Recombinant human thyrotropin versus thyroid hormone withdrawal in differentiated thyroid carcinoma follow-up: a single center experience

Abstract

Introduction

Our goal was to evaluate and compare the diagnostic utility of thyroid hormone withdrawal (THW) and recombinant thyroid-stimulating hormone (rhTSH) methods in detecting recurrence/persistence (R/PD) of differentiated thyroid carcinoma (DTC).

Methods

The study included 413 patients with DTC who underwent total thyroidectomy and had remnant ablation. DxWBS, s-Tg levels, R/PD were evaluated retrospectively. A s-Tg level≥2 ng/mL was considered as “positive s-Tg”.

Results

DxWBS and s-Tg levels were evaluated with rhTSH in 116 and THW in 297 subjects, respectively. The sensitivity and specificity of “positive s-Tg” for R/PD in THW group were 77.3% and 92.7%, with 90.3% accuracy, respectively. The sensitivity and specificity of “positive s-Tg” for R/PD in rhTSH group were 58.8% and 100% with 93.9 % accuracy, respectively. An uptake outside thyroid bed at WBS showed a sensitivity of 17.1%, specificity of 100% for R/PD with 89.4% accuracy in THW group. An uptake outside thyroid bed at WBS showed a sensitivity of 7.7%, specificity of 100% for R/PD with 88.8% accuracy in rhTSH group.

Conclusion

Method of TSH stimulation did not influence the reliability of DxWBS. The “positive s-Tg level” had a higher sensitivity with THW when compared to rhTSH in detecting R/PD.

Introduction

Thyroid carcinoma is the most common endocrine cancer and accounts for 1-2% among all malignancies (1). Approximately 20% of DTC patients experience the recurrence throughout their lifetime, and more than 80% of recurrences occur within the 10 years after diagnosis (2). However, recurrences may occur even 20 years after the initial treatment (3). Thus, long term follow-up strategy has crucial role in management of DTCs.

The recommendation of the American Thyroid Association (ATA) for follow-up of DTC patients is the measurement of basal thyroglobulin (b-Tg) on thyroxine therapy or stimulated thyroglobulin (s-Tg) after thyroid stimulating hormone (TSH) stimulation in low and intermediate-risk patients who had radioactive iodine (RAI) remnant ablation (with moderate quality evidence) (4). Development of high sensitive Tg assays caused a shift to monitoring DTC with b-Tg measurements instead of s-Tg. On the other hand, it was suggested that recurrent or persistent disease can be overlooked in more than 20% of DTC patients by measurement of b-Tg alone (5, 6).

Although the role of diagnostic whole body scan (DxWBS) in the follow-up of DTC has been controversial, it has still been extensively used in routine practice (7-16). ATA (2015) recommended use of DxWBS in the follow-up of patients with high or intermediate risk status for recurrence with low quality evidence (4). Some clinical settings were described which DxWBS should be given priority. These conditions were: uptake outside the thyroid bed at post-therapeutic WBS (PtWBS), uninformative PtWBS caused by large remnants and existence of anti-thyroglobulin antibodies (TgAb). Although DxWBS has not been a preferred method in monitoring DTC for a while, benefits in detecting recurrence were reported in a number of recent studies, even for patients with low risk status or early stage of disease (9, 15, 17).

The two methods of preparation to s-Tg measurement and DxWBS are THW and rhTSH (Thyrogen®; Genzyme Corporation, Cambridge, MA, USA) administration (4). The morbidity of hypothyroidism and decreased quality of life after THW caused a shift to testing with rhTSH stimulation (6, 18, 19). Phase III clinical trials of Ladenson et al. and Haugen et al., revealed that rhTSH administration was safe and effective in stimulating radioiodine uptake and serum Tg (6, 20). However, data in diagnostic utility of these two methods is limited (6, 20, 21). Some studies reported that preparation to WBS with THW was superior to rhTSH, whereas two methods were similar in others (21-26).

The purpose of our study was to evaluate the diagnostic utility of DxWBS and s-Tg measurements in detecting recurrence/persistence of DTC, and to compare the performance of preparation with THW or rhTSH methods.

Material and Methods

Patients

The study included 413 patients with DTC who underwent total thyroidectomy and RAI ablation/adjuvant therapy, and had follow-up data for at least 60 months in Ankara University Faculty of Medicine. The study was approved by the local ethical committee.

Demographic characteristics, histopathological findings (tumor size, extrathyroidal invasion, multifocality, tumor stage, and capsular invasion), presence of lymph node metastasis (LNM), distant metastasis, Dx/PtWBS, Tg levels, s-TSH, neck ultrasonography (US) and other imaging modalities [computed tomography (CT), magnetic resonance imaging (MRI), fluorine-18 fluorodeoxyglucose positron emission tomography (FDG-PET)] results were assessed retrospectively. Anti-thyroglobulin antibody positive patients were excluded from the study.

The risk stratification system for recurrence/persistence was defined according to the Revised American Thyroid Association Management Guidelines for Patients with Thyroid Nodules 2009, because size of lymph node metastases (LNMs) were not available in a high proportion of pathology reports. Clinico-pathological staging was performed according to the 8^th edition of American Joint Committee on Cancer TNM staging system.

Recurrence/persistence of disease (R/PD) was assessed from records of b-Tg levels, neck ultrasonography, other imaging modalities (US, MRI/CT/FDG PET-CT, PtWBS) and biopsy results. Cervical LNMs were confirmed with fine needle aspiration, Tg washout procedure or histopathological diagnosis. Response to treatment was first assessed after 6-12 months of initial treatment and then every 6-12 months. DxWBS and s-Tg levels were not used to define R/PD. Persistence was defined as presence of structural or persistent biochemical (b-Tg ≥1 ng/mL) evidence of disease at initial evaluation after total thyroidectomy and remnant ablation. Recurrence was defined as new biochemical (b-Tg ≥1 ng/mL), or structural evidence of tumor after total thyroidectomy and remnant ablation.

Thyroglobulin and 131I whole body scan

Serum thyroglobulin measurements were performed by radioimmunoassay (IRMA) method using a commercial kit (Beckman Access Systems, Beckman Coulter Ltd). The functional sensitivity of assay was 0.1ng/mL. The samples were obtained 72 hours after the administration of second dose in rhTSH group and before the administration of 131I in THW group. A s-Tg level lower than 2 ng/mL was considered as “negative s-Tg”. A s-Tg level equal to or greater than 2 ng/mL was considered as “positive s-Tg”. (6, 17, 21).

Thyroid hormone replacement was stopped and low iodine intake diet was applied four weeks before diagnostic procedure in THW group. WBS was performed with 5 mCi 131I, anterior and posterior projections were obtained at the 24 and 72 hours after administration of 131I, using a gamma camera with high energy collimator (GE 4000İXC-T/STARCAM). Patients in rhTSH group continued treatment with suppressive doses of thyroid hormone and received intramuscular injection of 0.9 mg rhTSH (Thyrogen®; Genzyme Corporation, Cambridge, MA, USA) 24 and 48 hours before the 131I administration. Scans were performed at least 6 months after, if a procedure containing iodine was relevant.

A random procedure was selected and evaluated in the statistical analysis, if patient underwent multiple testing. Patients who had only uptake at thyroid bed were not included in the statistical analysis of accuracy.

Statistical analysis

Categorical data were compared using the chi-square or Fisher’s exact test. Group data with a normal distribution were compared using the Student t test or analysis of variance, and nonparametric data were compared using the Mann-Whitney U or Kruskal-Wallis test. Values are expressed as mean ± standard deviation or median, as appropriate, with P<.05 considered statistically significant. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of THW and rhTSH methods in detecting persistence/recurrence of disease were calculated. A receiver operating characteristic (ROC) analysis was performed to determine the cut-off scores. All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) software, version 18.0 (IBM Corp, Armonk, NY).

Results

Characteristics of patients

A total of 413 DTC patients (149 male, 264 female) were included in this study. Diagnostic procedures were performed with THW (n=297) and rhTSH (n=116). As shown in Table 1, there were no differences between THW and rhTSH groups with respect to mean age at diagnosis, gender, age groups, type of DTC, tumor size, tumor stage and ATA risk groups. Mean RAI doses were comparable between two groups (121.6±27.2 mCi in THW and 119.0±34.5 mCi in rhTSH, p=0.41). The frequency of R/PD was 14.8% (n=61) in the whole group. The presence of R/PD was comparable between two groups [14.8% (n=44) in THW vs. 14.7% (n=17) in rhTSH, p=0.96].

Table 1.

Patient characteristics and histopathological features of tumors inTHW and rhTSH groups

	THW (n=297)	rhTSH (n=116)	p
Age at diagnosis	42.5±12.4	40.6±12.0	0.15
Age groups; <55 ≥55	241 56	101 15	0.19
Gender (female/male),n	197/100	49/67	0.10
Type of DTC;n (%)  Papillary  Follicular	282 (94.9) 15 (5.1)	109 (94.0) 7 (6.0)	0.68
PTC subtypes;n (%)  Conventional  Follicular	221 (78.4) 61 (21.6)	84 (77.0) 25 (23.0)	0.76
Tumor size, mm	13.5±0.79	11.9±1.10	0.20
T; T1 T2 T3 T4	215 (72.4) 52 (17.5) 21 (7.1) 9 (3.0)	75 (64.7) 27 (23.3) 12 (10.3) 2 (1.7)	0.28
N; N0 N1a N1b	228 (76.8) 47 (15.8) 22 (7.4)	76 (65.5) 31 (26.7) 9 (7.8)	0.04
M; M0 M1	291 (98) 6 (2.0)	115 (99.1) 1 (0.9)	0.41
Stage; I II III IV	278 (93.6) 11 (3.7) 1 (0.3) 7 (2.4)	110 (94.8) 4 (3.4) 1 (0.9) 1 (0.9)	0.69
Risk status for R/PD Low, Intermediate, High,	171 (57.6) 114 (38.4) 12 (4.0)	63 (54.3) 46 (39.7) 7 (6.0)	0.6

Mean peak TSH level was higher in THW group when compared to rhTSH (95.4±44.6 mIU/L vs. 80.8±37.0 mIU/L, p=0.02). Frequency of patients with peak TSH level below 30 mIU/L was higher in rhTSH group when compared to THW [12.7% (n=15) vs. 3.7% (n=11), p=0.001].

S-Tg levels

The median s-Tg level was 0.1 ng/mL (min-max: 0.1-5.58) in DTC patients without persistent/recurrent (R/PD) disease and 16 ng/mL (0.1-655) in patients with R/PD (<0.001). The median s-Tg level was higher in THW group when compared to rhTSH [0.1 ng/mL (0.1-655) vs. 0.1 ng/mL (0.1-198), p=0.001). The median s-Tg level was positively associated with the presence of R/PD in rhTSH group (p<0.001) and THW group (p<0.001). The s-Tg level was higher in THW group when compared to rhTSH [median 0.1 ng/mL (0.1-655) vs. 0.1 (0.1-198), p=0.01].

The ROC curve analysis showed that 2.57 ng/mL was the best cut-off value for the s-Tg level in THW group to indicate R/PD, with a sensitivity of 85.4% (95% CI, 72.8% to 92.8%), specificity of 97.2% (95% CI, 94.3% to 98.6%), respectively (AUC=0.915, p<0.001). The best cut-off value for the s-Tg level in rhTSH group to indicate R/PD was 1.35 ng/mL, with a sensitivity of 76.5% (95% CI, 52.7% to 90.4%), specificity of 99.0% (95% CI, 94.5% to 99.8%), respectively (AUC=0.922, p<0.001).

Negative s-Tg levels in R/PD

Among 61 patients with R/PD; the s-Tg level was ≥2 ng/mL in 37 (84.1%) patients in THW and 11 (61.1%) patients in rhTSH groups, respectively. Seven patients (15.9%) with R/PD in THW group and 6 patients (38.9%) with R/PD in rhTSH group had “negative s-Tg” level (p=0.09). None of the patients with R/PD and “negative s-Tg” had uptake outside the thyroid bed in DxWBS. Among “s-Tg negative” 7 patients with R/PD in THW group; lymph node metastasis (LNM) was demonstrated with neck US+lymph node biopsy in 6 patients and with FDG-PET+lymph node biopsy in one patient. Six patients with R/PD and “negative s-Tg” in rhTSH group had LNM diagnosed with neck US+biopsy.

Positive s-Tg level

A “positive s-Tg” level showed a sensitivity of 78.7% (95% CI, 66.9%-87.1%), specificity of 94.6% (95% CI, 91.7%-96.5%), PPV of 71.6% (95%CI, 66.9%-75.9%), NPV of 96.2 % (95% CI, 93.8%-97.8%) for R/PD with 92.3% accuracy in whole group. The false positive and negative rates for “positive s-Tg” were 5% and 21%, respectively.

Diagnostic accuracy of “positive s-Tg” with two methods in detecting R/PD was summarized in Table 2. A “positive s-Tg” level with THW showed a sensitivity of 84.1%, specificity of 92.9% for R/PD with 91.6% accuracy. A “positive s-Tg” level with rhTSH showed a sensitivity of 64.7%, specificity of 98.9% for R/PD with 93.9% accuracy. The false negative rate of “positive s-Tg” was 35.2% and 15.9% in rhTSH and THW groups, respectively (p=0.02). The false positive rate of “positive s-Tg” was 1.0% and 7.1%, in rhTSH and THW groups, respectively (p=0.001).

Table 2.

Diagnostic power of s-Tg level in detecting R/PD in THW and rhTSH groups

			R/PD+,n (%)	R/PD-,n (%)	Se % (95%Cl)	Spe % (95%CI)	PPV% (95%CI)	NPV% (95%CI)	Acc %
Whole group n=413	THW, n=297	s-Tg≥2, n(%)	37 (67.3)	18 (32.7)	84.1 (70.6-92.1)	92.9 (88.6-95.5)	67.3 (61.6-72.5)	97.1 (94.3-98.6)	91.6
		s-Tg<2, n(%)	7 (2.9)	235 (97.1)
	rhTSH n=116	s-Tg≥2, n(%)	11 (100)	1 (0)	64.7 (41.3-82.7)	98.9 (94.5-99.8)	91.7 (84.7-95.8)	94.2 (87.9-97.5)	93.9
		s-Tg<2, n(%)	6 (5.8)	98 (94.2)
Intermediate- high risk n=179	THW, n=126	s-Tg≥2, n(%)	29 (80.6)	7 (19.4)	93.6 (79.3-98.2)	92.6 (85.6-96.4)	80.6 (72.4-86.8)	97.8 (92.9-99.5)	92.9
		s-Tg<2, n(%)	2 (2.2)	88 (97.8)
	rhTSH n=53	s-Tg≥2, n(%) s-Tg<2, n(%)	9 (100) 5 (11.4)	0 (0) 39 (86.6)	64.2 (38.8-83.7)	100 (91.0-100)	100 (91.6-99.8)	88.6 (76.2-95.3)	88.7

Forty eight patients who experienced R/PD had “positive s-Tg” level and 21 (43.7%) of them had b-Tg level lower than 1 ng/mL. Among 21 patients with “negative b-Tg” and “positive s-Tg”, 85% (n=18) had stage I disease.

Among 352 patients without evidence of R/PD, 60 patients (17.3%) had b-Tg level ≥ 0.2 ng/mL, although they had “negative stimulated Tg”. The false negative and positive rates of “b-Tg level ≥0.2 ng/mL” was 18% and 20%, respectively. All patients with distant metastases had s-Tg levels higher than 10 ng/mL (10.4-655 ng/mL).

s-Tg levels in intermediate-high risk patients

A “positive s-Tg” level with THW showed a sensitivity of 93.6%, specificity of 92.6%, PPV of 80.6%, NPV of 97.8 % for R/PD with 92.9% accuracy in intermediate-high risk patients (Table 2). The false negative and positive rates of “positive s-Tg” with THW were 6.4% and 7.4% in this group, respectively. A “positive s-Tg” level with rhTSH showed a sensitivity of 64.3%, specificity of 100%, PPV of 100%, NPV of 88.6% for R/PD with 90.6% accuracy in intermediate-high risk patients (Table 2). The false negative rate of “positive s-Tg” with rhTSH was 35.7% in this group, whereas there were no false positive cases. All s-Tg negative cases with R/PD (n=7) in intermediate/high risk group, had LNM diagnosed with neck US and LN biopsy.

DxWBS

The DxWBS results in two groups were summarized in Table 3. An uptake outside thyroid bed at DxWBS with THW showed a sensitivity of 26.3%, specificity of 100%, PPV of 100%, NPV of 89.5% for R/PD with 89.9% accuracy (Table 4). False negative rate of DxWBS was 73.6% in THW group. An uptake outside thyroid bed at DxWBS with rhTSH showed a sensitivity of 26.7%, specificity of 100%, PPV of 100%, NPV of 89.5 % for R/PD with 89.9% accuracy in rhTSH group (Table 4). False negative rate of DxWBS was 73.3% in rhTSH group. Six distant metastases were identified by DxWBS and 4 (83.3%) of them were also identified by PtWBS.

Table 3.

Diagnostic whole body scans in THW and rhTSH groups

	THW, n(%)	rhTSH
No uptake	267 (89.9)	105 (90.4)
Thyroid bed only	20 (6.8)	7 (6.0)
Lateral neck	5 (1.7)	2 (1.8)
Mediastinum	1 (0.3)	-
Lung only	3 (1.0)	1 (0.9)
Bone only	-	1 (0.9)
Lung and bone	1 (0.3)	-

Table 4.

Diagnostic power of WBS in detecting R/PD in THW and rhTSH groups*

			R/PD+	R/PD-	Se % (95%Cl)	Spe % (95%CI)	PPV% (95%CI)	NPV% (95%CI)	Acc %
Whole group* (n=386)	THW	WBS +, n(%)	10 (100)	0 (0)	26.3 (14.9-42.1)	100 (98.4-100)	100 (98.3-99.4)	89.5 (85.1-92.8)	89.9
		WBS −, n(%)	28 (10.5)	239 (89.5)
	rhTSH	WBS +, n(%)	4 (100)	0 (0)	26.7 (10.9-51.9)	100 (96.1-100)	100 (95.8-99.9)	89.5 (81.8-94.3)	89.9
		WBS −, n(%)	11 (10.5)	94 (89.5)
Intermediate- high risk** (n=167)	THW	WBS +, n(%)	9 (100)	0 (0)	34.6 (19.4-53.8)	100 (95.9-100)	100 (96.0-99.9)	84.3 (76.1-90.1)	85.4
		WBS −, n(%)	17 (15.7)	91 (84.3)
	rhTSH	WBS +, n(%)	4 (100)	0 (0)	26.7 (10.9-51.9)	100 (96.1-100)	100 (95.8-99.9)	89.5 (81.8-94.3)	89.9
		WBS −, n(%)	11 (10.5)	94 (89.5)

*Patients with uptake at thyroid bed were excluded from the statistical analysis (n=27).

** Patients with uptake at thyroid bed were excluded from the statistical analysis (n=12).

DxWBS in intermediate-high risk patients

An uptake outside thyroid bed at DxWBS with THW showed a sensitivity of 34.6% specificity of 100%, PPV of 100%, NPV of 84.3% for R/PD with 85.4% accuracy in intermediate-high risk patients. The false negative rate of DxWBS was 65% in this group. An uptake outside thyroid bed with rhTSH showed a sensitivity of 26.7% specificity of 100%, PPV of 100%, NPV of 89.5% for R/PD with 89.9% accuracy in intermediate-high risk patients. The false negative rate of DxWBS was 73% in this group.

Negative DxWBS and positive s-Tg levels

Thirty nine patients had R/PD and negative WBS (28 patients in THW, 11 patients in rhTSH groups, respectively). Among patients with R/PD and negative WBS; 75% (n=21) and 54.5% (n=6) had positive s-Tg levels in THW and rhTSH groups, respectively (p=0.24). Only one patient with R/PD and negative WBS had TSH level lower than 30 mIU/L (in THW group).

In whole group frequency of negative DxWBS and “positive s-Tg” level was 10% (n=41). Twenty seven (66%) of them had R/PD. In THW group; 21 patients with R/PD had negative DxWBS and “positive s-Tg” level. In this group, metastatic disease was diagnosed with cervical US+LN biopsy, FDG-PET (mediastinal LNM) and Thorax CT (lung metastases) in 76.2% (n=16), 9.5% (n=2) and 4.8% (n=1), respectively. Two patients (9.5%) did not have structural evidence of disease in THW group.

In rhTSH group; 6 patients with R/PD had negative DxWBS and “positive s-Tg” level. In this group, metastatic disease was diagnosed with neck US+LN biopsy in 83.3% (n=5) and FDG-PET (lateral LNM) in 16.7% (n=1) of patients. All LNMs without 131I uptake which were confirmed by neck US and LN biopsy (n=21), were smaller than 15 mm and 80% of them were ≤ 10 mm in maximal diameter.

WBS uptake at thyroid bed

In whole group, 27 patients (6.5%) had uptake only at thyroid bed on DxWBS. Among them 20 patients (74.1%) were in THW and 7 (25.9%) were in rhTSH group. R/PD was diagnosed in 8 (29.6%) patients with uptake only at thyroid bed (n=27). There was no difference between rhTSH and THW groups with regard to presence of R/PD in patients who had uptake at only thyroid bed on DxWBS [28.5% (n=2) vs. 30% (n=6), p=0.9]. Among patients who had uptake only at thyroid bed with THW method (n=20); central LNM/locoregional disease was found by US+LN biopsy in 4 patients (25%), central+lateral LNM was found with FDG-PET in one patient (5%) and lung metastasis was found with Thorax CT in one patient (5%). All of these patients had “positive s-Tg” level. Among patients who had uptake only at thyroid bed with rhTSH method (n=20), central LNM/locoregional disease was found in two patients (25%) by neck US+LN biopsy and one of them had a negative s-Tg level.

Discussion

In the present study, we assessed the diagnostic utility of s-Tg measurement and DxWBS in detecting recurrence/persistence of DTC, and compared the performance of preparation with rhTSH and THW methods. Mean peak TSH level was higher, and frequency of “insufficient TSH” elevation was lower in THW group when compared to rhTSH. The sensitivity of DxWBS was approximately ~25% with both methods, but was slightly higher with THW method in patients with intermediate-high risk status. The frequency of “DxWBS negative, s-Tg positive” cases was 10%, and 66% of them had R/PD demonstrated by other imaging techniques. Specificity of DxWBS in detecting R/PD was 100% with both methods. Positive s-Tg level had higher sensitivity in THW group when compared to rhTSH in detecting R/PD. Specificity of positive s-Tg level was slightly higher with rhTSH than THW.

The introduction of high sensitive, second generation Tg assays raised concerns about the necessity of s-Tg measurement in detecting R/PD. It was suggested that measurement of high sensitive b-Tg may be sufficient in monitoring DTC with low risk status (27, 28). Besides, Moreno et al. reported that s-Tg measurement was not necessary for patients with intermediate risk status, if an excellent response to initial therapy was achieved according to the definition in ATA 2015 guidelines (14). However, these findings were not verified in other clinical studies (5, 15). A “consensus report” of eight studies was designed to examine the need for s-Tg measurement in follow-up of 1028 patients with low risk status (5). Their results showed that 21% of patients with b-Tg level below 1 ng/mL, had s-Tg level above 2 ng/mL. A recent study by Campenni et al. demonstrated that undetectable/low (<1 ng/mL) b-Tg level may not exclude the metastatic disease in early stages of DTC (15). Additionally, it was reported that a slightly increased b-Tg level may not be an accurate indicator of recurrence (29). In the study of Castagna et al., 15% of patients without R/PD had detectable basal Tg, although their s-Tg level was lower than 1 ng/mL (29). Consistent with Castagna et al., we observed that 17.3% of patients without evidence of R/PD had b-Tg level ≥ 0.2 ng/mL, although they had “negative stimulated Tg”. In our study, the 43.7% of patients with R/PD and positive s-Tg had b-Tg level lower than 1 ng/mL. Our findings demonstrated that metastatic disease may be overlooked if disease status is evaluated only with b-Tg level.

The majority of the studies suggested that s-Tg levels with THW were higher than rhTSH (6, 30, 31). Additionally, the time needed for stimulating Tg may be variable in rhTSH method (32). On the contrary, Robbins et al. reported similar diagnostic accuracy of s-Tg with both methods (21). In our study, peak s-Tg level and cut-off value that predicted R/PD were higher in THW group when compared to rhTSH. The sensitivity of “positive s-Tg” was significantly higher in THW group when compared to rhTSH, whereas specificity was slightly higher with rhTSH. Also some patients in rhTSH group did not reach optimal peak TSH levels.

The routine use of DxWBS in follow-up of DTC has still been debated (4, 7, 13, 15). Because of the accumulating data regarding the high diagnostic efficacy of s-Tg and neck US in detecting R/PD, DxWBS was recommended only for intermediate/high risk patients with specific conditions by ATA and British Thyroid Association with low quality of evidence (4, 13). The study by Rosario et al., demonstrated that DxWBS may be avoided for patients with intermediate risk status who had negative Pt-WBS and s-Tg<1 ng/mL with negative TgAb (11). In the study of Jeon et al., it was found that 75% of LNMs of intermediate risk DTCs could not be visualized on DxWBS (10). No benefit of routine DxWBS in addition to s-Tg determination was found for patients with high risk features by Verburg et al. and de Meer et al. (8, 33). These results advocated the use of s-Tg measurement without DxWBS, for the majority of patients with intermediate and high risk status for R/PD. Contrary to previous reports, Carvalho et al. suggested that DxWBS has still been beneficial in follow-up of DTCs from all risk groups (17). Authors proposed that DxWBS may be omitted if initial DxWBS was negative and s-Tg level was below functional sensitivity, since the risk of recurrence was very low in this group. Our results were concordant with the previous data, reporting low sensitivity of DxWBS when compared to “positive s-Tg” levels (17). Two distant metastases which did not have uptake at PtWBS were identified by DxWBS. Also, one mediastinal LNM was confirmed by DxWBS. It may be speculated that DxWBS ensured clinical information for only 4.9% of R/PD in our study group.

Patients with “negative DxWBS” but “positive s-Tg” pose as a diagnostic challenge and they account for up to 20-43% of DTCs (34, 35). In a preliminary study, Pacini et al. demonstrated that 12.6% of DTC patients, who underwent diagnostic procedure, had “positive s-Tg and negative DxWBS” result, whereas 94% of them had metastatic foci which became apparent at WBS with therapeutic doses of 131I (36). In our previous study, we have reported that empirical therapeutic doses of 131I may help in localization of the disease in Tg positive patients without anatomical evidence of persistent disease (37). Investigations revealed that negative DxWBS and positive s-Tg levels in patients with metastases were attributable to improper patient preparation, presence of small tumors that cannot be detected by scans, or impaired organification and uptake of iodine in tumor tissue because of dedifferentiation (38). The frequency of negative DxWBS and positive s-Tg was 10% in our series and this finding was consistent with the data of Pacini et al., but lower than others (34-36). We applied low iodine intake diet and performed WBS at least 6 months after if an iodine containing procedure was relevant. Only one patient with R/PD and “negative WBS” had insufficient TSH elevation in THW group. So, improper preparation was out of question in our study. The majority of lymph node metastases with “negative DxWBS, positive s-Tg” were confirmed by “neck US+biopsy” in both groups. Our findings confirm that neck US is a very effective method in detecting metastatic foci for both DxWBS positive and negative patients with R/PD. However, sensitivity and specificity of neck US or other imaging methods were not addressed in this study. We also observed that LNMs without 131I uptake which were confirmed by “neck US and LN biopsy” were smaller than 15 mm and 80% of them were ≤ 10 mm in maximal diameter. The high frequency of WBS negative lateral LNMs in our study may be due to inadequate uptake of 131I by small tumoral foci, as well as low resolution capacity of gamma camera for subcentimeter foci (9). The 131I SPECT/CT (single photon emission computerized tomography) was shown to improve diagnostic accuracy of planar WBS in the long term follow-up of DTC with availability of anatomic localization (39-42). Routine complementary use of 131I-SPECT/CT may be considered to improve the diagnostic utility of WBS.

Preparation to WBS with THW vs. rhTSH methods were investigated with both therapeutic and diagnostic doses of radioiodine, but results were incompatible (20-23, 32, 43-46). The studies by Ladenson et al. and Van Nostrand et al., compared the efficacy of two methods and showed that THW was significantly better than rhTSH in detecting metastatic foci on DxWBS (20, 22). Also, some case reports showed that THW was superior to rhTSH in preparation to Dx/PtWBS (23, 24, 32, 45). The studies by Freudenberg et al. and Pötzi et al. demonstrated that iodine uptake by metastatic foci was lower with rhTSH preparation than THW (24, 25). This results were explained by the high clearance of 131I, reduction of the dose absorbed by remnant/metastasis in rhTSH method and insufficient expression of sodium iodide symporter expression, due to short stimulation period after rhTSH method (23-26, 47, 48). On the other hand, the comparison of two methods by Liepe et al. and Robbins et al. showed statistically similar sensitivity of DxWBS in detecting metastases of DTC (21, 43). Recently, Hong et al., investigated 131I biokinetics in THW and rhTSH methods, and observed that the retention rate and effective half time was lower for LNM than thyroid remnant, whereas effectiveness of two methods was similar (44). Our findings were concordant with their data, as uptake at lateral neck was rare in patients who proved to have LNM with other imaging modalities. There was no significant difference between the two methods in detecting R/PD. However, sensitivity of DxWBS was slightly higher with THW when compared to rhTSH in intermediate- high risk patients.

In conclusion, the method of TSH stimulation did not influence the sensitivity and specificity of testing for DxWBS in our study. We observed that s-Tg measurement was beneficial and should not be replaced by b-Tg alone in routine follow-up procedure. The “positive s-Tg level” had a higher sensitivity with THW when compared to rhTSH in detecting R/PD. The results of our study suggested that DxWBS did not provide additional information to neck US and s-Tg measurement, especially in detecting cervical LNMs. Routine complementary use of 131I-SPECT/CT may be beneficial for this group of patients. Besides, WBS should be preserved for follow-up of patients with known radioiodine avid systemic metastases, TgAb positivity and high s-Tg values with normal neck US.

Conflict of interest

The authors declare that they have no conflict of interest.