Differentiated thyroid cancer: Role of serum thyroglobulin

INTRODUCTION — Thyroglobulin (Tg) is a storage form of thyroxine (T4) and triiodothyronine (T3). It is synthesized only by thyroid follicular cells and released into serum along with the thyroid hormones. Given the cellular specificity of Tg, its detection in biopsy specimens provides proof of the thyroid origin of the tissue. In addition, measurements of serum Tg provide important information about the presence or absence of residual, recurrent, or metastatic disease in patients with differentiated thyroid cancer. This topic review will describe the methods for measuring serum Tg and the use of serum Tg measurements in patients with differentiated thyroid cancer. Other aspects of the management of thyroid cancer are reviewed separately. (See "Differentiated thyroid cancer: Overview of management".)

THYROGLOBULIN ASSAY — Testing of serum thyroglobulin (Tg) should be done using a sensitive assay, ideally using the same assay for each sample.

Methodology — Serum Tg is now generally measured by two-antibody "sandwich" immunometric assays (the antigen is sandwiched between the two antibodies) in which the capture antibody is bound to a solid support and the detection antibody is labeled with either an isotopic (immunoradiometric assay, IRMA) or non-isotopic (usually immunochemiluminescent assay, ICMA) marker. The values in normal subjects in most laboratories range from 1 to approximately 30 ng/mL. These immunometric assays are quicker, readily automated, and have greater sensitivity (0.1 to 1 ng/mL) than most radioimmunoassays [1].

Interassay variation — We recommend that serial Tg measurements in thyroid cancer patients be done using the same assay.

Despite a trend toward assay standardization, serum Tg values obtained with different assays cannot be directly compared, as interassay variability remains substantial [2-4]. The variability in assay results is due to [1,5-7]:

●Variations in the anti-Tg antibodies used.

●The heterogeneity of Tg, a consequence of alternative processing and differences in iodination of Tg.

●Tg produced by thyroid cancer cells can be even more heterogeneous (because of dysregulation of the enzymatic glycosylation and iodination within malignant thyroid cells) and occasionally has enough conformational difference that it may not be recognized by a standard Tg assay.

The net effect can be widely variable antigen (Tg) detection among different assays.

Intraassay variation — Even using the same assay, between-run variability can affect the comparability of serial determinations over time [2,8]. While these differences are far less than the between-assay differences, they can be responsible for small fluctuations in Tg measurements over time within the same patient.

Functional sensitivity — Functional sensitivity is defined as the lowest Tg concentration that an assay can reliably and consistently measure under clinically relevant conditions with less than 20 percent coefficient of variation. For many years, the functional sensitivity of most Tg assays had been approximately 0.9 ng/mL. However, several assays with functional sensitivities of 0.1 to 0.2 ng/mL are commercially available [4,9,10].

To further enhance the sensitivity of serum Tg in the detection of persistent/recurrent thyroid cancer, serum Tg levels can be measured during thyroid-stimulating hormone (TSH) stimulation (either thyroid hormone withdrawal or with recombinant human TSH [rhTSH, thyrotropin alfa] administration) [11]. When using the less sensitive assays (functional sensitivities of approximately 1 ng/mL), TSH stimulation will result in a previously undetectable serum Tg value becoming measurable in as many as 20 to 25 percent of patients [12].

In the newer, more sensitive Tg assays, serum Tg concentrations (measured while receiving levothyroxine [T4, thyroxine] suppression therapy) correlate with rhTSH-stimulated Tg concentrations and, therefore, decrease the need for rhTSH-stimulated measurements [4,9,13-16]. This was illustrated in a study of 849 Tg antibody-negative patients being monitored after treatment for differentiated thyroid cancer [9]. Patients with a TSH-suppressed serum Tg concentration <0.1 ng/mL (measured with an assay with a functional sensitivity of 0.05 ng/mL) were unlikely to have an rhTSH stimulated Tg above 2.0 ng/mL. Similar findings were noted in a study of 178 low-risk patients that compared basal and post thyroid hormone withdrawal Tg levels. Basal serum Tg levels were <0.1 ng/mL in 130 patients. After withdrawal of thyroid hormone, 5 of 130 (3.8 percent) had a Tg >1 ng/mL and recurrence was diagnosed in only one patient. Among the 48 patients with Tg >0.1 ng/mL, 42 percent had Tg >1 ng/mL after withdrawal and 11 percent had recurrences [15].

Hook effect — Occasionally, immunometric assays may fail to detect very high serum Tg concentrations due to the so-called "hook effect," in which extremely high concentrations of Tg bind to each antibody, preventing the formation of the two-antibody sandwich upon which the assay depends [17]. If this effect is suspected, the sample should be reanalyzed after dilution to obtain a reliable Tg measurement.

Heterophilic antibodies — False-positive Tg results have been reported as a result of interference by heterophilic anti-mouse antibodies (HAMA) in an immunometric assay [18]. In this report, approximately 3 percent of the 1100 blood specimens tested had substantial differences in the measured Tg value before and after the specimens were treated to block potential heterophile antibody interference.

While an uncommon clinical problem, the presence of HAMA antibody interference should be considered in patients in whom an elevated Tg does not seem appropriate for their clinical condition [19]. Since radioimmunoassay is very unlikely to be affected by HAMA antibodies, repeat determination with a radioimmunoassay technique is likely to reveal the true serum Tg level [20].

Potential assay interference with biotin — Some automated assays utilizing a biotin-streptavidin separation system report falsely low or falsely high Tg levels. As an example, in one report, biotin (10 mg/day) was associated with a falsely lower Tg value, with maximal biotin interference observed two hours after ingestion of the biotin [21]. While additional studies are needed to clarify the likelihood of biotin interference associated with various biotin doses commonly being used as over-the-counter products in an effort to improve hair and joint health, repeating Tg values in the absence of biotin supplementation for at least two days can be considered if the measured Tg value is not consistent with the clinical scenario.

THYROGLOBULIN ANTIBODIES — We recommend measuring antithyroglobulin (anti-Tg) antibodies with each measurement of serum Tg. Anti-Tg antibodies are detectable in as many as 10 percent of the general population [22] and 20 percent of thyroid cancer patients [1,23]. These antibodies represent a challenge, because thyroglobulin (Tg) values obtained in the presence of anti-Tg antibodies may not be clinically reliable. Just as with serum Tg measurements, anti-Tg antibody levels are dependent upon the assay used, and serial values need to be compared within the same assay.

In patients with anti-Tg antibodies, serum Tg concentrations alone cannot be used as a marker to detect persistent or recurrent disease after thyroidectomy and ablation of residual normal thyroid tissue (see 'Monitoring response to therapy' below). Nevertheless, serum Tg and anti-Tg antibodies should be measured as in patients without Tg antibodies because disease recurrence can be heralded by a rise in Tg antibodies with or without a corresponding rise in serum Tg. (See 'Surrogate tumor marker' below.)

Impact on Tg assay — Because anti-Tg antibodies can have a major impact on the measurement of serum Tg, all laboratories that measure serum Tg should test for anti-Tg antibodies in any serum sample submitted for Tg assay [24].

By binding with the circulating serum Tg, anti-Tg antibodies may decrease the amount of unbound (free) Tg available for detection. Since immunometric assay systems appear to detect only the unbound (free) Tg, this effect can result in falsely low values. The magnitude of the interference is most noticeable at low Tg levels where the immunometric assays can falsely report an undetectable serum Tg if the anti-Tg antibodies are competing for the low-level circulating Tg molecules. This effect is not abrogated by use of monoclonal antibodies directed against epitopes of Tg that do not react with the autoantibodies [25]. Even very low anti-Tg antibody levels can interfere with the Tg assay performance. There does not appear to be a threshold value of anti-Tg antibodies that is predictive of antibody interference [1]. In some anti-Tg antibody assays, more reliable results are obtained if any level above the analytic sensitivity of the assay is considered to be Tg antibody-positive rather than using the manufacturer recommended cut-off levels [26].

Conversely, radioimmunoassays tend to report falsely high Tg values in the presence of anti-Tg antibodies. This is probably because, unlike the immunometric assay, the radioimmunoassay detects both unbound (free) Tg and bound Tg (complexed to anti-Tg antibodies). As a result, the Tg value reported may be higher than the actual "free" Tg circulating in the blood stream but is much less likely to be false low value. Because of this discordance between radioimmunoassay and immunometric assays, some experts have recommended that in the presence of anti-Tg antibodies, radioimmunoassays should be used to measure the serum Tg (recognizing the possibility of a false high result) but assuming a low value is likely to be real and reflect the clinical situation [20]. While this approach appears to be valid, radioimmunoassays are not widely available and still need to be interpreted with caution in this setting.

In an attempt to determine whether a circulating anti-Tg antibody is causing assay interference, several commercial labs offer Tg measurements by recovery assays, using standard immunometric assays to determine serum Tg levels before and after addition of a known amount of Tg. If 80 to 120 percent of the added quantity of Tg is detected, the antibodies are not considered to be interfering and a Tg value is reported. If the value falls outside that range, antibody interference is assumed and no Tg value is reported. While these assays have gained popularity in both the United States and Europe, many experts feel strongly that recovery assays are not reliable [20].

Assays using liquid chromatography-tandem mass spectrometry (LC-MS/MS) for the measurement of serum Tg are currently being evaluated. This technology uses trypsin to digest Tg bound to Tg antibody in order to release a specific target Tg peptide that can be enriched and subsequently detected by LC-MS/MS. While this assay was reported to detect Tg in the presence of anti-Tg antibodies [27], another study found that LC-MS/MS failed to detect Tg in at least 40 percent of Tg antibody-positive patients with known structural disease [28]. In addition, functional sensitivity (0.5 to 1.0 ng/mL) remains inferior to radioimmunoassays [28,29]. The cause for an undetectable MS Tg reading in the setting of known structural disease has not been determined but could be related to a wide variety of factors, including suboptimal functional sensitivity, poor Tg secretion by some tumors, tumor Tg polymorphisms, or enhanced metabolic clearance of serum Tg-anti-Tg antibody complexes [28-31].

Surrogate tumor marker — Following total thyroidectomy and radioiodine ablation, serum anti-Tg antibodies usually fall to undetectable levels over three to five years, while patients with persistent disease typically maintain detectable or rising anti-Tg. The decline in anti-Tg antibody levels has prognostic significance as a >50 percent decline during the first three years of follow-up is associated with a <3 percent risk of recurrence [30,32]. Stable values are associated with an approximately 20 percent risk of recurrence and rising values with approximately 40 percent risk of recurrence.

In two studies of thyroid cancer patients with undetectable serum Tg concentrations, 18 and 49 percent of patients with serum anti-Tg antibody concentrations >100 units/mL had a recurrence, compared with only 1 and 3 percent of patients with serum anti-Tg antibody concentrations <100 units/mL [32,33]. Even in patients previously documented to be free of anti-Tg antibodies, disease recurrence can be heralded by a rise in Tg antibodies with or without a corresponding rise in serum Tg [20,33,34]. However, in a subsequent study evaluating 812 patients with undetectable Tg antibodies on at least three serial assays over a minimum of three years of follow-up, 5 percent (n = 40) of patients converted to having detectable Tg antibodies, which was not associated with an increased risk of structural disease recurrence when compared with the group that were consistently Tg antibody undetectable (n = 772 patients) [35].

Just as with serum Tg measurements, anti-Tg antibody levels are dependent upon the assay used, and serial values need to be compared within the same assay [36]. In patients with coexistent autoimmune thyroid disease at the time of surgery, anti-Tg antibodies may persist far longer. In a study of 116 patients with anti-Tg antibodies prior to thyroidectomy, antibodies remained detectable for up to 20 years in some patients without detectable thyroid tissue, and the median time to disappearance of antibodies was three years [37].

MONITORING RESPONSE TO THERAPY

When to measure — Serum thyroglobulin (Tg) levels are used to monitor patients with differentiated thyroid cancer for persistent or recurrent disease after initial therapy (thyroidectomy with or without radioiodine ablation). Serum Tg can be measured while taking suppressive doses of thyroid hormone (TSH-suppressed) or with TSH stimulation (after thyroid hormone withdrawal or after administration of rhTSH). An important caveat pertains to patients in whom rhTSH, instead of withdrawal, is used for remnant ablation. In that protocol, radioiodine is given 48 hours before the measurement of the "stimulated" Tg, which may cause release of Tg from the remnants, potentially causing spuriously higher values [38].

A suggested schedule for Tg monitoring during the first year after thyroid surgery (table 1) as well as for long-term monitoring of patients with differentiated thyroid cancer (table 2) is discussed in more detail separately. (See "Differentiated thyroid cancer: Overview of management", section on 'Monitoring response to therapy'.)

The three major determinants of the serum Tg concentration are [20]:

●The mass of thyroid tissue present (both normal and malignant thyroid cells)

●The presence of injury to the thyroid (eg, after fine-needle aspiration [FNA], thyroidectomy or radioiodine therapy, or during thyroiditis)

●The degree of TSH receptor stimulation (eg, endogenous TSH, rhTSH [thyrotropin alfa], serum human chorionic gonadotropin [hCG], TSH receptor antibodies associated with autoimmune thyroid disease)

Detecting persistent thyroid cancer — Serum Tg measurements (post thyroidectomy) are useful for detecting persistent disease after initial treatment of differentiated thyroid cancer. Serum Tg measured before thyroidectomy is a reflection of both the normal and malignant thyroid cells and therefore is not a reliable predictor of thyroid cancer. Serum Tg is cleared with a half-life of approximately 30 hours following thyroidectomy [39].

The interpretation of the serum Tg level depends upon the initial therapy. (See "Differentiated thyroid cancer: Overview of management", section on 'Serum thyroglobulin measurements'.)

In patients with metastatic thyroid cancer, the serum Tg appears to be influenced by the volume of disease, the specific histology of the lesion, and the anatomic location of the metastasis. As an example, in a retrospective study of 417 thyroid cancer patients, basal Tg concentrations (TSH-suppressed) correlated with the volume and location of metastases (higher with bone and/or lung than cervical metastases), and with the histologic type of cancer (higher levels in follicular and Hürthle cell cancers compared with papillary cancers) [40]. Stimulated Tg concentrations (in response to recombinant TSH) correlated only with histologic type of cancer (highest in papillary and lowest in Hürthle cell cancers).

A meta-analysis found that the sensitivity and specificity of Tg measurement to detect persistent thyroid cancer was 96 and 95 percent, respectively, after thyroid hormone withdrawal; 93 and 88 percent, respectively, after rhTSH; and 78 and 98 percent, respectively, when measured while taking suppressive doses of thyroid hormone [41]. Combining rhTSH stimulation with cervical ultrasound improved sensitivity and negative predictive value to 93 and 99 percent, respectively, in a study of 340 consecutive patients [42]. These data indicate that neck ultrasonography will occasionally identify structural disease even when the Tg is undetectable.

Detecting recurrent thyroid cancer — Because serum Tg concentration is dependent on the mass of thyroid tissue present (both normal and malignant), serum Tg values have the highest sensitivity and specificity for detection of recurrent disease after the patient has had a total thyroidectomy and radioiodine ablation of any microscopic residual normal thyroid cells remaining after total thyroidectomy [43-45]. However, serum Tg may be useful for identifying persistent or metastatic disease in patient undergoing thyroidectomy without radioiodine. In a systematic review of studies examining the utility of serum Tg for detection of recurrent disease in patients who had total or near-total thyroidectomy without radioiodine, Tg levels less than 1 to 2.5 ng/mL appeared to identify patients at low risk for persistent or metastatic disease [46].

Because Tg production and release from the cell are significantly influenced by the degree of TSH stimulation, it is not surprising that Tg measurements are likely to be more sensitive for detection of recurrent disease following TSH stimulation (endogenous TSH or rhTSH) than when measured during TSH suppression (on T4). Nevertheless, others have utilized T4-suppressed Tg measurements along with ultrasound to assess for disease recurrence. In one such study of 495 patients, a detectable Tg while taking T4 indicated recurrence in 23 of 44 patients, while the addition of ultrasound detected 42 of the 44 recurrences [47].

Although initial measurement of rhTSH-stimulated Tg is useful, repeat rhTSH-stimulated Tg testing in patients whose first stimulated Tg is undetectable may not be helpful. As an example, in one study of 68 patients with an rhTSH-stimulated Tg that was initially <1 ng/mL at the time of remnant ablation, only 1 of 67 patients had a detectable stimulated value two to three years later [48].

Predictor of clinical outcomes — Serum Tg concentrations may predict disease-free remission at various times in the patient's course. The serum Tg concentration in low-risk patients after initial surgery while hypothyroid, just prior to administration of radioiodine for remnant ablation, has been correlated with patient outcomes [49-51]. As an example, in a meta-analysis of 15 studies (3947 patients with differentiated thyroid cancer) evaluating TSH-stimulated (thyroid hormone withdrawal) Tg prior to radioiodine ablation, the negative predictive value of a serum Tg below a threshold of 10 ng/mL was 94.2 percent for the absence of disease recurrence [52].

In a study that assessed Tg levels months after remnant ablation and with up to five years of follow-up, recurrent tumor was identified in 1.6 percent of patients with initial Tg after rhTSH <0.5 ng/mL, 5.5 percent of patients with Tg 0.6 to 2.0 ng/mL, and 80 percent of patients with Tg greater than 2.0 ng/mL [53].

It has become increasingly apparent that the results of repeated measurements of serum Tg over time are often more useful than a single measurement [54,55]. A progressive increase in serum Tg concentrations strongly suggests progressive disease and should lead to a search for the disease, usually starting with a chest radiograph, neck ultrasonography, and radioiodine scanning. Serum Tg doubling time of less than one year was associated with significantly higher rates of 10-year, disease-specific mortality, locoregional metastases, and identification of distant metastases than Tg doubling times of one to three years or greater than three years [56]. (See "Differentiated thyroid cancer: Overview of management", section on 'Monitoring response to therapy'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Thyroid nodules and cancer".)

SUMMARY AND RECOMMENDATIONS

●Utility of thyroglobulin – Thyroglobulin (Tg) is a storage form of thyroxine (T4) and triiodothyronine (T3). It is synthesized only by thyroid follicular cells and released into serum along with the thyroid hormones. Given the cellular specificity of Tg, its detection in biopsy specimens provides proof of the thyroid origin of the tissue. Serum Tg provides important information about the presence or absence of residual, recurrent, or metastatic disease in patients with differentiated thyroid cancer. In addition, serum Tg concentrations may predict disease-free remission at various times in the patient's course. (See 'Introduction' above and 'Monitoring response to therapy' above and "Differentiated thyroid cancer: Overview of management", section on 'Monitoring response to therapy'.)

●Thyroglobulin assay – Testing of serum Tg should be done using a sensitive assay, ideally using the same assay for each sample. Limitations to serum Tg assays include interassay variability and assay interference by antithyroglobulin (anti-Tg) antibodies, which are highly prevalent in the population. (See 'Thyroglobulin assay' above and 'Thyroglobulin antibodies' above and 'Monitoring response to therapy' above.)

●Thyroglobulin antibodies – Tg antibodies should always be measured with Tg, using the same Tg antibody assay over time. In patients with anti-Tg antibodies, serum Tg concentrations alone cannot be used as a marker to detect persistent or recurrent disease after thyroidectomy and ablation of residual normal thyroid tissue. Nevertheless, disease recurrence can be heralded by a rise in Tg antibodies with or without a corresponding rise in serum Tg. (See 'Surrogate tumor marker' above and "Differentiated thyroid cancer: Overview of management", section on 'Thyroglobulin antibodies'.)