thyroglobulin results?

Abstract

Introduction

Routine review of the patient with DTC

The thyroid nodule and nodular goitre

Metastatic disease – cervical lymphadenopathy

Metastatic disease – unknown primary

Preoperative/preradioiodine ablation Tg

Early measurement of Tg post-thyroid remnant ablation

Long-term follow-up

Recombinant thyrotropin

Protocol

Interpretation

Biological variability

In pregnancy

In children

Other indications for the measurement of serum Tg

Is the result correct?

Endogenous antibodies to Tg

Recovery of added Tg

IMA-RIA discordance

Heterophilic antibodies

Conclusions

DECLARATIONS

References