Abstract
Purpose
To determine whether thyroid-stimulating hormone level ≥ 30 mU/L is necessary for radioiodine (131I) remnant ablation (RRA) in patients with differentiated thyroid cancer (DTC), as well as its influencing factors and predictors.
Methods
A total of 487 DTC patients were retrospectively enrolled in this study. They were divided into two groups (TSH < 30 and ≥ 30 mU/L) and further divided into eight subgroups (0–<30, 30–<40, 40–<50, 50–<60, 60–<70, 70–<80, 80–<90, and 90–<100 mU/L). The simultaneous serum lipid level, successful rate of RRA and its influencing factors in different groups were analyzed. The receiver operating characteristic curves derived from pre-ablative thyroglobulin (pre-Tg) and pre-Tg/TSH ratio were compared for RRA success prediction performance.
Results
There was no statistical difference in success rates of RRA between the two groups (P = 0.247) and eight subgroups (P = 0.685). Levels of total cholesterol (P < 0.001), triglyceride (P = 0.006), high-density lipoprotein cholesterol (P = 0.024), low-density lipoprotein cholesterol (P = 0.001), apolipoprotein B (P < 0.001), and apolipoprotein E (P = 0.002) were significantly higher while apoA/apoB ratio (P = 0.024) was significantly lower at TSH ≥ 30 mU/L group. Pre-Tg level, gender, and N stage were influencing factors for RRA. The area under the curve of pre-Tg level and pre-Tg/TSH ratio was 0.7611 (P < 0.0001) and 0.7340 (P < 0.0001) for all enrolled patients and 0.7310 (P = 0.0145) and 0.6524 (P = 0.1068) for TSH < 30 mU/L, respectively.
Conclusion
TSH ≥ 30 mU/L may not be necessary for the success of RRA. Patients with higher serum TSH levels prior to RRA will suffer from severer hyperlipidemia. Pre-Tg level could be used as a predictor for the success of RRA, especially when TSH < 30 mU/L.
Introduction
Thyroid cancer, accounting for 586,000 cases worldwide, was ranked in ninth place for incidence in 2020 (1). Differentiated thyroid cancer (DTC) accounts for more than 90% of all thyroid malignancies (2). DTC patients usually have a good prognosis due to the biological behavior and standard treatments. Three conventional treatment regimens for thyroid cancer are surgery, radioiodine (131I) therapy, and thyroid-stimulating hormone (TSH) suppression. Among them, 131I is considered to be a significant treatment for eliminating the remnant thyroid tissue after thyroidectomy, which can increase the specificity of detectable serum thyroglobulin (Tg) as a tumor marker and potentially improve the quality of future 131I whole-body scan (3, 4).
In the late 1970s, Edmonds and colleagues suggested that tumor should not be considered incapable of concentrating 131I adequately until serum TSH levels have exceeded 30 mU/L (5). Ever since, this threshold has been quoted in the international literature, including many guidelines (6, 7, 8), as the minimum level of TSH stimulation for radioiodine therapy including thyroid remnant ablation. However, there is no definite evidence yet that TSH level ≥ 30 mU/L is necessary for successful radioiodine remnant ablation (RRA).
Moreover, patients who are in the process of elevating TSH level may develop hypothyroidism symptoms, which is a risk factor for physical health. Previous studies have demonstrated that TSH is an important modulator of serum lipid metabolism under the condition of hypothyroidism (9). Higher TSH level leads to an unfavorable lipid profile (10). Therefore, it still remains to be further explored whether a high TSH level is truly necessary before RRA considering the risk of hyperlipidemia or other side effects it may potentially cause.
The therapeutic effect of radioiodine for thyroid remnant ablation can be varied significantly among patients with different characteristics (11). However, the influencing factors are still not well understood. This problem creates a major obstacle to improve the success rates of RRA. Besides, to the best of our knowledge, no study has reported the predictor of the therapeutic effect of radioiodine for thyroid remnant ablation in patients with TSH < 30 mU/L. A valid predictor could suggest whether and when to start RRA for patients with low TSH levels. Thus, it has important implications for helping clinicians design a well-established treatment regimen.
Based on the earlier problems, the objectives of the current study were to determine whether a TSH level ≥ 30 mU/L is truly necessary for successful RRA, explore the influencing factors of RRA, and thus find predictors with the best predictive performance for RRA success, especially for patients with TSH < 30 mU/L.
Materials and methods
Patients and collection of variables
We retrospectively reviewed the clinical records and follow-up data of patients in the Department of Nuclear Medicine, Shanghai Sixth People's Hospital between January 2013 and May 2020. A total of 487 patients who satisfied the following criteria were finally enrolled: (i) differentiated thyroid carcinoma confirmed by histopathology after total thyroidectomy; (ii) first time 131I treatment with the purpose of thyroid remnant ablation; (iii) no evidence of neck lymph node metastases on ultrasound and no evidence of distant metastasis suspected by the clinical manifestations as well as medical imaging such as computed tomography (CT), magnetic resonance imaging, and/or fluorine-18-fluorodeoxyglucose positron emission tomography/CT before RRA; (iv) levothyroxine (LT4) withdrawal over 3–4 weeks before RRA; (v) pre-ablative Tg (pre-Tg) ranges from 0.1 to 10 ng/mL and Tg antibody (TgAb) < 100 IU/mL; (vi) simultaneous serum lipid values available; (vii) TSH-suppressed Tg after RRA (post-Tg) are available. Exclusion criteria were as follows: (i) suffering from other diseases; (ii) contrast-enhanced CT within 3 months before RRA; (iii) diagnostic 131I whole-body scan before RRA; (iv) abnormal radioiodine uptake lesions outside the thyroid bed in 131I whole-body scan or metastases detected on 131I single-photon emission computed tomography/computed tomography after 131I treatment; (v) TSH level ≥ 100 mU/L (as the exact value cannot be obtained when TSH ≥ 100 mU/L and all reported as TSH level ≥ 100 mU/L under this condition during clinical practice in our institution); (vi) abnormal serum lipid level; (vii) on medications that affect iodine uptake or metabolism; (viii) loss of follow-up and inadequate information.
The surgical and histopathologic data collected in our study were sex, age, lesions condition, maximum tumor diameter, pathological types, T stage, and N stage. Among them, pathological types were divided into classical papillary thyroid cancer, subtypes of papillary thyroid cancer, and follicular thyroid cancer; lesions condition was divided into solitary and multiple. T stage and N stage were evaluated according to the 8th edition of the tumor-node-metastasis (TNM) classification system. T stage was divided into T1, T2, T3, and T4; N stage was divided into N0, N1a, N1b, and Nx. The following data regarding RRA were collected: the activity of radioiodine, TSH level 1 day before RRA, pre-Tg and post-Tg values. In addition, serum lipid levels detected 1 day before RRA were retrieved from the database to assess its relationship with TSH.
According to TSH level, the enrolled patients were categorized into two groups: 0–<30 and 30–<100 mU/L; and further subdivided into eight subgroups: 0–<30, 30–<40, 40–<50, 50–<60, 60–<70, 70–<80, 80–<90, 90–<100 mU/L. Thyroid remnant ablation success was determined once TSH suppressed Tg < 0.1 ng/mL 6 months after RRA.
Radioiodine treatment
Patients were asked to have a low-iodine diet for 2 weeks and withdraw LT4 for 3–4 weeks prior to RRA. Routine measurements such as free T3 (FT3), free T4 (FT4), TSH, pre-Tg, TgAb, neck ultrasonography, and chest CT scan were completed before RRA. We subsequently used an empirical active regimen to determine the prescribed activity of 131I. After RRA, patients were followed up for at least 12 months with measurement of FT3, FT4, TSH, Tg, and TgAb every 3–6 months.
Laboratory tests
A Cobas analyzer (Roche Diagnostics GmbH) was used to determine serum TSH, Tg, and TgAb using the same high-sensitivity electrochemiluminescence immunoassay method in the same laboratory of our hospital based on the manufacturer’s instructions. The exact reportable range of TSH and Tg is 0.005–100 mU/L and 0.04–25,000 ng/mL, respectively.
Statistical analysis
All statistical analyses were performed using the SPSS statistical software version 26.0 (SPSS) and GraphPad Prism Version 9.0 (GraphPad Software, Inc). We used several statistical methods to test the normality of continuous variables including histograms, skewness and kurtosis values, and the Kolmogorov–Smirnov/Shapiro–Wilk's test. Continuous results were expressed as mean ± s.d. if normally distributed or median and range if not. Categorical data were presented as numbers with percentages. Correlation between variables was evaluated by chi-square test, eta correlation coefficient, and Kendall's correlation coefficient. Univariate analysis was performed using chi-square test and Mann–Whitney U test. Any variable that showed a P value < 0.2 in the univariate analysis was then included in the multivariate regression analysis. A multivariate binary logistic regression was performed to assess the predictors of successful RRA and linear regression was performed to determine the influencing factors of TSH, which was regarded as a continuous variable. Receiver operating characteristic (ROC) curves were constructed to determine the cut-off value with maximum sensitivity and specificity for pre-Tg and pre-Tg/TSH ratio. Differences in the therapeutic effect of RRA and serum lipids in different TSH groups were evaluated using Mann–Whitney U test and Kruskal–Wallis test. The level of statistical significance was P < 0.05.
Results
Clinical characteristics of patients
The characteristics of all patients was shown in Table 1. A total of 487 patients aged 17–72 years were analyzed in the study, including 171 males and 316 females. Patients were divided into two groups: TSH < 30 (n = 44) and TSH ≥ 30 mU/L (n = 443) and further divided into eight groups: 0–<30 (n = 44), 30–<40 (n = 36), 40–<50 (n = 55), 50–<60 (n = 52), 60–<70 (n = 52), 70–<80 (n = 73), 80–<90 (n = 73), and 90–<100 mU/L (n = 102). Out of the whole enrolled patients, 367 had successful RRA, while 120 patients had unsuccessful RRA, with a success rate of 75.36%. The characteristics of patients with TSH < 30 mU/L are shown in Table 2. Of these 44 patients, 30 patients had successful RRA, while 14 had unsuccessful RRA, with a success rate of 68.18%.
Patients’ clinical characteristics and univariate analyses.
| Characteristics | Total (n = 487) | Successful RRA (n = 367) | Unsuccessful RRA (n = 120) | Univariate analyses | |
|---|---|---|---|---|---|
| χ2/z | P value | ||||
| Gender | 9.330 | 0.002a | |||
| Males | 171 (35.1) | 115 (31.3) | 56 (46.7) | ||
| Females | 316 (64.9) | 252 (65.7) | 64 (53.3) | ||
| Age (years) | 48 (17, 72) | 48 (21, 72) | 47 (17, 72) | −1.832 | 0.067 |
| Lesions condition | 1.709 | 0.191 | |||
| Solitary | 224 (46.0) | 175 (47.7) | 49 (40.8) | ||
| Multiple | 263 (54.0) | 192 (52.3) | 71 (59.2) | ||
| Maximum tumor diameter (cm) | 1.2 (0.2, 5.0) | 1.2 (0.2, 5.0) | 1.2 (0.2, 4.3) | −0.132 | 0.895 |
| Pathological types | 5.760 | 0.056 | |||
| Classical PTC | 470 (96.5) | 350 (95.4) | 120 (100.0) | ||
| Subtypes of PTC | 6 (1.2) | 6 (1.6) | 0 (0.0) | ||
| FTC | 11 (2.3) | 11 (3.0) | 0 (0.0) | ||
| T stage | 1.646 | 0.649 | |||
| T1 | 418 (85.8) | 316 (86.1) | 102 (85.0) | ||
| T2 | 46 (9.4) | 32 (8.7) | 14 (11.7) | ||
| T3 | 19 (3.9) | 16 (4.4) | 3 (2.5) | ||
| T4 | 4 (0.8) | 3 (0.8) | 1 (0.8) | ||
| N stage | 14.475 | 0.002a | |||
| N0 | 48 (9.9) | 43 (11.7) | 5 (4.2) | ||
| N1a | 245 (50.3) | 193 (52.6) | 52 (43.3) | ||
| N1b | 187 (38.3) | 125 (34.1) | 62 (51.7) | ||
| Nx | 7 (1.4) | 6 (1.6) | 1 (0.8) | ||
| Pre-Tg level (ng/mL) | 3.84 (0.13, 9.95) | 3.06 (0.13, 9,90) | 6.21 (1.02, 9.95) | −8.594 | <0.001a |
| TSH level (mU/L) | 70.30 (0.11, 99.93) | 70.20 (0.11, 99.93) | 73.01 (0.54, 99.73) | −0.439 | 0.661 |
| Activity (mCi) | 1.105 | 0.954 | |||
| 30 | 8 (1.6) | 7 (87.5) | 1 (0.8) | ||
| 50 | 25 (5.1) | 19 (76.0) | 6 (5.0) | ||
| 70 | 3 (0.6) | 2 (66.7) | 1 (0.8) | ||
| 80 | 1 (0.2) | 1 (100.0) | 0 (0.0) | ||
| 100 | 426 (87.5) | 320 (75.1) | 106 (88.3) | ||
| 150 | 24 (5.0) | 18 (75.0) | 6 (5.0) | ||
Values are presented as number (%) of patients or median (range).
FTC, follicular thyroid carcinoma; pre-Tg, pre-ablative thyroglobulin; PTC, papillary thyroid carcinoma; RRA, radioiodine (131I) remnant ablation; TSH, thyroid-stimulating hormone. aStatistically significant.
Patients’ clinical characteristics and univariate analyses in TSH < 30 mU/L group.
| Characteristics | Total (n = 44) | Successful RRA (n = 30) | Unsuccessful RRA (n = 14) | Univariate analyses | |
|---|---|---|---|---|---|
| χ2/z | P value | ||||
| Gender | 2.313 | 0.128 | |||
| Males | 15 (34.1) | 8 (26.7) | 7 (50.0) | ||
| Females | 29 (65.9) | 22 (73.3) | 7 (50.0) | ||
| Age (years) | 48 (24, 64) | 47 (24, 64) | 48 (25, 64) | −0.379 | 0.705 |
| Lesions condition | 0.043 | 0.837 | |||
| Solitary | 23 (52.3) | 16 (53.3) | 7 (50.0) | ||
| Multiple | 21 (47.7) | 14 (46.7) | 7 (50.0) | ||
| Maximum tumor diameter (cm) | 1.2 (0.2, 4.2) | 1.1 (0.2, 4.2) | 1.6 (0.4, 3.7) | −1.716 | 0.086 |
| Pathological types | 0.478 | 0.490 | |||
| Classical PTC | 43 (97.7) | 29 (96.7) | 14 (100.0) | ||
| Subtypes of PTC | 1 (2.3) | 1 (3.3) | 0 (0.0) | ||
| FTC | 0 (0.0) | 0 (0.0) | 0 (0.0) | ||
| T stage | 5.28 | 0.152 | |||
| T1 | 35 (79.5) | 26 (86.7) | 9 (64.3) | ||
| T2 | 7 (15.9) | 3 (10.0) | 4 (28.6) | ||
| T3 | 1 (2.3) | 1 (3.3) | 0 (0.0) | ||
| T4 | 1 (2.3) | 0 (0.0) | 1 (7.1) | ||
| N stage | 6.795 | 0.079 | |||
| N0 | 2 (4.5) | 1 (3.3) | 1 (7.1) | ||
| N1a | 21 (47.7) | 18 (60.0) | 3 (21.4) | ||
| N1b | 20 (45.5) | 10 (33.3) | 10 (21.4) | ||
| Nx | 1 (2.3) | 1 (3.3) | 0 (0.0) | ||
| Pre-Tg level (ng/mL) | 3.34 (1.17, 9.81) | 2.74 (1.24, 9.81) | 6.67 (1.17, 9.76) | −2.444 | 0.015 |
| TSH level (mU/L) | 12.30 (0.11, 29.83) | 13.99 (0.11, 29.83) | 11.68 (0.54, 28.97) | −0.454 | 0.650 |
| Activity (mCi) | 3.575 | 0.467 | |||
| 30 | 1 (2.3) | 0 (0.0) | 1 (7.1) | ||
| 50 | 4 (9.1) | 2 (6.7) | 2 (14.3) | ||
| 70 | 2 (4.5) | 1 (3.3) | 1 (7.1) | ||
| 100 | 28 (63.6) | 20 (66.7) | 8 (57.2) | ||
| 150 | 9 (20.5) | 7 (23.3) | 2 (14.3) | ||
Values are presented as number (%) or median (range).
FTC, follicular thyroid carcinoma; pre-Tg, pre-ablative thyroglobulin; PTC, papillary thyroid carcinoma; RRA, radioiodine (131I) remnant ablation; TSH, thyroid-stimulating hormone.
TSH level and radioiodine (131I) remnant ablation
The scatter plot with TSH as a continuous variable is shown in Fig. 1A. There was no statistically significant difference in TSH between the successful and unsuccessful RRA groups (Fig. 1A). The histograms dividing TSH as a categorical variable into two groups and eight subgroups are shown in Fig. 1B and 1C. Although 30/44 patients (68.2%) with TSH < 30 mU/L had a lower success rate than 337/443 patients (76.1%) with TSH ≥ 30 mU/L, the difference between the two groups was not statistically significant (P = 0.247). There was seemingly a first increased and then decreased trend of RRA success rates of eight subgroups with increasing TSH levels, which were 68.2, 77.8, 76.4, 76.9, 80.8, 80.8, 69.9, 73.5%, respectively; however, their differences were also not statistically significant (P = 0.685), as shown in Fig. 1C.
Percentages of patients with successful 131I remnant ablation in different groups according to TSH level. Scatter dot plot with TSH level as a continuous variable (A). Histograms with TSH level as a categorical variable, divided into TSH < 30 or ≥30 mU/L (B) and further divided into eight subgroups (C). ns, no significance; TSH, thyroid-stimulating hormone.
Citation: European Thyroid Journal 12, 4; 10.1530/ETJ-22-0219
TSH level and serum lipids
We collected the levels of nine clinical indicators commonly used to reflect serum lipid, and the results are shown in Fig. 2. Of them, total cholesterol (T-CHO) (P < 0.001), triglyceride (TG) (P = 0.006), high-density lipoprotein cholesterol (HDL-C) (P = 0.024), low-density lipoprotein cholesterol (LDL-C) (P = 0.001), apolipoprotein B (apoB) (P < 0.001), and apolipoprotein E (apoE) (P = 0.002) were significantly higher while apoA/apoB ratio (P = 0.024) was significantly lower in TSH ≥ 30 mU/L group when compared to TSH < 30 mU/L group.
Levels of T-CHO (A), TG (B), HDL-C (C), LDL-C (D), Lp(A) (E), apoA-I (F), apoB (G), apoE (H), and apoA/apoB ratio (I) were compared between TSH < 30 and TSH ≥ 30 mU/L groups. All the values represent the mean ± s.e.m. ***P< 0.001; **P< 0.01; *P< 0.05, ns, no significance; TSH, thyroid-stimulating hormone.
Citation: European Thyroid Journal 12, 4; 10.1530/ETJ-22-0219
Independent affecting factors of radioiodine (131I) remnant ablation
Analyses of factors associated with successful RRA of all patients are shown in Table 3. Univariate binary logistic regression showed that TSH level was not related to the success rate of RRA (P = 0.661), nor maximum tumor diameter (P = 0.895), T stage (P = 0.649), activity (P = 0.954), age (P = 0.067), lesions condition (P = 0.191), and pathological types (P = 0.056). While pre-Tg level (P < 0.001), gender (P = 0.002), and N stage (P = 0.002) were related to the success rate of RRA. Multivariate analyses identified that pre-Tg level (P < 0.001), gender (P = 0.009), and N stage (P = 0.044) were independent predictors for successful RRA.
Multivariate analyses of influencing factors of RRA.
| Multivariate analyses | ||||||
|---|---|---|---|---|---|---|
| B | s.e. | Wals | df | Sig. | Exp(B) | |
| Gender | −0.641 | 0.245 | 6.812 | 1 | 0.009a | 0.527 |
| Males | ||||||
| Females | ||||||
| Age (years) | −0.013 | 0.011 | 1.355 | 1 | 0.244 | 0.987 |
| Lesions condition | 0.417 | 0.246 | 2.868 | 1 | 0.090 | 1.518 |
| Solitary | ||||||
| Multiple | ||||||
| Maximum tumor diameter (cm) | ||||||
| Pathological types | 0.000 | 2 | 1.000 | |||
| Classical PTC | ||||||
| Subtypes of PTC | ||||||
| FTC | ||||||
| T stage | ||||||
| T1 | ||||||
| T2 | ||||||
| T3 | ||||||
| T4 | ||||||
| N stage | 8.112 | 3 | 0.044a | |||
| N0 | ||||||
| N1a | ||||||
| N1b | ||||||
| Nx | ||||||
| Pre-Tg level (ng/mL) | 0.387 | 0.049 | 63.331 | 1 | <0.001a | 1.472 |
| TSH level (mU/L) | ||||||
| Activity (mCi) | ||||||
| 30 | ||||||
| 50 | ||||||
| 70 | ||||||
| 80 | ||||||
| 100 | ||||||
| 150 | ||||||
aStatistically significant.
B, partial regression coefficient; Exp (B), the OR value of the corresponding variable; FTC, follicular thyroid carcinoma; pre-Tg, pre-ablative thyroglobulin; PTC, papillary thyroid carcinoma; RRA, radioiodine (131I) remnant ablation; TSH, thyroid-stimulating hormone; Wals, Wald statistics.
Analyses of factors associated with successful RRA of patients with TSH < 30 mU/L are shown in Table 4. Univariate binary logistic regression showed that TSH level was not related to the success rates of RRA (P = 0.650), nor age (P = 0.705), lesions condition (P = 0.837), pathological types (P = 0.490), activity (P = 0.467), gender (P = 0.128), maximum tumor diameter (P = 0.086), T stage (P = 0.152), and N stage (P = 0.079). While pre-Tg level (P = 0.015) was related to the success rates of RRA. Multivariate analyses showed that only pre-Tg level (P = 0.012) was identified as an independent predictor of successful RRA in patients with TSH < 30 mU/L.
Multivariate analyses of influencing factors of RRA in TSH < 30 mU/L group.
| B | s.e. | Wals | df | Sig. | Exp(B) | |
|---|---|---|---|---|---|---|
| Gender | −1.505 | 1.008 | 2.226 | 1 | 0.136 | 0.222 |
| Males | ||||||
| Females | ||||||
| Age (years) | ||||||
| Lesions condition | ||||||
| Solitary | ||||||
| Multiple | ||||||
| Maximum tumor diameter (cm) | 0.444 | 1.014 | 0.192 | 1 | 0.661 | 1.559 |
| Pathological types | ||||||
| Classical PTC | ||||||
| Subtypes of PTC | ||||||
| FTC | ||||||
| T stage | 0.996 | 3 | 0.802 | |||
| T1 | ||||||
| T2 | ||||||
| T3 | ||||||
| T4 | ||||||
| N stage | 2.535 | 2 | 0.281 | |||
| N0 | ||||||
| N1a | ||||||
| N1b | ||||||
| Nx | ||||||
| Pre-Tg level(ng/mL) | 0.486 | 0.193 | 6.358 | 1 | 0.012a | 1.626 |
| TSH level (mU/L) | ||||||
| Activity (mCi) | ||||||
| 30 | ||||||
| 50 | ||||||
| 70 | ||||||
| 80 | ||||||
| 100 | ||||||
| 150 |
aStatistically significant.
B, partial regression coefficient; Exp (B), the OR value of the corresponding variable; FTC: follicular thyroid carcinoma; pre-Tg, pre-ablative thyroglobulin; PTC: papillary thyroid carcinoma; RRA, radioiodine (131I) remnant ablation; s.e., standard error; TSH, thyroid-stimulating hormone; Wals, Wald statistics.
Pre-Tg and Pre-Tg/TSH ratio for predicting successful RRA
In the linear regression analysis, TSH level, entered as a continuous variable, was significantly related to pre-Tg (P = 0.002), age (P = 0.002), but not to gender (P = 0.518), maximum tumor diameter (P = 0.965), lesions condition (P = 0.559), T stage (P = 0.556), and N stage (P = 0.142).
Since the fact that TSH affects the value of Tg, pre-Tg/TSH ratio was used to correct the predictive value of pre-Tg. We compared the performance of pre-Tg/TSH ratio with pre-Tg in predicting the therapeutic effect of RRA. According to ROC curves of pre-Tg and pre-Tg/TSH ratio, the area under the curve (AUC) for pre-Tg level and pre-Tg/TSH ratio was 0.7611 (P < 0.0001) and 0.7340 (P < 0.0001), respectively, for all patients (Fig. 3A). The AUC for pre-Tg and pre-Tg/TSH ratio of patients with TSH < 30 mU/L was 0.7310 (P = 0.0145) and 0.6524 (P = 0.1068), respectively (Fig. 3B).
ROC curves analyses of pre-Tg and pre-Tg/TSH ratio to predict successful 131I remnant ablation in the whole 487 patients (A) and 44 patients with TSH < 30 mU/L (B). pre-Tg, pre-ablative TSH-stimulated Tg.
Citation: European Thyroid Journal 12, 4; 10.1530/ETJ-22-0219
The predictive performances of pre-Tg and pre-Tg/TSH ratio are illustrated in Table 5. In all patients, the cut-off value of pre-Tg to predict successful RRA was < 3.205 ng/mL with its sensitivity and specificity of 90.0 and 53.1%, while the cut-off value of pre-Tg/TSH ratio was < 0.0597 with its sensitivity and specificity of 80.8 and 58.3%, respectively. In the TSH < 30 mU/L group, the cut-off value of pre-Tg to predict successful RRA was < 3.335 ng/mL with its sensitivity and specificity of 85.70 and 66.70%, while the cut-off value of pre-Tg/TSH ratio was < 0.1480 with its sensitivity and specificity of 92.9 and 40.0%, respectively.
Cut-off valuesa for pre-Tg and pre-Tg/TSH ratio with sensitivity and specificity.
| Pre-Tg | Pre-Tg/TSH ratio | |
|---|---|---|
| All patients | ||
| Cut-off value | <3.205 | <0.0597 |
| Sensitivity | 90.0% | 80.8% |
| Specificity | 53.1% | 58.3% |
| Area under curve | 0.7611 | 0.7340 |
| P value | <0.0001 | <0.0001 |
| TSH < 30 mU/L | ||
| Cut-off value | <3.335 | <0.1480 |
| Sensitivity | 85.70% | 92.9% |
| Specificity | 66.70% | 40.0% |
| Area under curve | 0.7310 | 0.6524 |
| P value | 0.0145 | 0.1068 |
aValues with maximum sensitivity and specificity.
Pre-Tg, pre-ablative thyroglobulin; TSH, thyroid-stimulating hormone.
Discussion
In this study, we found that the success rates of RRA were not statistically different either between the two groups or the eight subgroups, thus presumed that TSH levels might not be necessary for successful RRA. Our results showed that patients in the TSH ≥ 30 mU/L group generally had higher serum lipid levels than those in the TSH < 30 mU/L. In addition, we analyzed the influencing factors of successful RRA and proposed corresponding predictors, especially in patients with TSH < 30 mU/L for the first time. And the results demonstrated that pre-Tg could be an independent predictor for successful RRA not only in the whole group but also in the TSH < 30 mU/L group. Furthermore, we applied ROC curves to examine the predictive performance of pre-Tg and compared it with its TSH-corrected value (pre-Tg/TSH). We found that pre-Tg showed a satisfied predictive performance when compared to the pre-Tg/TSH ratio in patients with TSH < 30 mU/L.
Residual thyroid tissue after total thyroidectomy in patients with DTCs could be an independent risk factor for recurrence and radioiodine therapy is the cornerstone of treatment to remove residual thyroid tissue. After all thyroid tissue is eliminated, using Tg as a tumor marker during follow-up will have a higher sensitivity and specificity because typically Tg is only produced by normal and cancerous thyroid follicular cells. Prior to RRA, LT4 replacement must be discontinued to increase the TSH level and the recommended level is ≥ 30 mU/L. Although this threshold came from a report more than 40 years ago and has not been proven conclusively, it is still cited in many guidelines. However, we did not find any significant difference in RRA success rates between TSH < 30 and TSH ≥ 30 mU/L groups and between eight subgroups in our study. This indicates that the successful rate of RRA might not be limited by low TSH levels.
The serum lipid level analysis confirmed that patients could suffer lipid metabolic disorders during the period of LT4 withdrawal prior to RRA. Our results are in good agreement with other studies, which demonstrate that TSH levels are associated with hyperlipidemia (10, 12). It also has been reported that high TSH levels could be associated with impaired endothelial function, in hypothyroidism and elevated TSH could promote endothelial dysfunction by altering gene expression in human umbilical vein endothelial cells (13, 14). Hence, we believed that the advantages of prompt RRA in patients with TSH < 30 mU/L may be more important than longer withdrawal time.
The factors that may influence the therapeutic outcome of RRA were analyzed among 487 patients enrolled. We found that pre-Tg, gender, and N stage were the major determinants while age, lesions condition, maximum tumor diameter, T stage, and TSH were not. Thyroglobulin is only synthesized in normal thyroid follicular cells or DTC cells, so pre-Tg level is an important parameter for monitoring the recurrence or metastasis of thyroid carcinoma (15, 16). Many studies reported that a high pre-Tg level was associated with a low success rate of RRA and had predictive value, which is consistent with our study (17, 18, 19, 20). The data in our study indicated that female was more likely to result in successful RRA than male. This is inconsistent with the results of most published studies, which have generally concluded that gender is not an influencing factor. This discrepancy may be related to differences in the distribution of enrolled patients. In this study, patients with the lower N stage were more likely to have successful RRA. This has been verified in other studies (11). We further analyzed the influencing factors of successful RRA in patients with TSH < 30 mU/L, which is conducted for the first time to the best of our knowledge. Interestingly, we found that TSH level was not an influencing factor either in all 487 patients or in TSH < 30 mU/L group, which further supports our previous conclusion that high TSH level may not be necessary for successful RRA.
Based on the earlier results, we hypothesized that pre-Tg level could be a predictor and used ROC curves to validate its predictive performance. It revealed that pre-Tg showed satisfied predictive performance for RRA success. Considering that elevated TSH is the most important stimulus of Tg, we used its corrected value pre-Tg/TSH ratio and evaluated its predictive performance, especially in patients with TSH < 30 mU/L for the first time. However, we found that it might not be appropriate to use pre-Tg/TSH ratio to predict RRA success specifically in patients with TSH < 30 mU/L. It may be due to the minimal effect of TSH on pre-Tg when TSH < 30 mU/L. Therefore, there is no need to use TSH to correct pre-Tg in this condition.
The results of this study have important implications for clinical practice. Because our results support that DTC patients with LT4 withdrawal after thyroidectomy can receive RRA even if TSH < 30 mU/L. We believe that the high TSH level prior to RRA has probably been exaggerated in recent decades. Hence, many patients have possibly been submitted to a period of hypothyroidism, resulting in a temporary loss of quality of life (21), such as suffering hyperlipidemia (22). Our findings may support to shorten the duration and/or activity of LT4 withdrawal prior to RRA, which could reduce or avoid adverse effects from high TSH levels. We identified some major influencing factors of RRA, which should be taken into consideration in determining individualized treatment for optimizing the efficacy of radioiodine for thyroid remnant ablation. Also, the use of pre-Tg rather than pre-Tg/TSH ratio for predicting the therapeutic outcome of RRA has promising applications, especially when patients with TSH < 30 mU/L. It can be used as a predictor for successful RRA to guide the treatment plan.
Several limitations of our study should be mentioned. Firstly, the study was retrospective in design, which may have caused deviations in data selection. Secondly, the upper limit of TSH in our study was 100 mU/L because the exact value cannot be determined when TSH ≥ 100 mU/L in our institute. Thirdly, we did not analyze patients with Tg ≥10 ng/mL because patients with Tg ≥10 ng/mL may show more risk of potential metastatic lesions and affect the accuracy to use the post-Tg level as an indicator for ablation success. Finally, we only conducted the study among patients with LT4 withdrawal rather than using recombinant human TSH (rh-TSH) because the use of rh-TSH has not been approved in China yet. Therefore, further research is still needed before adopting this strategy.
Conclusion
The results of this study found no statistical difference in RRA success rates in DTC patients with TSH <30 and TSH ≥ 30 mU/L stimulation. Acceptable therapeutic effect could be achieved even with TSH < 30 mU/L. Patients with higher TSH level will suffer from severer hyperlipidemia. Pre-Tg level could be used as a predictor for RRA success, especially when TSH < 30 mU/L prior radioiodine therapy. The minimum level of TSH stimulation for radioiodine thyroid remnant ablation therapy for DTC patients after total thyroidectomy may need further investigation.
Declaration of interest
The authors have no conflicts of interest to declare.
Funding
This study was supported by the Shanghai Municipal Health Commission scientific research project (No.20204Y0256) and partially supported by the Shanghai Sixth People’s Hospital scientific research project (No.YNLC201903), the National Natural Science Foundation of China (No.81901773) and the Shanghai Key Clinical Specialty of Medical Imaging (No.shslczdzk03203).
Statement of ethics
This retrospective study was approved by the ethical review board of the Shanghai Sixth People’s Hospital (No.2020-096). Written informed consent to publish their case and images was obtained from every patient before radioiodine therapy.
Author contribution statement
Every person listed as authors made significant contribution to the writing and revision and approved the final version for submission for publication.
Acknowledgement
The authors thank Prof L-N Zhang (Department of Biostatistics, Clinical Research Institute, Shanghai Jiao Tong University School of Medicine) for her valuable suggestions about the statistical method of our study.
References
- 1↑
Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, & Bray F. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians 2021 71 209–249. (https://doi.org/10.3322/caac.21660)
- 2↑
Xing M, Haugen BR, & Schlumberger M. Progress in molecular-based management of differentiated thyroid cancer. Lancet 2013 381 1058–1069. (https://doi.org/10.1016/S0140-6736(1360109-9)
- 3↑
Ciarallo A, & Rivera J. Radioactive iodine therapy in differentiated thyroid cancer: 2020 update. AJR. American Journal of Roentgenology 2020 215 285–291. (https://doi.org/10.2214/AJR.19.22626)
- 4↑
Reiners C, Hanscheid H, Luster M, Lassmann M, & Verburg FA. Radioiodine for remnant ablation and therapy of metastatic disease. Nature Reviews. Endocrinology 2011 7 589–595. (https://doi.org/10.1038/nrendo.2011.134)
- 5↑
Edmonds CJ, Hayes S, Kermode JC, & Thompson BD. Measurement of serum TSH and thyroid hormones in the management of treatment of thyroid carcinoma with radioiodine. British Journal of Radiology 1977 50 799–807. (https://doi.org/10.1259/0007-1285-50-599-799)
- 6↑
Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, Pacini F, Randolph GW, Sawka AM, Schlumberger M, et al.2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 2016 26 1–133. (https://doi.org/10.1089/thy.2015.0020)
- 7↑
Filetti S, Durante C, Hartl D, Leboulleux S, Locati LD, Newbold K, Papotti MG, & Berruti A. Thyroid cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-updagger. Annals of Oncology 2019 30 1856–1883. (https://doi.org/10.1093/annonc/mdz400)
- 8↑
Pacini F, Fuhrer D, Elisei R, Handkiewicz-Junak D, Leboulleux S, Luster M, Schlumberger M, & Smit JW. 2022 ETA Consensus Statement: what are the indications for post-surgical radioiodine therapy in differentiated thyroid cancer? European Thyroid Journal 2022 11. (https://doi.org/10.1530/ETJ-21-0046)
- 9↑
Su X, Peng H, Chen X, Wu X, & Wang B. Hyperlipidemia and hypothyroidism. Clinica Chimica Acta; International Journal of Clinical Chemistry 2022 527 61–70. (https://doi.org/10.1016/j.cca.2022.01.006)
- 10↑
van Vliet NA, Bos MM, Thesing CS, Chaker L, Pietzner M, Houtman E, Neville MJ, Li-Gao R, Trompet S, Mustafa R, et al.Higher thyrotropin leads to unfavorable lipid profile and somewhat higher cardiovascular disease risk: evidence from multi-cohort Mendelian randomization and metabolomic profiling. BMC Medicine 2021 19 266. (https://doi.org/10.1186/s12916-021-02130-1)
- 11↑
Wang C, Diao H, Ren P, Wang X, Wang Y, & Zhao W. Efficacy and affecting factors of (131)I thyroid remnant ablation after surgical treatment of differentiated thyroid carcinoma. Frontiers in Oncology 2018 8 640. (https://doi.org/10.3389/fonc.2018.00640)
- 12↑
Wang JJ, Zhuang ZH, Shao CL, Yu CQ, Wang WY, Zhang K, Meng XB, Gao J, Tian J, Zheng JL, et al.Assessment of causal association between thyroid function and lipid metabolism: a Mendelian randomization study. Chinese Medical Journal 2021 134 1064–1069. (https://doi.org/10.1097/CM9.0000000000001505)
- 13↑
Gazdag A, Nagy EV, Burman KD, Paragh G, & Jenei Z. Improved endothelial function and lipid profile compensate for impaired hemostatic and inflammatory status in iatrogenic chronic subclinical hyperthyroidism of thyroid cancer patients on L-t4 therapy. Experimental and Clinical Endocrinology and Diabetes 2010 118 381–387. (https://doi.org/10.1055/s-0029-1224156)
- 14↑
Tian L, Zhang L, Liu J, Guo T, Gao C, & Ni J. Effects of TSH on the function of human umbilical vein endothelial cells. Journal of Molecular Endocrinology 2014 52 215–222. (https://doi.org/10.1530/JME-13-0119)
- 15↑
Evans C, Tennant S, & Perros P. Thyroglobulin in differentiated thyroid cancer. Clinica Chimica Acta; International Journal of Clinical Chemistry 2015 444 310–317. (https://doi.org/10.1016/j.cca.2014.10.035)
- 16↑
Li S, Ren C, Gong Y, Ye F, Tang Y, Xu J, Guo C, & Huang J. The role of thyroglobulin in preoperative and postoperative evaluation of patients with differentiated thyroid cancer. Frontiers in Endocrinology (Lausanne) 2022 13 872527. (https://doi.org/10.3389/fendo.2022.872527)
- 17↑
Ha S, Oh SW, Kim YK, Koo do H, Jung YH, Yi KH, & Chung JK. Clinical outcome of remnant thyroid ablation with low dose radioiodine in Korean patients with low to intermediate-risk thyroid cancer. Journal of Korean Medical Science 2015 30 876–881. (https://doi.org/10.3346/jkms.2015.30.7.876)
- 18↑
Trevizam PG, Tagliarini JV, Castilho EC, de Alencar Marques M, Kiy Y, & Mazeto GMFS. Thyroglobulin levels and thyroglobulin/thyrotropin ratio could predict the success of the ablative/therapeutic (131)I in the differentiated thyroid cancers. Endocrine Research 2017 42 42–48. (https://doi.org/10.3109/07435800.2016.1173056)
- 19↑
Husseini MA. Implication of different clinical and pathological variables in patients with differentiated thyroid cancer on successful ablation for 3700 MBq (131)I: a single Egyptian institutional experience over 14 years. Annals of Nuclear Medicine 2016 30 468–476. (https://doi.org/10.1007/s12149-016-1084-9)
- 20↑
Lubin DJ, Tsetse C, Khorasani MS, Allahyari M, & McGrath M. Clinical predictors of I-131 therapy failure in differentiated thyroid cancer by machine learning: a single-center experience. World Journal of Nuclear Medicine 2021 20 253–259. (https://doi.org/10.4103/wjnm.WJNM_104_20)
- 21↑
Badihian S, Jalalpour P, Mirdamadi M, & Moslehi M. Quality of life, anxiety and depression in patients with differentiated thyroid cancer under short term hypothyroidism induced by levothyroxine withdrawal. Klinicka Onkologie: Casopis Ceske a Slovenske Onkologicke Spolecnosti 2016 29 439–444. (https://doi.org/10.14735/amko2016439)
- 22↑
Papadakis G, Kalaitzidou S, Triantafillou E, Drosou A, Kakava K, Dogkas N, Pappa T, Kaltzidou V, Tertipi A, Villiotou V, et al.Biochemical effects of levothyroxine withdrawal in patients with differentiated thyroid cancer. Anticancer Research 2015 35 6933–6940. (https://doi.org/10.1530/endoabs.41.EP1106)
