Warning: fopen(/home/virtual/enm-kes/journal/upload/ip_log/ip_log_2025-06.txt): failed to open stream: Permission denied in /home/virtual/lib/view_data.php on line 100 Warning: fwrite() expects parameter 1 to be resource, boolean given in /home/virtual/lib/view_data.php on line 101 Comparison of Ultrasensitive and Highly Sensitive Assay to Predict Stimulated Thyroglobulin Levels Using Unstimulated Levels in Differentiated Thyroid Cancer Patients
Skip Navigation
Skip to contents

Endocrinol Metab : Endocrinology and Metabolism

clarivate
OPEN ACCESS
SEARCH
Search
Advanced Search
mobile menu button

Articles

Page Path
HOME > Endocrinol Metab > Ahead-of print > Article
Original Article
Comparison of Ultrasensitive and Highly Sensitive Assay to Predict Stimulated Thyroglobulin Levels Using Unstimulated Levels in Differentiated Thyroid Cancer Patients
Jinsun Jang1*orcid, Hyun Joo Kim2,3*orcid, Seunggyun Ha2,4, Kyong Yeun Jung1,5, Gyeongseo Jung6, Sun Wook Cho1,7, Do Joon Park1, Gi Jeong Cheon2orcid, Young Joo Park1,8,9orcid

DOI: https://doi.org/10.3803/EnM.2025.2302
Published online: June 5, 2025

1Department of Internal Medicine, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea

2Department of Nuclear Medicine, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea

3Department of Nuclear Medicine, Korea University Anam Hospital, Seoul, Korea

4Division of Nuclear Medicine, Department of Radiology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea

5Department of Internal Medicine, Nowon Eulji Medical Center, Eulji University, Seoul, Korea

6Center for Medical Innovation, Seoul National University Hospital, Seoul, Korea

7Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Korea

8Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology, Seoul National University College of Medicine, Seoul, Korea

9Genomic Medicine Institute Medical Research Center, Seoul National University College of Medicine, Seoul, Korea

Corresponding authors: Young Joo Park Department of Internal Medicine, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 03080, Korea Tel: +82-2-2072-4183, Fax: +82-2-764-2199, E-mail: yjparkmd@snu.ac.kr
Gi Jeong Cheon Department of Nuclear Medicine, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 03080, Korea Tel: +82-2-2072-3376, Fax: +82-2-745-7690, E-mail: larrycheon@gmail.com
*These authors contributed equally to this work.
• Received: January 6, 2025   • Revised: February 28, 2025   • Accepted: March 20, 2025

Copyright © 2025 Korean Endocrine Society

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 215 Views
  • 9 Download
Full Article

Download PDF Download PDF

  • Background
    Thyroglobulin (Tg) measurement is an essential aspect of monitoring for differentiated thyroid cancer (DTC) patients. This study compared the performances of ultrasensitive Tg (ultraTg) and highly sensitive Tg (hsTg) assays in predicting stimulated Tg levels without thyroid-stimulating hormone stimulation.
  • Methods
    Overall, 268 DTC patients who had undergone total thyroidectomy and either radioiodine treatment or I-123 diagnostic scanning were included. Unstimulated and stimulated Tg levels were measured using hsTg (BRAHMS Dynotest Tg-plus) and ultraTg (RIAKEY Tg immunoradiometric assay) assays. Correlations of each assay with the ability of unstimulated Tg levels to predict stimulated Tg ≥1 ng/mL were analyzed.
  • Results
    hsTg and ultraTg showed a strong correlation (R=0.79, P<0.01); the correlation was weaker in Tg antibody-positive patients (R=0.52). UltraTg demonstrated higher sensitivity in predicting stimulated Tg ≥1 ng/mL compared with hsTg. The optimal cut-off for ultraTg was 0.12 ng/mL (sensitivity, 72.0%; specificity, 67.2%). hsTg at 0.105 ng/mL had lower sensitivity (39.8%) but higher specificity (91.5%). Eight discordant cases with low hsTg (<0.2 ng/mL) but elevated ultraTg (>0.23 ng/mL) were identified; three developed structural recurrence within 3.4 to 5.8 years. Two patients had an excellent response according to hsTg but an indeterminate or biochemical incomplete response according to ultraTg.
  • Conclusion
    UltraTg demonstrated higher sensitivity in predicting positive stimulated Tg levels and potential recurrence compared with hsTg. However, its lower specificity may lead to more frequent classifications of biochemical incomplete response. UltraTg may be beneficial in clinically suspicious cases where hsTg falls below the cut-off, but its broader applicability requires further investigation.
  • Keywords: Thyroglobulin; Immunoassay; Thyroid neoplasms
Differentiated thyroid cancer (DTC) is one of the most common endocrine malignancies, and its incidence has steadily increased worldwide over the past few decades [1]. Despite its generally favorable prognosis, long-term follow-up is essential due to the potential for disease recurrence, which may occur even decades after initial treatment. Thyroglobulin (Tg) has been regarded as the gold standard for detecting residual or recurrent diseases in DTC patients.
Substantial advancements in Tg assay technology have been achieved, leading to the development of increasingly sensitive tests [2]. The evolution of Tg assays can be broadly categorized into three generations (Supplemental Tables S1-S3). First-generation assays, which served as initial tests, had limited sensitivity, with a limit of detection (LOD) of 0.2 ng/mL and a functional sensitivity of 0.9 ng/mL. Second-generation (highly sensitive) assays offered improved sensitivity and reduced interference; LODs ranged from 0.035 to 0.1 ng/mL and functional sensitivity values were between 0.15 and 0.2 ng/mL. These assays are currently the most widely used in clinical practice [3-5]. Finally, third-generation (ultrasensitive) assays represent the latest development in Tg measurement: they can detect Tg at extremely low levels, with a detection limit of 0.01 ng/mL and functional sensitivity of 0.06 ng/mL.
Traditionally, this approach has involved measuring Tg levels after thyroid-stimulating hormone (TSH) stimulation, either through thyroid hormone withdrawal or the administration of recombinant human TSH (rhTSH). Among patients who have undergone radioactive iodine (RAI) treatment after total thyroidectomy, Tg levels are used to evaluate dynamic risk stratification. An excellent response is indicated by a stimulated Tg level of less than 1 ng/mL with second-generation Tg assays and by an unstimulated Tg level of less than 0.2 ng/mL in thyroglobulin antibody (TgAb)-negative patients [6]. Although stimulated Tg testing offers high sensitivity, it has several limitations, including patient discomfort due to hypothyroid symptoms during hormone withdrawal, increased healthcare costs associated with rhTSH administration, and the need for multiple clinic visits. These factors have prompted researchers and clinicians to explore alternative approaches that might provide comparable diagnostic accuracy while reducing the burden on patients and healthcare systems [7,8].
Highly sensitive Tg (hsTg) assays have been extensively studied, but research concerning the clinical utility of ultrasensitive Tg (ultraTg) assays remains limited. In this study, we aimed to determine whether unstimulated hsTg levels could predict a stimulated-hsTg value of ≥1 ng/mL without TSH stimulation and whether the more sensitively developed unstimulated ultraTg levels offer greater predictive accuracy regarding stimulated-hsTg values above 1 ng/mL.
Subjects
This study included 268 individuals who underwent total thyroidectomy, including completion thyroidectomy after lobectomy, due to DTC. These individuals underwent remnant ablation or an I-123 diagnostic scan at Seoul National University Hospital between November 2013 and December 2018. Patients were enrolled when planning to undergo radioiodine treatment or an I-123 diagnostic scan. After patients had provided informed consent, blood samples were collected for the measurement of TSH, Tg, and TgAb levels using two different immunoradiometric assay (IRMA) kits. Unstimulated serum samples were collected after total thyroidectomy. Stimulated samples were obtained after TSH stimulation, which was performed either through levothyroxine withdrawal or intramuscular injection of rhTSH (Thyrogen, Genzyme Corp., Cambridge, MA, USA). Stimulation with rhTSH was conducted through a 2-day course of intramuscular injections of 0.9 mg rhTSH, administered 2 days prior to RAI intake. Patients who underwent more than two rounds of TSH stimulation provided either one or two sets of Tg measurements. This study was approved by the Institutional Review Board of Seoul National University Hospital (IRB No. 1608-105-786). Written informed consent was obtained from all patients.
Serum assay
Serum samples were stored at –20℃ until evaluation. For Tg measurement, two IRMA kits were used, representing two generations of Tg assays. The second-generation Tg IRMA, also referred to as hsTg, utilized the Dynotest Tg-plus kit (BRAHMS Diagnostic GmbH, Berlin, Germany). For third-generation Tg IRMA, regarded as ultraTg IRMA, the RIAKEY Tg IRMA kit (Shinjin Medics, Goyang, Korea) was used. The Dynotest Tg-plus (Tg-plus) hsTg assay measures serum Tg levels with a functional sensitivity of 0.2 ng/mL and an analytical sensitivity of 0.1 ng/mL. In contrast, ultraTg was measured using the RIAKEY Tg IRMA (Tg-RIAKEY) kit, which provides a functional sensitivity of 0.06 ng/mL and an analytical sensitivity of 0.01 ng/mL. TgAb levels were determined using the Dynotest Anti-Tg kit (BRAHMS GmbH) to adjust for interference in serum Tg measurement due to serum autoantibodies. A TgAb titer below 60 U/mL, the manufacturer’s cut-off level, was considered TgAb-negative [9,10]. TSH levels were measured using the RIAKEY TSH IRMA kit, which has a standard range of 0 to 150 μIU/mL (Shinjin Medics).
Statistical analysis
Statistical analyses were performed using R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria). In analyses involving continuous variables, the analytical sensitivity value was used as the substitute for Tg values below the analytical sensitivity threshold within each Tg assay. Reliability between the hsTg and ultraTg assays was assessed by calculating the Pearson correlation coefficient using the merged dataset, irrespective of TSH stimulation status. All other analyses excluded data from individuals with TgAb levels ≥60 U/mL to address TgAb interference. Two clinically validated stimulated Tg values of 1 ng/mL, measured using hsTg, were regarded as reference values for Tg detection. The diagnostic performance of unstimulated Tg levels in predicting positive stimulated Tg values was evaluated using the functional and analytical sensitivities of the Tg IRMA assay as cut-off thresholds. Functional sensitivity, defined as the lowest Tg concentration measurable with acceptable precision (typically a coefficient of variation <20%), represents the assay’s clinical reliability in real-world settings. In contrast, analytical sensitivity denotes the minimum detectable Tg concentration under optimal conditions, reflecting the assay’s technical detection limits. These sensitivity thresholds were applied to assess their utility in predicting positive stimulated Tg values. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. The optimal cut-off value for unstimulated Tg levels was determined using receiver operating characteristic (ROC) curve analysis. Values of P<0.05 were considered statistically significant.
Patient characteristics
In this study, 287 individuals were enrolled. Eighteen patients did not undergo ultraTg testing, and one individual declined RAI treatment. Among the 268 individuals analyzed, 56 were identified as TgAb-positive, defined by a TgAb level exceeding 60 U/L. The mean age was 45.7 years, and 205 (76.5%) patients were women in the total group (i.e., all analyzed patients). Central lymph node dissection was performed in 211 (78.7%) individuals, whereas lateral lymph node dissection was performed in six (2.2%) individuals. Both central and lateral lymph node dissections were performed in 38 (14.2%) individuals. The mean tumor sizes were 1.5±2.3 cm in the total group, 1.4±1.0 cm in the TgAb-negative group, and 2.1±4.6 cm in the TgAbpositive group. Multifocality was observed in 48.5% of the total group, 49.5% of the TgAb-negative group, and 44.6% of the TgAb-positive group. Minimal extrathyroidal extension (miETE) was present in 162 (60.4%) patients, while gross extrathyroidal extension was observed in 27 (10.1%). Central lymph node metastasis was identified in 163 (60.8%) patients, lateral lymph node metastasis was present in 28 (10.4%) patients, and lymph node metastasis was evident in 171 (63.8%) patients. Lymphatic invasion was observed in 119 (44.4%) patients in the total group, including 94 (44.4%) in the TgAb-negative group. Angioinvasion was identified in five (1.9%) patients in the total group, all of whom were TgAb-negative. Positive resection margins were reported in 11 (4.1%) patients, including eight (3.8%) in the TgAb-negative group. One individual with initial distant metastasis was TgAb-positive. RAI treatment was administered an average of 1.7±0.7 times in both the total group and the TgAb-negative group. The mean cumulative doses were 66.5±42.0 mCi in the total group and 68.0±45.7 mCi in the TgAb-negative group. Additional patient characteristics are presented in Table 1.
Correlation between serum Tg assays
Before comparing the effectiveness of Tg assays, a correlation analysis between two serum Tg assays was performed. The correlation between the two assays was analyzed using all available samples with paired hsTg and ultraTg values, regardless of TSH stimulation status or TgAb positivity. A strong correlation was observed between hsTg and ultraTg, with a correlation coefficient of R=0.79 (P<0.01), as shown in Supplemental Fig. S1A. This strong correlation, reflected by favorable R values, was consistent in the TgAb-negative group (Fig. 1A). In the TgAb-positive group, a significant correlation was also detected (P<0.01), but the correlation coefficient decreased (R=0.52) (Fig. 1B). To address potential interference by TgAb, further analyses were conducted within the TgAb-negative group. Strong correlations were observed irrespective of the stimulation methods used, with correlation coefficients of R=0.76 for rhTSH stimulation and R=0.87 for levothyroxine withdrawal (Supplemental Fig. S1B, C).
Within the subgroup where hsTg exceeded 1 ng/mL, the threshold for defining a biochemical incomplete response, the strong correlation remained consistent (Fig. 1C). The analysis was not feasible in the TgAb-positive subgroup with hsTg >1 ng/mL, as it included only three patients (Supplemental Fig. S1D). However, when hsTg was below 1 ng/mL (Fig. 1D), the correlation coefficient decreased to R=0.32, indicating a weaker correlation. This decline was more pronounced in the subgroup with hsTg <1 ng/mL and TgAb-positive individuals, where the correlation coefficient further decreased to R=0.12 (P<0.01) (Supplemental Fig. S1E). These findings suggest the potential for false-negative results, where responses considered excellent based on low hsTg values might be reclassified as biochemical incomplete or indeterminate according to ultraTg. Notably, for eight patients displaying low hsTg levels within the excellent response range (LOD <0.1), ultraTg levels were substantially elevated, ranging from 0.23 to 7.31 (Table 2, patients 1–8).
Correlation between unstimulated Tg and stimulated Tg
To compare hsTg and ultraTg in predicting stimulated Tg levels, correlation analyses were performed between unstimulated Tg and stimulated Tg using both assays. Significant correlations were observed for both hsTg and ultraTg, with correlation coefficients of R=0.27 for hsTg and R=0.44 for ultraTg (Fig. 2A). When examining TSH stimulation methods, hsTg showed similar correlation coefficients of R=0.36 and R=0.35 for levothyroxine withdrawal and rhTSH stimulation, respectively. In contrast, ultraTg displayed variable correlation coefficients depending on the TSH stimulation method: R=0.60 for levothyroxine withdrawal and R=0.25 for rhTSH stimulation (Fig. 2B).
To evaluate individual responses, unstimulated Tg and stimulated Tg results were analyzed (Table 3). A greater proportion of individuals was categorized as indeterminate with ultraTg relative to hsTg (85 individuals, 28.9%, vs. 32 individuals, 10.9%) and as biochemical incomplete response (17 individuals, 5.8%, with ultraTg vs. 12 individuals, 4.1%, with hsTg). Among those considered to exhibit an excellent response in both unstimulated and stimulated Tg measurements, 189 individuals (75.6%) were identified with hsTg, while 154 (80.2%) were identified with ultraTg. Notably, 60 individuals (24.0%) considered to exhibit an excellent response with hsTg in unstimulated Tg shifted to an indeterminate response upon stimulation. Similarly, 37 individuals (19.3%) considered to exhibit an excellent response with ultraTg in unstimulated Tg shifted to an indeterminate response upon stimulation. Only one individual, initially classified as excellent response in both hsTg and ultraTg assays of unstimulated Tg, shifted to a biochemical incomplete response upon stimulation.
Among individuals considered to exhibit an indeterminate response with unstimulated Tg, further analysis after stimulation revealed the following outcomes. With hsTg, eight individuals (25.0%) achieved an excellent response, 23 (71.9%) maintained an indeterminate response, and one (3.1%) was considered to exhibit a biochemical incomplete response. With ultraTg, 43 individuals (50.6%) achieved an excellent response, 40 (47.1%) maintained an indeterminate response, and two (2.4%) were considered to exhibit a biochemical incomplete response. In the biochemical incomplete response category, the results from four individuals were reclassified as excellent response with both hsTg and ultraTg after stimulation. However, the proportion of individuals considered to exhibit a biochemical incomplete response after stimulation was higher with ultraTg (11 individuals, 64.7%) than with hsTg (three individuals, 25.0%).
Optimal cut-off for unstimulated Tg
To determine the optimal cut-off for estimating clinically significant stimulated Tg levels using unstimulated Tg levels, ROC analysis was performed with a stimulated Tg cut-off of 1 ng/mL, utilizing both hsTg and ultraTg. To predict a stimulated Tg level of 1 ng/mL, the area under the ROC curve (AUC-ROC) of hsTg was 0.66 (95% confidence interval [CI], 0.60 to 0.71; P<0.01) (Fig. 2C), with an optimal cut-off of 0.105 ng/mL; the sensitivity was 39.8% and specificity was 91.5%. For ultraTg, the AUC-ROC was 0.74 (95% CI, 0.68 to 0.80; P<0.01) (Fig. 2C), with an optimal cut-off of 0.12 ng/mL; the sensitivity was 72.0% and specificity was 67.2%.
Considering the functional sensitivity of hsTg at 0.2 ng/mL, the diagnostic performance was also evaluated using this cut-off for hsTg. Values above this cut-off demonstrated high specificity (94.0%) but low sensitivity (34.4%), resulting in an NPV of 76.7%. Additionally, the optimal cut-off for ultraTg was 0.12 ng/mL, which provided higher sensitivity but relatively lower specificity and an NPV of 50.4%. This cut-off resulted in a PPV of 83.8%, compared with the PPV of hsTg at its functional sensitivity (72.7%) (Fig. 2A).
When individual results of unstimulated Tg and stimulated Tg were analyzed (Table 3), a tendency for ultraTg to exhibit higher sensitivity but lower specificity was observed. Based on the optimal cut-off determined by ROC analysis, 240 individuals (81.6%) were initially considered to exhibit an excellent response using hsTg; 184 (76.7%) individuals maintained an excellent response after stimulation. In contrast, the ultraTg assay initially indicated that only 153 individuals (52.0%) exhibited an excellent response, but a slightly greater proportion of 127 individuals (83.0%) maintained an excellent response after stimulation compared with hsTg.
In the indeterminate response category, hsTg initially indicated that 42 individuals (14.3%) exhibited an indeterminate response, whereas ultraTg identified a substantially greater proportion of 124 individuals (42.2%). Upon stimulation, 13 individuals (30.9%) in the hsTg group shifted to an excellent response, 28 (66.7%) maintained an indeterminate response, and one (2.4%) was considered to exhibit a biochemical incomplete response. In contrast, 70 individuals (56.4%) in the ultraTg group shifted to an excellent response, 52 (41.9%) maintained an indeterminate response, and two (1.6%) were considered to exhibit a biochemical incomplete response.
In the biochemical incomplete response category, ultraTg initially indicated that a greater proportion of individuals (5.8%) exhibited a biochemical incomplete response compared with hsTg (4.1%). After stimulation, four individuals (33.3%) in the hsTg group shifted to an excellent response, five (41.7%) maintained an indeterminate response, and three (25.0%) maintained a biochemical incomplete response. In the ultraTg group, four individuals (23.5%) shifted to an excellent response, 11 (64.7%) shifted to an indeterminate response, and two (11.8%) maintained a biochemical incomplete response.
Utility of unstimulated ultraTg in patients with unstimulated hsTg
To compare the utility of unstimulated Tg, paired unstimulated Tg and stimulated Tg levels were analyzed in individuals considered to exhibit an excellent response (Fig. 3). Among 294 individuals considered to exhibit an excellent response based on the optimal cut-off for unstimulated hsTg (<0.105 ng/mL), 56 individuals (23.3%) did not meet the criteria for an excellent response based on stimulated Tg. Of these, 143 individuals (59.6%) were considered to exhibit an excellent response using the optimal cut-off for unstimulated ultraTg. However, 23 of these individuals (16.1%) subsequently did not meet the criteria for an excellent response. Furthermore, among the 56 individuals (23.3%) falsely considered to exhibit an excellent response according to unstimulated hsTg, 33 did not meet the criteria for an excellent response based on unstimulated ultraTg. Conversely, 64 individuals who were correctly considered to exhibit an excellent response according to both unstimulated hsTg and stimulated Tg were incorrectly indicated to not meet the criteria for an excellent response based on unstimulated ultraTg, highlighting ongoing issues with the lower specificity of ultraTg.
When analyzed using the standard unstimulated Tg cut-off value of 1 ng/mL, 61 of 250 patients considered to exhibit an excellent response with unstimulated hsTg were reclassified as not meeting the criteria for excellent response according to stimulated Tg (Fig. 4). Among them, 27 individuals (40.3%) were classified as not in excellent response based on unstimulated ultraTg. Forty individuals remained incorrectly within the group considered not to exhibit an excellent response with unstimulated ultraTg.
Discordant cases: low hsTg with elevated ultraTg
Eight discordant cases were identified, in which unstimulated hsTg levels were below 0.2 ng/mL but unstimulated ultraTg levels exceeded 0.23 ng/mL (Table 2). All patients were categorized as intermediate risk according to the American Thyroid Association (ATA) risk stratification system. Six patients, except for patient no. 2, had miETE. Patient nos. 2 and 8 had RAI remnants during the last RAI treatment. Patient nos. 5, 6, 7, and 8 experienced structural recurrence during follow-up, ranging from 3.4 to 5.8 years. Three other patients with structural recurrence exhibited minimal differences between hsTg and ultraTg levels (less than 0.1 ng/mL) or even lower ultraTg levels relative to hsTg. Patient nos. 8, 10, and 11 were considered to exhibit an excellent response during the first evaluation using both unstimulated hsTg and ultraTg. In contrast, patient no. 9 was considered to exhibit a biochemical incomplete response in both the first and second RAI treatments with both unstimulated Tg assays. However, patient no. 9 was TgAb-positive, which may have caused interference in Tg measurements. All patients with structural recurrence were categorized as intermediate risk according to the ATA risk stratification system and had miETE. They also showed positive findings concerning the B-type Raf kinase (BRAF) mutation and negative findings concerning the telomerase reverse transcriptase (TERT) mutation.
When diagnostic performance was analyzed for individuals with structural recurrence and TgAb negativity, all five patients were considered to exhibit either an indeterminate or biochemical incomplete response based on stimulated Tg levels (Table 2). Among them, two individuals were considered to exhibit an indeterminate response with unstimulated ultraTg, despite being categorized as excellent response based on unstimulated hsTg. This discrepancy was observed using both the standard cut-off and the optimal cut-off determined by ROC analysis (Table 3, Supplemental Figs. S2, S3).
This study evaluated the clinical utility of ultraTg assays relative to hsTg assays for predicting stimulated Tg levels in patients with DTC. The findings highlight the value of unstimulated Tg assays, particularly in cases where hsTg measurements fall below the excellent response threshold of 0.2 ng/mL. Considering their enhanced sensitivity, ultraTg assays enable more precise stratification of patient risk status, offering a potentially impactful tool for managing recurrence risk in DTC patients.
Although several studies have compared the efficacies of Tg assays between first- and second-generation tests, revealing that hsTg detects minimal residual disease more effectively than conventional assays, no published studies have compared the performances of second- and third-generation assays [11]. The ultraTg assay demonstrated superior sensitivity in predicting stimulated Tg levels greater than 1 ng/mL compared with the hsTg assay. At the optimal cut-off of 0.12 ng/mL, ultraTg achieved a sensitivity of 72.0% and specificity of 67.2%, significantly outperforming the sensitivity of 34.4% for hsTg at its functional sensitivity cut-off of 0.2 ng/mL.
Given that unstimulated ultraTg has lower specificity and a reduced NPV compared to hsTg, the risk of false-positive results increases. This could lead to unnecessary follow-up evaluations, which may be excessive considering the low recurrence rate of DTC. Interestingly, as shown in Table 2, patients who eventually experienced recurrence had lower ultraTg levels than those with no evidence of disease. This suggests that ultraTg alone may not be a reliable predictor of recurrence.
UltraTg indicated that a greater proportion of individuals had an excellent response before and after stimulation, compared with hsTg. This difference was observed both at the standard cut-off (80.2% vs. 75.6%) and at the optimal cut-off determined by ROC analysis (83.0% vs. 76.7%). However, ultraTg also indicated that a greater proportion of individuals shifted from an indeterminate response to an excellent response after stimulation. Additionally, ultraTg revealed that a greater proportion of individuals shifted from a biochemical incomplete response to an indeterminate response after stimulation. This higher proportion of category shifts may be because stimulated Tg was measured using the hsTg assay. Nevertheless, given the proportion of structural recurrence in patients with stimulated Tg levels >1 ng/mL, the ultraTg assay remains a reliable diagnostic tool.
The difference in correlation coefficients depending on the TSH stimulation method in the ultraTg assay (R=0.60 for levothyroxine withdrawal vs. R=0.25 for rhTSH stimulation) was observed. Traditionally, stimulated Tg levels have differed between TSH stimulation methods, with levothyroxine withdrawal resulting in values approximately two to five times higher than those observed with rhTSH stimulation [12,13].
Eight discordant cases were identified, in which hsTg levels were below 0.2 ng/mL but ultraTg levels exceeded 0.23 ng/mL. Notably, four of these patients developed structural recurrence within 3.4 to 5.8 years. This finding underscores the potential of ultraTg to detect early signs of recurrence that might be missed by hsTg assays, particularly in patients considered to exhibit an excellent response based on hsTg levels. Based on these analysis, ultraTg could be suggested as a supplementary screening tool for detecting high-risk patients who might not have been identified using unstimulated hsTg alone.
Interference of TgAb is a major limitation of serum Tg measurement in both qualitative and quantitative, which may cause falsely low values of serum Tg. Autologous TgAb is presented in 25% of DTC patients and 10% of the general population [14]. Also, TgAb concentration may rise transiently after thyroidectomy or ablation therapy as an immune reaction [15]. The measurement of TgAb is highly dependent on TgAb assays which may show a 100-fold variation of concentrations and marked discordance. Interference between Tg and TgAb is not easily estimated by a simple relationship, because even high titters of TgAbs possibly show no distinct evidence of interference with the Tg measurement [14]. Also, there are 20% to 30% of DTC patients with a borderline level of TgAb between upper and lower limit of reference range, which is not ignorable in clinical situations [16]. Accordingly, although the ultraTg assay has superior sensitivity and specificity compared to the hsTg assay, there is still a lack of evidence supporting its superiority in TgAb-positive patients.
The present study had some limitations. First, the analyzed data, sampled before ablation, differs from the clinical context of using hsTg as a tumor marker during long-term follow-up after ablation. Additionally, some patients received further RAI treatments without ultraTg measurement, which could have provided a more comprehensive analysis. Second, although ultraTg offers higher sensitivity, it demonstrated lower specificity compared with hsTg. Further analysis among patients with structural recurrence did not reveal distinct or specific features. This lower specificity could lead to an increased number of biochemical incomplete response classifications, potentially resulting in unnecessary additional testing or treatment [17]. Finally, the variability in TgAb presence across the sample may impact the generalizability of these findings [16,18].
In cases where unstimulated hsTg indicates an excellent response, measurements of unstimulated ultraTg could be clinically useful if further evaluation is warranted, such as for low-risk patients exhibiting structurally indeterminate findings during follow-up. Notably, an unstimulated ultraTg level exceeding 0.12 ng/mL may indicate an increased risk of recurrence; caution is thus warranted. For example, among patients with structural recurrence and TgAb negativity, two individuals who were classified as excellent response according to unstimulated hsTg were reclassified as exhibiting an indeterminate response based on unstimulated ultraTg. These findings suggest that, although limited, unstimulated ultraTg may have substantial potential utility in dynamic risk stratification during follow-up, particularly for assessing Tg response and identifying patients who require closer monitoring. Future studies should examine large multi-institutional cohorts to validate these cut-off thresholds and assess how the integration of ultraTg can improve prognostic accuracy.
In conclusion, although ultraTg has high sensitivity in predicting positive stimulated Tg levels and recurrence, its low specificity results in a high frequency of cases exhibiting clinically insignificant biochemical incomplete responses. The clinical utility of ultraTg appears limited but may be valuable in clinically suspicious cases where hsTg is below the cut-off.

Supplemental Table S1.

Limit of Detection and Functional Sensitivity of Tg Assay Kits
enm-2025-2302-Supplemental-Table-S1.pdf

Supplemental Table S2.

Characteristics of Individuals with Low hsTg and High ultraTg Levels
enm-2025-2302-Supplemental-Table-S2.pdf

Supplemental Table S3.

Characteristics of Individuals with Structural Recurrence
enm-2025-2302-Supplemental-Table-S3.pdf

Supplemental Fig. S1.

Correlation between high sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg) stratified by stimulation method and thyroglobulin antibody (TgAb) status. Samples with (A) all sample, (B) recombinant human thyroid-stimulating hormone (rhTSH), TgAb < 60 U/L, (C) off-levothyroxine (T4), TgAb < 60 U/L, (D) hsTg ≥1 ng/mL, TgAb ≥60 U/L, and (E) hsTg < 1 ng/mL, Ab ≥60 U/L. Corr r, correlation coefficient r.
enm-2025-2302-Supplemental-Fig-S1.pdf

Supplemental Fig. S2.

Performance comparison of unstimulated-high sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg) in relation to stimulated thyroglobulin (Tg) using optimal cut-off values with individuals of structural recurred. Two individuals with structural recurrence were excluded from the figure results due to thyroglobulin antibody (TgAb) levels ≥60 U/L.
enm-2025-2302-Supplemental-Fig-S2.pdf

Supplemental Fig. S3.

Performance comparison of unstimulated-high sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg) in relation to stimulated thyroglobulin (Tg) using standard cut-off values with individuals of structural recurred. Two individuals with structural recurrence were excluded from the figure results due to thyroglobulin antibody (TgAb) levels ≥60 U/L.
enm-2025-2302-Supplemental-Fig-S3.pdf

CONFLICTS OF INTEREST

Young Joo Park is the editor-in-chief of the journal and Sun Wook Cho is the deputy editor of the journal. But they were not involved in the selection of peer reviewers, evaluation, or decision-making process for this article. No other potential conflicts of interest relevant to this article were reported.

ACKNOWLEDGMENTS

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (RS-2024-00453506) and the BK21FOUR Program of the National Research Foundation of Korea (NRF) funded by the Ministry of Education (5120200513755). The biospecimens and data used in this study were provided by the Biobank of Seoul National University Hospital, a member of the Korea Biobank Network.

We gratefully acknowledge the generous support of Ok Nam Cho, whose contribution was instrumental to the completion of this research.

AUTHOR CONTRIBUTIONS

Conception or design: G.J.C., Y.J.P. Acquisition, analysis, or interpretation of data: J.J., H.J.K., S.H., K.Y.J., G.J., S.W.C., D.J.P., G.J.C., Y.J.P. Drafting the work or revising: J.J., H.J.K., Y.J.P. Final approval of the manuscript: J.J., H.J.K., S.H., K.Y.J., G.J., S.W.C., D.K.P., G.J.C., Y.J.P.

Fig. 1.
Correlation between highly sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg). Correlation shown for samples with (A) TgAb <60 U/L; (B) TgAb ≥60 U/L; (C) hsTg ≥1 ng/mL, TgAb <60 U/L; and (D) hsTg <1 ng/mL, TgAb <60 U/L. Corr r, correlation coefficient r.
enm-2025-2302f1.jpg
Fig. 2.
Correlation between unstimulated-thyroglobulin (Tg) and stimulated Tg, and determination of optimal cut-off. Correlation shown for samples with (A) unstimulated highly sensitive Tg (hsTg) and stimulated Tg and (B) unstimulated ultrasensitive Tg (ultraTg) and stimulated Tg. (C) Optimal cut-off for predicting stimulated Tg ≥1 ng/mL using area under the receiver operating characteristic curve (AUC-ROC) analysis. TSH, thyroid-stimulating hormone; Corr r, correlation coefficient r; FS, functional sensitivity; PPV, positive predictive value; NPV, negative predictive value.
enm-2025-2302f2.jpg
Fig. 3.
Comparison of unstimulated highly sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg) performances in relation to stimulated Tg using optimal cut-off values.
enm-2025-2302f3.jpg
Fig. 4.
Comparison of unstimulated highly sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg) performances in relation to stimulated Tg using standard cut-off values
enm-2025-2302f4.jpg
Table 1.
Baseline Characteristics
Thyroglobulin antibody Total <60 U/L ≥60 U/L
Number 268 212 56
Age, yr 45.7±12.4 45.6±12.2 46.0±13.2
Female sex 205 (76.5) 156 (73.6) 49 (87.5)
LN dissection
 None 13 (4.9) 12 (5.7) 1 (1.8)
 Central 249 (92.9)a 194 (91.6)a 46 (82.2)a
 Lateral 44 (16.4)a 35 (16.5)a 9 (16.1)a
Tumor size, cm 1.5±2.3 1.4±1.0 2.1±4.6
Multifocal 130 (48.5) 105 (49.5) 25 (44.6)
ETE
 Microscopic 162 (60.4) 126 (59.4) 36 (64.3)
 Gross 27 (10.1) 20 (9.4) 7 (12.5)
Lymphatic invasion 119 (44.4) 94 (44.4) 25 (44.6)
Angioinvasion 5 (1.9) 5 (2.4) 0
LN metastasis, total 171 (63.8) 136 (64.1) 34 (60.7)
 Central 163 (60.8) 128 (60.4) 34 (60.7)
 Lateral 28 (10.4) 22 (10.3) 5 (8.9)
Resection margin positive 11 (4.1) 8 (3.8) 3 (5.4)
Radioiodine treatment
 Number 1.7±0.7 1.7±0.7 1.6±0.6
 Total cumulative dose, mCi 66.5±42.0 68.0±45.7 60.5±21.9
Distant metastasis 1 (0.4) 0 1 (1.8)
Measurement of stimulated Tgb
 1st total/rhTSH/off-T4 263/113/150 (98.1/42.2/56.0) 207/90/117 (97.6/42.4/55.2) 56/23/33 (100/41.1/58.9)
 2nd total/rhTSH/off-T4 97/48/49 (36.2/49.5/18.3) 77/40/37 (28.7/51.9/48.1) 20/9/11 (35.7/45.0/55.0)
Dynamic risk stratification
 Duration of follow-up, yr 8.5±10.0 8.3±8.3 9.4±14.8
 Excellent 233 (86.9) 182 (85.8) 51 (91.1)
 Indeterminate 16 (6.0) 14 (6.6) 2 (3.6)
 Biochemical incompletec 10 (3.7) 9 (4.2) 1 (1.8)
 Structural incomplete 9 (3.4) 7 (3.3) 2 (3.6)

Values are expressed as mean±standard deviation or number (%). The group with thyroglobulin antibody (TgAb) levels greater than 60 U/L includes individuals with at least one TgAb measurement exceeding 60 U/L, either during unstimulated or stimulated Tg measurements, in the first or second test.

LN, lymph node; ETE, extrathyroidal extension; Tg, thyroglobulin; rhTSH, recombinant human thyroid-stimulating hormone; T4, levothyroxine.

a A total of 38 individuals (14.2%) underwent both central and lateral lymph node dissection, including 29 individuals (13.7%) in the group with TgAb levels less than 60 U/L and nine individuals (16.1%) in the group with TgAb levels greater than 60 U/L;

b Two methods were used for Tg stimulation measurement. rhTSH stimulation involved a 2-day course of intramuscular injections of 0.9 mg rhTSH, administered 2 days before radioactive iodine intake. Off-T4 stimulation was achieved via levothyroxine withdrawal; cBiochemical incomplete response was defined as suppressed Tg >1 ng/mL, thyroidstimulating hormone-stimulated Tg >10 ng/mL, or rising anti-Tg antibody levels in the absence of structural disease [6].

Table 2.
Characteristics of Patients Showing Discordant ultraTg Levels among Those with Unstimulated hsTg <0.2 ng/mL or Structural Recurrence during Follow-up
No. Age, yr Sex 1st Unstimulated
 1st Stimulated
 2nd Unstimulated
 2nd Stimulated
 TgAb (≥60 U/L) TSHa Size, mm ATA risk ETE No. of LN Meta (meta/examined) Total no. of RAI Total RAI dose, mCi Last RAI remnant Recurrence Recurrence site
hsTg ultraTg hsTg ultraTg hsTg ultraTg hsTg ultraTg
Patients with discordant ultraTg levels among those with unstimulated hsTg <0.2 ng/mL
 1 47 F 0.1 3.46 1.26 1.92 0.1 0.43 0.1 2.57 – 0.14 0.7 Intermediate +, mi ETE 0/14 1 50 – No
 2 31 F 0.1 2.68 0.83 3.07 0.1 - 0.1 - + 0.05 0.7 Intermediate - 3/8 2 100 + No
 3 43 M 0.1 1.94 4.12 7.82 0.1 - 1.23 - – 0.05 0.6 Intermediate +, mi ETE 1/7 1 50 – No
 4 34 M 0.1 5.37 3.38 12.74 0.1 7.31 4.01 17.66 + 0.77 1.6 Intermediate +, mi ETE 9/12 3 110 – No
 5b 38 M 1.29 0.98 14.08 22.49 0.1 0.43 0.1 6.65 – 0.05 1.1 Intermediate +, mi ETE 0/0 2 80 – Yes (4.0 yr) LN
 6b 25 F 0.1 0.23 1.45 0.95 0.1 NA 1.03 NA – 0.06 0.6 Intermediate +, mi ETE 5/34 2 80 – Yes (3.5 yr) LN
 7b 31 F 0.2 0.50 29.94 34.57 0.1 NA 4.67 NA – 0.05 0.5 Intermediate +, mi ETE 0/1 2 60 – Yes (5.8 yr) LN
 8b 62 F 0.1 1.15 5.01 NA 0.1 NA 8.22 NA – 0.05 1.2 Intermediate +, mi ETE 2/7 3 130 + Yes (3.4 yr) LN
Other patients with structural recurrence
 9b 79 F 7.31 2.36 66.57 27.80 1.66 1.13 NA NA + 0.05 1.5 Intermediate +, mi ETE 5/9 1 50 + Yes (4.7 yr) LN
 10b 26 F 0.1 0.11 0.10 0.64 0.1 NA 0.1 NA + 0.05 0.7 Intermediate +, mi ETE 9/21 2 80 – Yes (2.5 yr) LN
 11b 28 F 0.1 0.01 2.14 2.08 0.1 NA 0.19 NA – 0.24 0.8 Intermediate +, mi ETE 2/3 2 60 – Yes (10.5 yr) LN

All individuals were B-type Raf kinase (BRAF; +) and telomerase reverse transcriptase (TERT; –).

ultraTg, ultrasensitive thyroglobulin; hsTg, highly sensitive thyroglobulin; TgAb, thyroglobulin antibody; TSH, thyroid-stimulating hormone; ATA, American Thyroid Association; ETE, extrathyroidal extension; LN, lymph node; RAI, radioactive iodine; miETE, minimal extrathyroidal extension, NA, missing value.

a TSH levels were measured at the time of the first unstimulated Tg measurement;

b Patients 5, 6, 7, 8, 9, 10, and 11 experienced structural recurrence.

Table 3.
Comparison of Diagnostic Performances in Predicting Stimulated Tg
By unstimulated Tg Total By stimulated Tg
Excellent Indeterminate Biochemical incomplete
Unstimulated Tg <0.2 ng/mL cut-off
 hsTg (<0.2 ng/mL) Excellent 250 [3]/294 189/250 60 [3]/250 1/250
(85.0) (75.6) (24.0) (0.4)
Indeterminate 32 [1]/294 8/32 23/32 1 [1]/32
(10.9) (25.0) (71.9) (3.1)
Biochemical incomplete 12 [1]/294 4/12 5/12 3 [1]/12
(4.1) (33.3) (41.7) (25.0)
 ultraTg (<0.2 ng/mL) Excellent 192 [1]/294 154/192 37 [1]/192 1/192
(65.3) (80.2) (19.3) (5.2)
Indeterminate 85 [3]/294 43/85 40 [1]/85 2 [2]/85
(28.9) (50.6) (47.1) (2.3)
Biochemical incomplete 17 [1] 4/17 11 [1]/17 2/17
(5.8) (23.5) (64.7) (11.8)
Best cut-off according to ROC analysis
 hsTg (<0.105 ng/mL) Excellent 240 [3]/294 184/240 55 [3]/240 1/240
(81.6) (76.7) (22.9) (0.4)
Indeterminate 42 [1]/294 13/42 28/42 1 [1]/42
(14.3) (30.9) (66.7) (2.4)
Biochemical incomplete 12 [1]/294 4/12 5/12 3 [1]/12
(4.1) (33.3) (41.7) (25.0)
 ultraTg (<0.12 ng/mL) Excellent 153 [1]/294 127/153 25 [1]/153 1/153
(52.0) (83.0) (16.4) (0.6)
Indeterminate 124 [3]/294 70/124 52 [1]/124 2 [2]/124
(42.2) (56.4) (41.9) (1.6)
Biochemical incomplete 17 [1] 4/17 11 [1]/17 2/17
(5.8) (23.5) (64.7) (11.8)

Values are expressed as number (%). The number of patients with confirmed structural recurrence is recorded in [ ], based on the first measurement. Two individuals with structural recurrence were excluded from the table results due to thyroglobulin antibody levels ≥60 U/L. For unstimulated Tg, an excellent response was defined as a level below the cut-off; levels between the cut-off and 1.0 ng/mL were considered indeterminate response; and levels exceeding 1.0 ng/mL indicated a biochemical incomplete response. For stimulated Tg, an excellent response was defined as a level below 1.0 ng/mL; levels between 1.0 and 10.0 ng/mL were considered indeterminate response; and levels exceeding 10.0 ng/mL indicated a biochemical incomplete response.

Tg, thyroglobulin; hsTg, highly sensitive Tg; ultraTg, ultrasensitive Tg.

  • 1. Davies L, Welch HG. Increasing incidence of thyroid cancer in the United States, 1973-2002. JAMA 2006;295:2164–7.ArticlePubMed
  • 2. Giovanella L. Highly sensitive thyroglobulin measurements in differentiated thyroid carcinoma management. Clin Chem Lab Med 2008;46:1067–73.ArticlePubMed
  • 3. Giovanella L, D’Aurizio F, Algeciras-Schimnich A, Gorges R, Petranovic Ovcaricek P, Tuttle RM, et al. Thyroglobulin and thyroglobulin antibody: an updated clinical and laboratory expert consensus. Eur J Endocrinol 2023;189:R11–27.ArticlePDF
  • 4. Schlumberger M, Hitzel A, Toubert ME, Corone C, Troalen F, Schlageter MH, et al. 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.ArticlePubMedPDF
  • 5. Spencer C, Fatemi S, Singer P, Nicoloff J, Lopresti J. Serum basal thyroglobulin measured by a second-generation assay correlates with the recombinant human thyrotropin-stimulated thyroglobulin response in patients treated for differentiated thyroid cancer. Thyroid 2010;20:587–95.ArticlePubMed
  • 6. Filetti S, Durante C, Hartl D, Leboulleux S, Locati LD, Newbold K, et al. Thyroid cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 2019;30:1856–83.ArticlePubMedPDF
  • 7. Giovanella L, Treglia G, Sadeghi R, Trimboli P, Ceriani L, Verburg FA. Unstimulated highly sensitive thyroglobulin in follow-up of differentiated thyroid cancer patients: a meta-analysis. J Clin Endocrinol Metab 2014;99:440–7.ArticlePubMed
  • 8. Giovanella L, Clark PM, Chiovato L, Duntas L, Elisei R, Feldt-Rasmussen U, et al. Thyroglobulin measurement using highly sensitive assays in patients with differentiated thyroid cancer: a clinical position paper. Eur J Endocrinol 2014;171:R33–46.PubMedPMC
  • 9. Massart C, Maugendre D. Importance of the detection method for thyroglobulin antibodies for the validity of thyroglobulin measurements in sera from patients with Graves disease. Clin Chem 2002;48:102–7.ArticlePubMedPDF
  • 10. Morgenthaler NG, Froehlich J, Rendl J, Willnich M, Alonso C, Bergmann A, et al. Technical evaluation of a new immunoradiometric and a new immunoluminometric assay for thyroglobulin. Clin Chem 2002;48:1077–83.ArticlePubMedPDF
  • 11. Flores-Rebollar A, Perez-Diaz I, Lagunas-Barcenas S, Garcia-Martinez B, Rivera-Moscoso R, Fagundo-Sierra R. Clinical utility of an ultrasensitive thyroglobulin assay in the follow-up of patients with differentiated thyroid cancer: can the stimulation test be avoided in patients with an intermediate recurrence risk? Acta Otorhinolaryngol Ital 2018;38:188–93.ArticlePubMedPMC
  • 12. Kowalska A, Palyga I, Gasior-Perczak D, Walczyk A, Trybek T, Sluszniak A, et al. The cut-off level of recombinant human TSH-stimulated thyroglobulin in the follow-up of patients with differentiated thyroid cancer. PLoS One 2015;10:e0133852.ArticlePubMedPMC
  • 13. Pellegriti G, Scollo C, Regalbuto C, Attard M, Marozzi P, Vermiglio F, et al. The diagnostic use of the rhTSH/thyroglobulin test in differentiated thyroid cancer patients with persistent disease and low thyroglobulin levels. Clin Endocrinol (Oxf) 2003;58:556–61.ArticlePubMedPDF
  • 14. Spencer CA, Takeuchi M, Kazarosyan M, Wang CC, Guttler RB, Singer PA, et al. Serum thyroglobulin autoantibodies: prevalence, influence on serum thyroglobulin measurement, and prognostic significance in patients with differentiated thyroid carcinoma. J Clin Endocrinol Metab 1998;83:1121–7.ArticlePubMed
  • 15. Gorges R, Maniecki M, Jentzen W, Sheu SN, Mann K, Bockisch A, et al. 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.ArticlePubMed
  • 16. Spencer C, Petrovic I, Fatemi S. 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.PubMed
  • 17. McGrath RT, Preda VA, Clifton-Bligh P, Robinson B, Sywak M, Delbridge L, et al. Is there a role for an ultrasensitive thyroglobulin assay in patients with serum antithyroglobulin antibodies?: a large (Australian) cohort study in differentiated thyroid cancer. Clin Endocrinol (Oxf) 2016;84:271–7.Article
  • 18. Dekker BL, van der Horst-Schrivers AN, Sluiter WJ, Brouwers AH, Lentjes EG, Heijboer AC, et al. Clinical applicability of low levels of thyroglobulin autoantibodies as cutoff point for thyroglobulin autoantibody positivity. Thyroid 2019;29:71–8.Article

Figure & Data

References

    Citations

    Citations to this article as recorded by  

      • PubReader PubReader
      • ePub LinkePub Link
      • Cite
        Cite
        export Copy Download
        Close
        Download Citation
        Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.
        Format:
        • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
        • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
        Include:
        • Citation for the content below
        Comparison of Ultrasensitive and Highly Sensitive Assay to Predict Stimulated Thyroglobulin Levels Using Unstimulated Levels in Differentiated Thyroid Cancer Patients
        Close
      • XML DownloadXML Download
      Figure
      • 0
      • 1
      • 2
      • 3
      Comparison of Ultrasensitive and Highly Sensitive Assay to Predict Stimulated Thyroglobulin Levels Using Unstimulated Levels in Differentiated Thyroid Cancer Patients
      Image Image Image Image
      Fig. 1. Correlation between highly sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg). Correlation shown for samples with (A) TgAb <60 U/L; (B) TgAb ≥60 U/L; (C) hsTg ≥1 ng/mL, TgAb <60 U/L; and (D) hsTg <1 ng/mL, TgAb <60 U/L. Corr r, correlation coefficient r.
      Fig. 2. Correlation between unstimulated-thyroglobulin (Tg) and stimulated Tg, and determination of optimal cut-off. Correlation shown for samples with (A) unstimulated highly sensitive Tg (hsTg) and stimulated Tg and (B) unstimulated ultrasensitive Tg (ultraTg) and stimulated Tg. (C) Optimal cut-off for predicting stimulated Tg ≥1 ng/mL using area under the receiver operating characteristic curve (AUC-ROC) analysis. TSH, thyroid-stimulating hormone; Corr r, correlation coefficient r; FS, functional sensitivity; PPV, positive predictive value; NPV, negative predictive value.
      Fig. 3. Comparison of unstimulated highly sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg) performances in relation to stimulated Tg using optimal cut-off values.
      Fig. 4. Comparison of unstimulated highly sensitive thyroglobulin (hsTg) and ultrasensitive thyroglobulin (ultraTg) performances in relation to stimulated Tg using standard cut-off values

      Fig. 1.

      Fig. 2.

      Fig. 3.

      Fig. 4.

      Comparison of Ultrasensitive and Highly Sensitive Assay to Predict Stimulated Thyroglobulin Levels Using Unstimulated Levels in Differentiated Thyroid Cancer Patients
      Thyroglobulin antibody Total <60 U/L ≥60 U/L
      Number 268 212 56
      Age, yr 45.7±12.4 45.6±12.2 46.0±13.2
      Female sex 205 (76.5) 156 (73.6) 49 (87.5)
      LN dissection
       None 13 (4.9) 12 (5.7) 1 (1.8)
       Central 249 (92.9)a 194 (91.6)a 46 (82.2)a
       Lateral 44 (16.4)a 35 (16.5)a 9 (16.1)a
      Tumor size, cm 1.5±2.3 1.4±1.0 2.1±4.6
      Multifocal 130 (48.5) 105 (49.5) 25 (44.6)
      ETE
       Microscopic 162 (60.4) 126 (59.4) 36 (64.3)
       Gross 27 (10.1) 20 (9.4) 7 (12.5)
      Lymphatic invasion 119 (44.4) 94 (44.4) 25 (44.6)
      Angioinvasion 5 (1.9) 5 (2.4) 0
      LN metastasis, total 171 (63.8) 136 (64.1) 34 (60.7)
       Central 163 (60.8) 128 (60.4) 34 (60.7)
       Lateral 28 (10.4) 22 (10.3) 5 (8.9)
      Resection margin positive 11 (4.1) 8 (3.8) 3 (5.4)
      Radioiodine treatment
       Number 1.7±0.7 1.7±0.7 1.6±0.6
       Total cumulative dose, mCi 66.5±42.0 68.0±45.7 60.5±21.9
      Distant metastasis 1 (0.4) 0 1 (1.8)
      Measurement of stimulated Tgb
       1st total/rhTSH/off-T4 263/113/150 (98.1/42.2/56.0) 207/90/117 (97.6/42.4/55.2) 56/23/33 (100/41.1/58.9)
       2nd total/rhTSH/off-T4 97/48/49 (36.2/49.5/18.3) 77/40/37 (28.7/51.9/48.1) 20/9/11 (35.7/45.0/55.0)
      Dynamic risk stratification
       Duration of follow-up, yr 8.5±10.0 8.3±8.3 9.4±14.8
       Excellent 233 (86.9) 182 (85.8) 51 (91.1)
       Indeterminate 16 (6.0) 14 (6.6) 2 (3.6)
       Biochemical incompletec 10 (3.7) 9 (4.2) 1 (1.8)
       Structural incomplete 9 (3.4) 7 (3.3) 2 (3.6)
      No. Age, yr Sex 1st Unstimulated
      		 1st Stimulated
      		 2nd Unstimulated
      		 2nd Stimulated
      		 TgAb (≥60 U/L) TSHa Size, mm ATA risk ETE No. of LN Meta (meta/examined) Total no. of RAI Total RAI dose, mCi Last RAI remnant Recurrence Recurrence site
      hsTg ultraTg hsTg ultraTg hsTg ultraTg hsTg ultraTg
      Patients with discordant ultraTg levels among those with unstimulated hsTg <0.2 ng/mL
       1 47 F 0.1 3.46 1.26 1.92 0.1 0.43 0.1 2.57 – 0.14 0.7 Intermediate +, mi ETE 0/14 1 50 – No
       2 31 F 0.1 2.68 0.83 3.07 0.1 - 0.1 - + 0.05 0.7 Intermediate - 3/8 2 100 + No
       3 43 M 0.1 1.94 4.12 7.82 0.1 - 1.23 - – 0.05 0.6 Intermediate +, mi ETE 1/7 1 50 – No
       4 34 M 0.1 5.37 3.38 12.74 0.1 7.31 4.01 17.66 + 0.77 1.6 Intermediate +, mi ETE 9/12 3 110 – No
       5b 38 M 1.29 0.98 14.08 22.49 0.1 0.43 0.1 6.65 – 0.05 1.1 Intermediate +, mi ETE 0/0 2 80 – Yes (4.0 yr) LN
       6b 25 F 0.1 0.23 1.45 0.95 0.1 NA 1.03 NA – 0.06 0.6 Intermediate +, mi ETE 5/34 2 80 – Yes (3.5 yr) LN
       7b 31 F 0.2 0.50 29.94 34.57 0.1 NA 4.67 NA – 0.05 0.5 Intermediate +, mi ETE 0/1 2 60 – Yes (5.8 yr) LN
       8b 62 F 0.1 1.15 5.01 NA 0.1 NA 8.22 NA – 0.05 1.2 Intermediate +, mi ETE 2/7 3 130 + Yes (3.4 yr) LN
      Other patients with structural recurrence
       9b 79 F 7.31 2.36 66.57 27.80 1.66 1.13 NA NA + 0.05 1.5 Intermediate +, mi ETE 5/9 1 50 + Yes (4.7 yr) LN
       10b 26 F 0.1 0.11 0.10 0.64 0.1 NA 0.1 NA + 0.05 0.7 Intermediate +, mi ETE 9/21 2 80 – Yes (2.5 yr) LN
       11b 28 F 0.1 0.01 2.14 2.08 0.1 NA 0.19 NA – 0.24 0.8 Intermediate +, mi ETE 2/3 2 60 – Yes (10.5 yr) LN
      By unstimulated Tg Total By stimulated Tg
      Excellent Indeterminate Biochemical incomplete
      Unstimulated Tg <0.2 ng/mL cut-off
       hsTg (<0.2 ng/mL) Excellent 250 [3]/294 189/250 60 [3]/250 1/250
      (85.0) (75.6) (24.0) (0.4)
      Indeterminate 32 [1]/294 8/32 23/32 1 [1]/32
      (10.9) (25.0) (71.9) (3.1)
      Biochemical incomplete 12 [1]/294 4/12 5/12 3 [1]/12
      (4.1) (33.3) (41.7) (25.0)
       ultraTg (<0.2 ng/mL) Excellent 192 [1]/294 154/192 37 [1]/192 1/192
      (65.3) (80.2) (19.3) (5.2)
      Indeterminate 85 [3]/294 43/85 40 [1]/85 2 [2]/85
      (28.9) (50.6) (47.1) (2.3)
      Biochemical incomplete 17 [1] 4/17 11 [1]/17 2/17
      (5.8) (23.5) (64.7) (11.8)
      Best cut-off according to ROC analysis
       hsTg (<0.105 ng/mL) Excellent 240 [3]/294 184/240 55 [3]/240 1/240
      (81.6) (76.7) (22.9) (0.4)
      Indeterminate 42 [1]/294 13/42 28/42 1 [1]/42
      (14.3) (30.9) (66.7) (2.4)
      Biochemical incomplete 12 [1]/294 4/12 5/12 3 [1]/12
      (4.1) (33.3) (41.7) (25.0)
       ultraTg (<0.12 ng/mL) Excellent 153 [1]/294 127/153 25 [1]/153 1/153
      (52.0) (83.0) (16.4) (0.6)
      Indeterminate 124 [3]/294 70/124 52 [1]/124 2 [2]/124
      (42.2) (56.4) (41.9) (1.6)
      Biochemical incomplete 17 [1] 4/17 11 [1]/17 2/17
      (5.8) (23.5) (64.7) (11.8)
      Table 1. Baseline Characteristics

      Values are expressed as mean±standard deviation or number (%). The group with thyroglobulin antibody (TgAb) levels greater than 60 U/L includes individuals with at least one TgAb measurement exceeding 60 U/L, either during unstimulated or stimulated Tg measurements, in the first or second test.

      LN, lymph node; ETE, extrathyroidal extension; Tg, thyroglobulin; rhTSH, recombinant human thyroid-stimulating hormone; T4, levothyroxine.

      A total of 38 individuals (14.2%) underwent both central and lateral lymph node dissection, including 29 individuals (13.7%) in the group with TgAb levels less than 60 U/L and nine individuals (16.1%) in the group with TgAb levels greater than 60 U/L;

      Two methods were used for Tg stimulation measurement. rhTSH stimulation involved a 2-day course of intramuscular injections of 0.9 mg rhTSH, administered 2 days before radioactive iodine intake. Off-T4 stimulation was achieved via levothyroxine withdrawal; cBiochemical incomplete response was defined as suppressed Tg >1 ng/mL, thyroidstimulating hormone-stimulated Tg >10 ng/mL, or rising anti-Tg antibody levels in the absence of structural disease [6].

      Table 2. Characteristics of Patients Showing Discordant ultraTg Levels among Those with Unstimulated hsTg <0.2 ng/mL or Structural Recurrence during Follow-up

      All individuals were B-type Raf kinase (BRAF; +) and telomerase reverse transcriptase (TERT; –).

      ultraTg, ultrasensitive thyroglobulin; hsTg, highly sensitive thyroglobulin; TgAb, thyroglobulin antibody; TSH, thyroid-stimulating hormone; ATA, American Thyroid Association; ETE, extrathyroidal extension; LN, lymph node; RAI, radioactive iodine; miETE, minimal extrathyroidal extension, NA, missing value.

      TSH levels were measured at the time of the first unstimulated Tg measurement;

      Patients 5, 6, 7, 8, 9, 10, and 11 experienced structural recurrence.

      Table 3. Comparison of Diagnostic Performances in Predicting Stimulated Tg

      Values are expressed as number (%). The number of patients with confirmed structural recurrence is recorded in [ ], based on the first measurement. Two individuals with structural recurrence were excluded from the table results due to thyroglobulin antibody levels ≥60 U/L. For unstimulated Tg, an excellent response was defined as a level below the cut-off; levels between the cut-off and 1.0 ng/mL were considered indeterminate response; and levels exceeding 1.0 ng/mL indicated a biochemical incomplete response. For stimulated Tg, an excellent response was defined as a level below 1.0 ng/mL; levels between 1.0 and 10.0 ng/mL were considered indeterminate response; and levels exceeding 10.0 ng/mL indicated a biochemical incomplete response.

      Tg, thyroglobulin; hsTg, highly sensitive Tg; ultraTg, ultrasensitive Tg.

      Table 1.

      Table 2.

      Table 3.

      Endocrinol Metab : Endocrinology and Metabolism
      © Korean Endocrine Society.
      TOP