Abstract
Objectives: This study aimed to assess whether low-iodine diet (LID) adherence is associated with therapeutic response in papillary thyroid carcinoma (PTC), specifically in relation to post-therapeutic thyroglobulin (Tg) release as a surrogate marker for the acute radiation-induced response following radioactive iodine (RAI) therapy. Methods: This retrospective study included 895 patients with PTC treated with RAI. LID adherence was assessed using the urine iodine-to-creatinine (I/Cr) ratio, with <66.2 μg/g Cr defined as good adherence. The Tg ratio (ratioTg), calculated by dividing post-RAI Tg (measured 7 days after RAI) by pre-RAI Tg, was used to reflect the magnitude of the radiation-induced Tg release. Patients were stratified by ratioTg (≤1 vs. >1), and associations between LID adherence and therapeutic response were analyzed within each group. Results: Well-adherent patients exhibited significantly higher ratioTg compared to poorly adherent patients (15.7 ± 2.2 vs. 8.9 ± 1.3, p = 0.007). Among patients with ratioTg > 1 (n = 630), LID adherence was independently associated with improved therapeutic response (OR, 2.004; 95% CI, 1.270–3.162; p = 0.003). No such association was observed in patients with ratioTg ≤ 1 (n = 265; p = 0.546). Conclusions: The clinical benefit of LID appears to depend on the presence of a certain magnitude of radiation-induced Tg release. RatioTg may serve as a useful marker for identifying patients likely to benefit from LID.
Keywords: papillary thyroid carcinoma, low-iodine diet, radioactive iodine, thyroglobulin
1. Introduction
Differentiated thyroid carcinoma (DTC), including papillary and follicular subtypes, is one of the most prevalent endocrine cancers, with its incidence steadily increasing worldwide [1]. The prognosis of DTC is generally favorable, reflecting both its indolent biology and the effectiveness of standard treatments, including surgery, radioactive iodine (RAI) therapy, and thyroid-stimulating hormone (TSH) suppression, in appropriately selected patients [2].
RAI therapy plays a pivotal role in the management of DTC by exploiting the thyroid follicular cells’ unique ability to uptake and retain iodine [3,4]. This targeted approach has been shown to reduce the risk of recurrence in selected patients, particularly those with intermediate- or high-risk disease [5,6]. However, the therapeutic efficacy of RAI is influenced not only by the administration protocol, but also by individual patient factors, including iodine status and tumor biology [7].
A low-iodine diet (LID) is typically recommended for 1–2 weeks prior to RAI therapy to reduce the body’s iodine pool and enhance RAI uptake by thyroid cells [3,8]. Despite its routine use, the clinical impact of LID adherence on treatment outcomes remains controversial [9]. Some studies have shown that strict adherence to LID improves the therapeutic response, whereas others have found no significant benefit, highlighting the need to determine the specific biological conditions under which LID exerts a meaningful effect.
Serum thyroglobulin (Tg), a glycoprotein produced by normal and neoplastic thyroid follicular cells, is a well-established biomarker for monitoring disease burden and response to therapy in DTC [10,11,12]. Serum Tg levels measured before RAI therapy primarily reflect the burden of residual thyroid tissue or metastatic disease, whereas it is hypothesized that a transient rise in Tg levels after RAI administration may serve as a surrogate marker reflecting the acute biological response of thyroid tissue to radiation [13].
LID adherence may enhance treatment response under specific conditions, such as when a certain degree of acute radiation-induced response occurs. In this study, we investigated whether LID adherence is associated with therapeutic benefit in such cases, using post-RAI Tg elevation as a surrogate marker of radiation-induced Tg release in patients with papillary thyroid carcinoma (PTC).
2. Materials and Methods
2.1. Patients
We initially included 1102 patients with PTC who underwent total thyroidectomy followed by the first RAI therapy between 2014 and 2019. Patients were aged ≥ 18 years and had no gross residual disease or distant metastasis before RAI therapy. All patients were instructed to follow a standardized LID for two weeks prior to RAI therapy, which included avoidance of iodine-rich foods such as seaweed, seafood, dairy products, and iodized salt. To elevate TSH, recombinant human TSH (rhTSH) or thyroid hormone withdrawal (THW) was applied before RAI therapy. The exclusion criteria were as follows: the presence of other primary malignancies at the time of diagnosis (n = 27); a time interval between surgery and RAI therapy of less than two weeks or more than 12 months (n = 2); serum anti-Tg antibody (TgAb) levels greater than 100 U/mL (n = 82); loss to follow-up (n = 16); and RAI dose of 2.96 GBq (n = 80). In total, 895 PTC patients were retrospectively enrolled in this study. This retrospective study was approved by the institutional review board of our hospital in accordance with the Declaration of Helsinki (Approval No. CNUHH-2021-043), and the informed consent was waived.
2.2. Measurement of Clinical Parameters at the Time of RAI Therapy or Follow-Up
Serum Tg levels were measured twice: immediately before RAI therapy (pre-RAI Tg) and 7 days after RAI therapy (post-RAI Tg). TgAb and TSH levels were also evaluated prior to RAI therapy. Tg, TgAb, and TSH levels were measured using an immunoradiometric assay (IRMA) with a lower detection limit of <1 ng/mL (RIA Tg-plus, BRAHMS GmbH, Hennigsdorf, Germany), a radioimmunoassay (RIA anti-Tgn, BRAHMS GmbH, Hennigsdorf, Germany), and an IRMA with a lower detection limit of 0.07 µIU/mL (TSH-CTK-3, DiaSorin, Saluggia, Italy), respectively.
Urine iodine and creatinine levels were measured immediately before RAI therapy. Urine iodine concentration was quantified using the AutoLab Iodine assay (IVD-LAB, Hwaseong-si, Gyeonggi, Republic of Korea) on a clinical chemistry analyzer (AU5822, Beckman Coulter Inc., Brea, CA, USA) utilizing the Sandell–Kolthoff reaction. Urine creatinine levels were determined on the same AU5822 analyzer using the CREATININE assay (Beckman Coulter Inc., Brea, CA, USA), based on the compensated kinetic Jaffe method, traceable to the isotope dilution mass spectrometry (IDMS) method. The urine iodine-to-creatinine (I/Cr) ratio (μg/g Cr) was then calculated.
Response assessments using serum Tg and neck ultrasound (US) were performed between 6 and 24 months after the initial RAI therapy [3]. Serum Tg was measured via electrochemiluminescent immunoassay (ECLIA) (Elecsys Tg II, Roche Diagnostics, Penzberg, Germany) using the Cobas e801 module (Roche Diagnostics, Mannheim, Germany). Additionally, TgAb (Elecsys Anti-Tg, Roche Diagnostics, Penzberg, Germany) and TSH (Elecsys TSH, Roche Diagnostics, Penzberg, Germany) levels were measured using the same e801 module on the same day as the Tg measurement.
2.3. Study Design
To investigate whether LID adherence is associated with therapeutic benefit, particularly when a certain degree of acute radiation-induced response occurred, we evaluated ratioTg changes according to LID status, its validity as a biomarker, and the resulting conditional efficacy of LID.
We examined whether LID adherence could influence the extent of RAI-induced Tg release. LID adherence was assessed using the urine I/Cr ratio, with a cut-off value of 66.2 μg/g Cr, as established in a previous study conducted in Korea [14]. Patients with a urine I/Cr ratio < 66.2 μg/g Cr were classified as well-adherent, whereas those with a urine I/Cr ratio ≥ 66.2 μg/g Cr were classified as poorly adherent, serving as an operational definition of dietary compliance in this study. The degree of RAI-induced Tg release was indirectly evaluated using the change in serum Tg levels before and after RAI therapy [15,16], expressed as the ratio of post-RAI to pre-RAI Tg levels (ratioTg). We compared ratioTg values between well-adherent and poorly adherent groups to determine whether LID was associated with greater thyroid cell response.
Next, we investigated whether the association between LID adherence and therapeutic response varied depending on the extent of Tg release. Therapeutic response was assessed based on the 2015 American Thyroid Association (ATA) guidelines: (1) excellent response (ER), defined as negative imaging and suppressed Tg < 0.2 ng/mL or TSH-stimulated Tg < 1 ng/mL; (2) indeterminate response (IR), characterized by non-specific findings on imaging studies or suppressed Tg that is detectable but < 1 ng/mL, or stimulated Tg between 1 and 10 ng/mL; (3) biochemical incomplete response (BIR), indicated by negative imaging with suppressed Tg ≥ 1 ng/mL, TSH-stimulated Tg ≥ 10 ng/mL, or increased TgAb values; and (4) structural incomplete response (SIR), identified as structural or functional evidence of disease regardless of Tg level [3]. Patients were categorized into the ER and the non-ER (nER) groups, with the nER group including patients with IR, BIR, and SIR. Patients were stratified into two groups based on a ratioTg threshold of 1.0, under the assumption that a ratioTg > 1 indicates a certain degree of RAI-induced tissue response. Within each group, the association between LID adherence and therapeutic response (ER vs. nER) was then evaluated.
Lastly, we evaluated whether the association between ratioTg and therapeutic response differed depending on LID adherence. This aimed to explore whether ratioTg functions as a biologically meaningful parameter that links LID adherence to therapeutic outcomes, thereby supporting its role as a surrogate marker of RAI-induced thyroid tissue response.
2.4. Statistical Analysis
Quantitative data were presented as mean ± standard deviation (SD), or standard error of the mean (SEM), while qualitative data were presented as percentages. Differences in variables were evaluated using t-tests for continuous variables and chi-square tests for categorical variables. Multivariate logistic regression analysis was performed to identify factors predicting ER and nER. Variables for multivariate logistic regression were selected based on a combination of univariate analysis results and established clinical relevance. p-values less than 0.05 were considered statistically significant. All statistical analyses were conducted using IBM SPSS Statistics for Windows, version 29.0.2.0 (IBM Corp., Armonk, NY, USA).
3. Results
3.1. Patient Characteristics
Table 1 summarizes the baseline characteristics of the 895 patients included in this study. The mean age was 49.1 ± 12.5 years (range, 18–83 years), and 74.2% were female. All patients had TSH levels greater than 30 µIU/mL. Tumor staging revealed that most patients were classified as T1 (65.2%) or T3 (20.0%), followed by T4 (8.5%) and T2 (6.3%). Lymph node status was categorized as N0a/N0b (20.8%), N1a (55.0%), and N1b (24.2%). The mean diameter of the largest tumor was 1.3 ± 0.9 cm (range, 0.1–7.0 cm). Tumor multiplicity was evenly distributed, with 45.5% of patients having solitary tumors and 54.5% having multiple tumors. Gross extrathyroidal extension was present in 27.4% of cases. The mean number of metastatic lymph nodes per patient was 4.3 ± 5.2. The mean pre-RAI Tg level was 2.6 ± 0.4 ng/mL (range, 0.1–308.0 ng/mL), and the mean value of ratioTg was 13.4 ± 1.5 (range, 0.007–1004.0). Most patients (81.1%) received rhTSH for stimulation, while 18.9% underwent THW.
Table 1.
Patient characteristics (n = 895).
| Variables | Values |
|---|---|
| Age (years) | |
| Mean ± SD (range) | 49.1 ± 12.5 (18–83) |
| Sex | |
| Male | 231 (25.8%) |
| Female | 664 (74.2%) |
| T stage | |
| T1 | 584 (65.2%) |
| T2 | 56 (6.3%) |
| T3 | 179 (20.0%) |
| T4 | 76 (8.5%) |
| N stage | |
| N0a or N0b | 186 (20.8%) |
| N1a | 492 (55.0%) |
| N1b | 217 (24.2%) |
| Diameter of the largest tumor (cm) | |
| Mean ± SD (range) | 1.3 ± 0.9 (0.1–7.0) |
| Multiplicity | |
| Solitary | 407 (45.5%) |
| Multiple | 488 (54.5%) |
| Gross extrathyroidal extension | |
| No | 650 (72.6%) |
| Yes | 245 (27.4%) |
| Number of metastatic lymph nodes | |
| Mean ± SD (range) | 4.3 ± 5.2 (0–36) |
| RAI dose | |
| Low does (≤1.85 GBq) | 452 (50.5%) |
| High does (≥3.70 GBq) | 443 (49.5%) |
| Pre-RAI Tg (ng/mL) | |
| Mean ± SEM (range) | 2.6 ± 0.4 (0.1–308.0) |
| RatioTg | |
| Mean ± SEM (range) | 13.4 ± 1.5 (0.007–1004.0) |
| Preparation method | |
| THW | 169 (18.9%) |
| rhTSH | 726 (81.1%) |
| Urine I/Cr ratio (μg/g Cr) | |
| <66.2 | 592 (66.1%) |
| ≥66.2 | 303 (33.9%) |
| Response assessment | |
| ER | 677 (75.6%) |
| nER | 218 (24.4%) |
SD, standard deviation; RAI, radioactive iodine; Tg, thyroglobulin; SEM, standard error of the mean; THW, thyroid hormone withdrawal; rhTSH, recombinant human thyrotropin; ER, excellent response; nER, non-excellent response.
LID adherence was assessed using the urine I/Cr ratio; 66.1% of patients (n = 592) had a ratio < 66.2 μg/g Cr, indicating good adherence, whereas 33.9% (n = 303) had a ratio ≥ 66.2 μg/g Cr, indicating poor adherence.
3.2. Comparison of ratioTg Values According to Low Iodine Diet Adherence
The serum pre-RAI Tg level, reflecting residual functioning thyroid tissue after surgery, did not significantly differ between the well-adherent (2.8 ± 0.6 ng/mL) and poorly adherent LID groups (2.2 ± 0.5 ng/mL; p = 0.754). To evaluate the impact of LID adherence on RAI-induced thyroid tissue response, we compared ratioTg values between groups. As shown in Figure 1, the well-adherent group had a significantly higher ratioTg (15.7 ± 2.2) than the poorly adherent group (8.9 ± 1.3; p = 0.007). There was no significant difference in RAI dose between the groups (p = 0.890), allowing us to exclude the effect of RAI dose as a confounding factor in the observed difference in thyroid tissue response.
Figure 1.
Comparison of ratioTg values according to low iodine diet (LID) adherence. The well-adherent LID group exhibited significantly higher ratioTg levels (15.7 ± 2.2) compared to the poorly adherent LID group (8.9 ± 1.3). The graph represents mean ± standard error of the mean.
3.3. Factors Associated with RAI Response in All Patients
Among the enrolled patients, 677 (75.6%) were classified into the ER group and 218 (24.4%) into the nER group (Table 1). We analyzed the factors associated with RAI response in the entire patient group (Table 2). In the multivariate logistic regression analysis, female sex (odds ratio [OR], 1.710; 95% confidence interval [CI], 1.150–2.544; p = 0.008), smaller tumor size (OR, 1.766; 95% CI, 1.192–2.617; p = 0.005), solitary tumor (OR, 1.653; 95% CI, 1.134–2.408; p = 0.009), number of metastatic lymph nodes (OR, 1.658; 95% CI, 1.035–2.659; p = 0.036), and pre-RAI Tg levels (OR, 1.588; 95% CI, 1.422–1.774; p < 0.001) were significantly associated with achieving ER. However, ratioTg (p = 0.200) and LID adherence (p = 0.177) were not significant predictors in the overall group.
Table 2.
Factors associated with therapeutic response in the overall study population.
| Variables | ER (n = 677) | nER (n = 218) | Univariate | Multivariate | |
|---|---|---|---|---|---|
| p Value | p Value | Odds Ratio (95% CI) | |||
| Age (years) | 0.971 | ||||
| <55 years | 445 (75.7%) | 143 (24.3%) | |||
| ≥55 years | 232 (75.6%) | 75 (24.4%) | |||
| Sex | < 0.001 | 0.008 | 1.710 (1.150–2.544) | ||
| Male | 150 (64.9%) | 81 (35.1%) | |||
| Female | 527 (79.4%) | 137 (20.6%) | |||
| T stage | 0.004 | 0.992 | 1.009 (0.175–5.832) | ||
| T1 + T2 | 501 (78.3%) | 139 (21.7%) | |||
| T3 + T4 | 176 (69.0%) | 79 (31.0%) | |||
| N stage | <0.001 | 0.238 | 1.235 (0.870–1.755) | ||
| N0a, N0b | 163 (87.6%) | 23 (12.4%) | |||
| N1a | 385 (78.3%) | 107 (21.7%) | |||
| N1b | 129 (59.4%) | 88 (40.6%) | |||
| Tumor size (cm) | < 0.001 | 0.005 | 1.766 (1.192–2.617) | ||
| ≤1.0 | 334 (81.9%) | 74 (18.1%) | |||
| >1.0 | 343 (70.4%) | 144 (29.6%) | |||
| Multiplicity | < 0.001 | 0.009 | 1.653 (1.134–2.408) | ||
| Solitary | 334 (82.1%) | 73 (17.9%) | |||
| Multiple | 343 (70.3%) | 145 (29.7%) | |||
| Gross extrathyroidal extension | 0.004 | 0.671 | 1.469 (0.250–8.648) | ||
| No | 508 (78.1%) | 142 (21.9%) | |||
| Yes | 169 (69.0%) | 76 (31.0%) | |||
| Number of metastatic lymph nodes | < 0.001 | 0.036 | 1.658 (1.035–2.659) | ||
| ≤5 | 536 (82.3%) | 115 (17.7%) | |||
| >5 | 141 (57.8%) | 103 (42.2%) | |||
| RAI dose (GBq) | 0.001 | 0.842 | 1.040 (0.706–1.532) | ||
| Low dose (≤1.85 GBq) | 364 (80.5%) | 88 (19.5%) | |||
| High dose (≥3.7 GBq) | 313 (70.7%) | 130 (29.3%) | |||
| Pre-RAI Tg (ng/mL) | < 0.001 | < 0.001 | 1.588 (1.422–1.774) | ||
| Mean ± SEM | 0.8 ± 0.1 | 8.3 ± 1.7 | |||
| RatioTg | < 0.001 | 0.200 | 0.994 (0.985–1.003) | ||
| Mean ± SEM | 15.5 ± 1.9 | 7.1 ± 1.3 | |||
| Urine I/Cr ratio (μg/g Cr) | 0.177 | ||||
| <66.2 | 456 (77.0%) | 136 (23.0%) | |||
| ≥66.2 | 221 (72.9%) | 82 (27.1%) | |||
| Preparation | 0.407 | ||||
| THW | 132 (78.1%) | 37 (21.9%) | |||
| rhTSH | 545 (75.1%) | 181 (24.9%) | |||
ER, excellent response; nER, non-excellent response; RAI, radioactive iodine; Tg, thyroglobulin; SEM, standard error of the mean; THW, thyroid hormone withdrawal; rhTSH, recombinant human thyrotropin.
3.4. Association Between LID Adherence and Therapeutic Response According to ratioTg
Based on the hypothesis that the therapeutic benefit of LID may emerge when a certain degree of RAI-induced thyroid tissue response has occurred, patients were stratified according to ratioTg (≤1 or >1). Within each group, the association between LID adherence and therapeutic response was assessed. In patients with a ratioTg > 1 (n = 630), LID adherence was significantly associated with therapeutic response in the multivariate logistic regression analysis (OR, 2.004; 95% CI, 1.270–3.162; p = 0.003) (Table 3). Tumor size (OR, 2.041; 95% CI, 1.278–3.261; p = 0.003), solitary tumor (OR, 1.740; 95% CI, 1.125–2.691; p = 0.013), and pre-RAI Tg level (OR, 1.705; 95% CI, 1.474–1.973; p < 0.001) were also independently associated with therapeutic response in this subgroup. However, in patients with a ratioTg ≤1 (n = 265), no significant association was observed between LID adherence and therapeutic response (p = 0.546).
Table 3.
Factors associated with therapeutic response in patients with a ratioTg > 1.
| Variables | ER (n = 474) | nER (n = 156) | Univariate | Multivariate | |
|---|---|---|---|---|---|
| p Value | p Value | Odds Ratio (95% CI) | |||
| Age (years) | 0.590 | ||||
| <55 years | 321 (75.9%) | 102 (24.1%) | |||
| ≥55 years | 153 (73.9%) | 54 (26.1%) | |||
| Sex | 0.003 | 0.068 | 1.558 (0.967–2.509) | ||
| Male | 117 (66.9%) | 58 (33.1%) | |||
| Female | 357 (78.5%) | 98 (21.5%) | |||
| T stage | 0.014 | 0.990 | 0.989 (0.172–5.675) | ||
| T1 + T2 | 347 (78.0%) | 98 (22.0%) | |||
| T3 + T4 | 127 (68.6%) | 58 (31.4%) | |||
| N stage | < 0.001 | 0.459 | 1.168 (0.774–1.762) | ||
| N0a, N0b | 109 (85.8%) | 18 (14.2%) | |||
| N1a | 276 (77.5%) | 80 (22.5%) | |||
| N1b | 89 (60.5%) | 58 (39.5%) | |||
| Tumor size (cm) | < 0.001 | 0.003 | 2.041 (1.278–3.261) | ||
| ≤1.0 | 228 (83.5%) | 45 (16.5%) | |||
| >1.0 | 246 (68.9%) | 111 (31.1%) | |||
| Multiplicity | < 0.001 | 0.013 | 1.740 (1.125–2.691) | ||
| Solitary | 245 (81.7%) | 55 (18.3%) | |||
| Multiple | 229 (69.4%) | 101 (30.6%) | |||
| Gross extrathyroidal extension | 0.013 | 0.798 | 1.262 (0.213–7.486) | ||
| No | 353 (77.9%) | 100 (22.1%) | |||
| Yes | 121 (68.4%) | 56 (31.6%) | |||
| Number of metastatic lymph nodes | < 0.001 | 0.057 | 1.719 (0.984–3.003) | ||
| ≤5 | 377 (81.3%) | 87 (18.8%) | |||
| >5 | 97 (58.4%) | 69 (41.6%) | |||
| RAI dose (GBq) | 0.004 | 0.827 | 0.949 (0.596–1.512) | ||
| Low dose (≤1.85 GBq) | 242 (80.4%) | 59 (19.6%) | |||
| High dose (≥3.7 GBq) | 232 (70.5%) | 97 (29.5%) | |||
| Pre-RAI Tg (ng/mL) | < 0.001 | <0.001 | 1.705 (1.474–1.973) | ||
| Mean ± SEM | 0.8 ± 1.3 | 5.2 ± 10.4 | |||
| Urine I/Cr ratio (μg/g Cr) | 0.045 | 0.003 | 2.004 (1.270–3.162) | ||
| <66.2 | 327 (77.7%) | 94 (22.3%) | |||
| ≥66.2 | 147 (70.3%) | 62 (29.7%) | |||
| Preparation | 0.507 | ||||
| THW | 103 (77.4%) | 30 (22.6%) | |||
| rhTSH | 371 (74.6%) | 126 (25.4%) | |||
ER, excellent response; nER, non-excellent response; RAI, radioactive iodine; Tg, thyroglobulin; SEM, standard error of the mean; THW, thyroid hormone withdrawal; rhTSH, recombinant human thyrotropin.
3.5. Predictive Value of ratioTg in Relation to LID Adherence
We analyzed the relationship between therapeutic response and ratioTg in two groups stratified by LID adherence (Figure 2). In the well-adherent LID group (Figure 2A), ratioTg levels were significantly higher in the ER group (18.7 ± 2.8) than in the nER group (5.8 ± 0.8) (p = 0.013). In multivariate logistic regression analysis, a higher ratioTg was significantly associated with a reduced risk of nER in the well-adherent group (OR, 0.979; 95% CI, 0.961–0.998; p = 0.030) (Table 4). Other significant predictors of ER included female sex (OR, 1.745; 95% CI, 1.077–2.826; p = 0.024), solitary tumor (OR, 1.675; 95% CI, 1.035–2.711; p = 0.036), number of metastatic lymph nodes (OR, 1.928; 95% CI, 1.079–3.446; p = 0.027), and pre-RAI Tg (OR, 1.632; 95% CI, 1.420–1.875; p < 0.001) (Table 4). In contrast, in the poorly adherent LID group (Figure 2B), there was no significant difference in ratioTg between the ER (8.8 ± 1.3) and nER (9.3 ± 3.3) groups (p = 0.862).
Figure 2.
The relationship between therapeutic response and ratioTg in two groups categorized based on low iodine diet (LID) adherence. (A) In the well-adherent LID group, the ratioTg levels were significantly higher in the excellent response (ER) group (18.7 ± 2.8) compared to the non-excellent response (nER) group (5.8 ± 0.8) (p = 0.013). (B) In the poorly adherent LID group, there was no significant difference in ratioTg between the ER group (8.8 ± 1.3) and the nER group (9.3 ± 3.3) (p = 0.862). The graph represents mean ± standard error of the mean.
Table 4.
Factors associated with therapeutic response in well-adherent low-iodine diet group.
| Variables | ER (n = 456) | nER (n = 136) | Univariate | Multivariate | |
|---|---|---|---|---|---|
| p Value | p Value | Odds Ratio (95% CI) | |||
| Age (years) | 0.740 | ||||
| <55 years | 334 (76.6%) | 102 (23.4%) | |||
| ≥55 years | 122 (78.2%) | 34 (21.8%) | |||
| Sex | < 0.001 | 0.024 | 1.745 (1.077–2.826) | ||
| Male | 114 (66.3%) | 58 (33.7%) | |||
| Female | 342 (81.4%) | 78 (18.6%) | |||
| T stage | 0.124 | ||||
| T1 + T2 | 340 (78.7%) | 92 (21.3%) | |||
| T3 + T4 | 116 (72.5%) | 44 (27.5%) | |||
| N stage | < 0.001 | 0.669 | 1.102 (0.705–1.723) | ||
| N0a, N0b | 97 (89.0%) | 12 (11.0%) | |||
| N1a | 269 (79.6%) | 69 (20.4%) | |||
| N1b | 90 (62.1%) | 55 (37.9%) | |||
| Tumor size (cm) | 0.062 | ||||
| ≤1.0 | 213 (80.7%) | 51 (19.3%) | |||
| >1.0 | 243 (74.1%) | 85 (25.9%) | |||
| Multiplicity | 0.004 | 0.036 | 1.675 (1.035–2.711) | ||
| Solitary | 218 (82.6%) | 46 (17.4%) | |||
| Multiple | 238 (72.6%) | 90 (27.4%) | |||
| Gross extrathyroidal extension | 0.118 | ||||
| No | 346 (78.6%) | 94 (21.4%) | |||
| Yes | 110 (72.4%) | 42 (27.6%) | |||
| Number of metastatic lymph nodes | < 0.001 | 0.027 | 1.928 (1.079–3.446) | ||
| ≤5 | 356 (83.2%) | 72 (16.8%) | |||
| >5 | 100 (61.0%) | 64 (39.0%) | |||
| RAI dose (GBq) | 0.171 | ||||
| Low dose (≤1.85 GBq) | 237 (79.5%) | 61 (20.5%) | |||
| High dose (≥3.7 GBq) | 219 (74.5%) | 75 (25.5%) | |||
| Pre-RAI Tg (ng/mL) | < 0.001 | < 0.001 | 1.632 (1.420–1.875) | ||
| Mean ± SEM | 0.8 ± 0.1 | 9.5 ± 2.5 | |||
| RatioTg | 0.013 | 0.030 | 0.979 (0.961–0.998) | ||
| Mean ± SEM | 18.7 ± 2.8 | 5.8 ± 0.8 | |||
| Preparation | 0.488 | ||||
| THW | 108 (79.4%) | 28 (20.6%) | |||
| rhTSH | 348 (76.3%) | 108 (23.7%) | |||
ER, excellent response; nER, non-excellent response; RAI, radioactive iodine; Tg, thyroglobulin; SEM, standard error of the mean; THW, thyroid hormone withdrawal; rhTSH, recombinant human thyrotropin.
Sensitivity analysis excluding extreme ratioTg values yielded consistent results; the clinical benefit of LID in patients with ratioTg > 1.0 and the independent predictive value of ratioTg for treatment response remained statistically significant. These findings confirm that our primary conclusions were not disproportionately influenced by extreme outliers.
4. Discussion
In this study, LID adherence, defined by the urine I/Cr ratio, was associated with a significant increase in ratioTg, reflecting enhanced Tg release after RAI therapy. When stratified by a ratioTg threshold of 1, the association between LID adherence and therapeutic response was observed only in patients with ratioTg > 1. Additionally, ratioTg served as a predictive factor for therapeutic response only in the well-adherent LID group.
Although LID is usually recommended for RAI therapy, high-quality studies on its association with therapeutic response are limited, and conflicting results have been reported [9]. One study suggested that in DTC patients, urine iodine excretion exceeding 200 µg/day at the time of RAI therapy was associated with a higher risk of disease progression [17]. In contrast, another prospective study reported that urine iodine excretion did not have predictive value for therapeutic outcomes [18,19]. Our study also found no significant association between LID adherence and therapeutic response in the entire patient cohort (p = 0.177 in Table 2).
While LID is widely used as a preparatory step to enhance RAI uptake, its clinical efficacy depends on whether the accumulated RAI can effectively induce thyroid cell destruction. In other words, reducing iodine pool through LID may increase intracellular RAI concentration, but this only translates into therapeutic benefit when thyroid cells are sufficiently susceptible to radiation-induced damage [20]. Our findings further support this concept. Among patients with poor therapeutic response (nER), the ratioTg in the well-adherent LID group (5.8 ± 0.8) tended to be lower than that in the poorly adherent LID group (9.3 ± 3.3) (Figure 2). This suggests that even when iodine uptake is optimized through LID, effective cell destruction may not occur if the biological conditions are unfavorable. Thus, LID alone cannot guarantee improved outcomes, and its impact should be evaluated in the context of whether thyroid cells can actually be destroyed by RAI. These findings highlight the need for more precise criteria or biomarkers to identify patients who are likely to benefit from LID.
Following RAI therapy, Tg stored within thyroid follicular cells is passively released into the bloodstream due to radiation-induced membrane disruption and apoptosis. This results in a transient rise in serum Tg levels [13], which can indirectly reflect the extent of tissue destruction. Several prior studies have explored the clinical relevance of this phenomenon by examining the ratio between pre- and post-RAI Tg levels (ratioTg), a metric that normalizes for baseline variations across different TSH stimulation methods, demonstrating that higher ratioTg values are associated with better short-term outcomes [15,16]. These findings support the use of ratioTg as a surrogate marker for RAI-induced thyroid tissue response and, by extension, a context in which LID adherence could meaningfully influence therapeutic results. Building on this rationale, we employed ratioTg in our study to stratify patients and evaluate the association between LID adherence and therapeutic response in a biologically informed manner.
In our study, we stratified patients based on a ratioTg threshold of 1.0 to examine the potential therapeutic relevance of LID adherence under varying degrees of RAI-induced thyroid tissue response. This decision was based on the hypothesis that LID contributes to treatment efficacy only when RAI induces a certain degree of RAI-induced thyroid tissue response that leads to Tg release due to loss of cell membrane integrity. A ratioTg greater than 1 reflects a net increase in serum Tg following RAI therapy, which can reasonably be interpreted as a surrogate for acute tissue response. While this biologically informed threshold provides a practical framework for analysis, it should be noted that the cutoff value of 1.0 has not been universally validated. The selection was intended to distinguish between patients with minimal and those with appreciable Tg release. Further studies are warranted to refine this threshold and to determine whether alternative cutoffs or continuous evaluation of Tg changes could more accurately reflect the biologically relevant extent of thyroid tissue response.
These results have several implications. They suggest that the effect of LID should not be assessed simply through a binary comparison with treatment response, but rather within the biological context of thyroid tissue response following RAI therapy. RatioTg may serve as a surrogate marker to identify the conditions under which LID is likely to confer therapeutic benefit. The therapeutic benefit of LID was significant only in patients with elevated ratioTg levels, indicating that even with adequate LID adherence, additional therapeutic strategies may be necessary for patients exhibiting limited Tg release after RAI therapy. Nonetheless, caution is warranted in making direct causal interpretations, as ratioTg represents an intermediate post-intervention marker that may be subject to mediation or collider bias.
This study has several limitations. First, its retrospective design may introduce selection bias, though the inclusion of a relatively large number of patients may help reduce this concern. Second, the ratioTg is influenced by a complex interplay of radiation-induced follicle damage, transient inflammation, and Tg leakage. A single-point measurement on day 7 may not fully capture the entire kinetic profile of this response, which may vary between normal remnants and metastatic tissues. Further studies are needed to better understand how these temporal dynamics can be leveraged to more accurately reflect the magnitude of the acute biological response to RAI. In addition, the ratioTg cutoff of 1.0 was selected to distinguish between minimal and measurable Tg release following RAI therapy, rather than to optimize predictive performance. Further studies are warranted to refine this threshold and validate its clinical utility. Lastly, the binary classification (ER vs. nER) may obscure subgroup heterogeneity, and this study focused on short-term treatment outcomes. Future research with long-term follow-up is needed to assess the prognostic relevance of LID adherence under specific conditions, such as those indicated by ratioTg.
5. Conclusions
LID adherence was associated with increased Tg release following RAI, as reflected by higher ratioTg values. However, the therapeutic benefit of LID was observed only in patients with ratioTg > 1, suggesting that LID enhances therapeutic efficacy primarily in cases where a certain magnitude of radiation-induced Tg release is achieved. RatioTg may serve as a useful indicator for identifying patients likely to benefit from LID adherence.
Acknowledgments
The authors thank Young Jin Moon for his assistance with data processing and Subin Jeon for her support with statistical analysis.
Author Contributions
All authors contributed to the study conception and design. S.-Y.K. and S.W.Y. contributed to the study methodology. S.W.Y. was involved in data curation and investigation. S.W.Y., Y.M.N., and Y.J.R. contributed to the formal analysis. S.W.Y. wrote the original manuscript draft. H.K.K., H.-J.C., and S.-Y.K. reviewed and edited the manuscript. S.-Y.K. contributed to project administration and funding acquisition. All authors have read and agreed to the published version of the manuscript.
Institutional Review Board Statement
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of Chonnam National University Hwasun Hospital (protocol code CNUHH-2021-043 and date of approval 11 March 2021).
Informed Consent Statement
This retrospective study was approved by the institutional review board of our hospital in accordance with the Declaration of Helsinki (Approval No. CNUHH-2021-043), and the informed consent was waived.
Data Availability Statement
The data presented in this study are available on request from the corresponding author due to privacy and ethical restrictions.
Conflicts of Interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.
Funding Statement
This research was supported by the Bio & Medical Technology Development Program of the National Research Foundation (NRF) funded by the Korean government (MSIT) (NRF-2022R1F1A1073869); by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2025-00556583); and by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare, Republic of Korea (RS-2025-19252970, RS-2025-25459531). This study was also financially supported by Chonnam National University (2024-0437-01).
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
References
- 1.Huang J., Ngai C.H., Deng Y., Pun C.N., Lok V., Zhang L., Xu Q., Lucero-Prisno D.E., Xu W., Zheng Z.-J., et al. Incidence and mortality of thyroid cancer in 50 countries: A joinpoint regression analysis of global trends. Endocrine. 2023;80:355–365. doi: 10.1007/s12020-022-03274-7. [DOI] [PubMed] [Google Scholar]
- 2.Schmidbauer B., Menhart K., Hellwig D., Grosse J. Differentiated Thyroid Cancer-Treatment: State of the Art. Int. J. Mol. Sci. 2017;18:1292. doi: 10.3390/ijms18061292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Haugen B.R., Alexander E.K., Bible K.C., Doherty G.M., Mandel S.J., Nikiforov Y.E., Pacini F., Randolph G.W., Sawka A.M., 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. doi: 10.1089/thy.2015.0020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Oh S.W., Park S., Chong A., Kim K., Bang J.-I., Seo Y., Hong C.M., Lee S.-W. Radioactive Iodine Therapy in Differentiated Thyroid Cancer: Summary of the Korean Thyroid Association Guidelines 2024 from Nuclear Medicine Perspective, Part-II. Nucl. Med. Mol. Imaging. 2025;59:8–26. doi: 10.1007/s13139-024-00886-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Tian T., Qi Z., Huang S., Wang H., Huang R. Radioactive Iodine Therapy Decreases the Recurrence of Intermediate-Risk PTC With Low Thyroglobulin Levels. J. Clin. Endocrinol. Metab. 2023;108:2033–2041. doi: 10.1210/clinem/dgad045. [DOI] [PubMed] [Google Scholar]
- 6.Kim H.I., Kim B., Hyeon J., Ko N.G.M., Jin M.M., Cho J.H.M., Han J.M.P., Suh S., Bae J.C., Chung M.K., et al. Impact of Therapeutic Radioactive Iodine on the Recurrence of Papillary Thyroid Cancer With Extranodal Extension in Metastatic Lymph Nodes: A Propensity Score-matched Analysis. Clin. Nucl. Med. 2025;50:830–839. doi: 10.1097/RLU.0000000000005940. [DOI] [PubMed] [Google Scholar]
- 7.Liu Y., Wang J., Hu X., Pan Z., Xu T., Xu J., Jiang L., Huang P., Zhang Y., Ge M. Radioiodine therapy in advanced differentiated thyroid cancer: Resistance and overcoming strategy. Drug Resist. Updat. 2023;68:100939. doi: 10.1016/j.drup.2023.100939. [DOI] [PubMed] [Google Scholar]
- 8.Li J.H., He Z.H., Bansal V., Hennessey J.V. Low iodine diet in differentiated thyroid cancer: A review. Clin. Endocrinol. 2016;84:3–12. doi: 10.1111/cen.12846. [DOI] [PubMed] [Google Scholar]
- 9.Herbert G., England C., Perry R., Whitmarsh A., Moore T., Searle A., Chotaliya S., Ness A., Beasley M., Atkinson C. Impact of low iodine diets on ablation success in differentiated thyroid cancer: A mixed-methods systematic review and meta-analysis. Clin. Endocrinol. 2022;97:702–729. doi: 10.1111/cen.14751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Silaghi H., Lozovanu V., Georgescu C.E., Pop C., Nasui B.A., Cătoi A.F., Silaghi C.A. State of the Art in the Current Management and Future Directions of Targeted Therapy for Differentiated Thyroid Cancer. Int. J. Mol. Sci. 2022;23:3470. doi: 10.3390/ijms23073470. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Filetti S., Durante C., Hartl D., Leboulleux S., Locati L., Newbold K., Papotti M., Berruti A. Thyroid cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-updagger. Ann. Oncol. 2019;30:1856–1883. doi: 10.1093/annonc/mdz400. [DOI] [PubMed] [Google Scholar]
- 12.Haddad R.I., Bischoff L., Ball D., Bernet V., Blomain E., Busaidy N.L., Campbell M., Dickson P., Duh Q.-Y., Ehya H., et al. Thyroid Carcinoma, Version 2.2022, NCCN Clinical Practice Guidelines in Oncology. J. Natl. Compr. Cancer Netw. 2022;20:925–951. doi: 10.6004/jnccn.2022.0040. [DOI] [PubMed] [Google Scholar]
- 13.Stevic I., Dembinski T.C., Pathak K.A., Leslie W.D. Transient early increase in thyroglobulin levels post-radioiodine ablation in patients with differentiated thyroid cancer. Clin. Biochem. 2015;48:658–661. doi: 10.1016/j.clinbiochem.2015.04.009. [DOI] [PubMed] [Google Scholar]
- 14.Kim H.K., Lee S.Y., Lee J.I., Jang H.W., Kim S.K., Chung H.S., Tan A.H.K., Hur K.Y., Kim J.H., Chung J.H., et al. Usefulness of iodine/creatinine ratio from spot-urine samples to evaluate the effectiveness of low-iodine diet preparation for radioiodine therapy. Clin. Endocrinol. 2010;73:114–118. doi: 10.1111/j.1365-2265.2009.03774.x. [DOI] [PubMed] [Google Scholar]
- 15.Yoo S.W., Chowdhury M.S.A., Jeon S., Kang S.-R., Cho S.-G., Kim J., Lee C., Ryu Y.J., Song H.-C., Bom H.-S., et al. Optimization of Predictive Performance for the Therapeutic Response Using Iodine Scan-Corrected Serum Thyroglobulin in Patients with Differentiated Thyroid Carcinoma. Cancers. 2020;12:262. doi: 10.3390/cancers12020262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhu Y., Yang X., Liu Z., Zhang Q., Li Z., Hou X., Zhu H. Predictive value of thyroglobulin after radioiodine therapy for excellent response to treatment in postoperative thyroid cancer. Nucl. Med. Commun. 2025;46:146–151. doi: 10.1097/MNM.0000000000001933. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Luo H., Tobey A., Auh S., Cochran C., Behairy N., Merino M., Zemskova M., Klubo-Gwiezdzinska J. The utility of low-iodine diet in preparation for thyroid cancer therapy with radioactive iodine-A cohort study. Front. Pharmacol. 2022;13:791710. doi: 10.3389/fphar.2022.791710. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ito S., Iwano S., Kato K., Naganawa S. Predictive factors for the outcomes of initial I-131 low-dose ablation therapy to Japanese patients with differentiated thyroid cancer. Ann. Nucl. Med. 2018;32:418–424. doi: 10.1007/s12149-018-1261-0. [DOI] [PubMed] [Google Scholar]
- 19.Cao J., Yun C., Zhu X., Sun Y., Li X., Zhang W. Could Urinary Iodine Be an Effective Predictive Factor for Thyroid Cancer After High Dose Radioactive Iodine Therapy? Endocr. Pract. 2022;28:586–592. doi: 10.1016/j.eprac.2022.03.005. [DOI] [PubMed] [Google Scholar]
- 20.Reiners C., Drozd V., Yamashita S. Hypothyroidism after radiation exposure: Brief narrative review. J. Neural Transm. 2020;127:1455–1466. doi: 10.1007/s00702-020-02260-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data presented in this study are available on request from the corresponding author due to privacy and ethical restrictions.