Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2026 Aug 17;21(8):e0349850. doi: 10.1371/journal.pone.0349850

The effects of stimulated thyroglobulin/ thyroid-stimulating hormone ratio and N1b status on the efficacy of the first I-131 treatment in patients with differentiated thyroid cancer

Supanida Mayurasakorn 1,2,#, Charoonsak Somboonporn 1, Daris Theerakulpisut 1,*,#
Editor: Antimo Migliaccio3
PMCID: PMC13480613  PMID: 42607083

Abstract

Background

Thyroglobulin (Tg) level is one of the well-known factors affecting the outcome of radioactive iodine treatment in differentiated thyroid cancer, though this tumor marker is affected by the thyroid-stimulating hormone (TSH) level. To negate this influence, there is an effort to use the stimulated thyroglobulin/ thyroid-stimulating hormone (sTg/TSH) ratio instead of stimulated thyroglobulin (sTg) level alone. N1b cervical lymph node metastasis is also another factor mentioned in literature. This study aims to demonstrate the significance of these two factors whether they affect the treatment outcome in differentiated thyroid cancer patients who received the first radioactive iodine treatment.

Methods

This was a retrospective analytical observational study. The population were patients with differentiated thyroid cancer with an age of 18 years and above who received the first radioactive iodine treatment at our center from 1 January 2020–31 December 2023. The total population was 227 patients. Logistic regression analysis was done to determine the significant factors affecting the outcome of the first radioactive iodine treatment.

Results

In 227 patients, the median value for sTg and sTg/TSH ratio were 9.4 ng/mL and 0.2, respectively. Multiple logistic regression found that sTg/TSH ratio and N1b status were not independent risk factors for non-excellent response after radioactive iodine treatment. The independent risk factors were male sex, intermediate and high risk of recurrence, stage II and sTg of more than or equal to 9.4 ng/mL. Receiver operating characteristic (ROC) analysis showed that sTg and sTg/TSH ratio had rather similar performance in differentiating between excellent and non-excellent response groups.

Conclusion

sTg/TSH ratio and N1b cervical lymph node metastasis were not significant independent factors for non-excellent response after the first radioactive iodine treatment, which was possibly due to multicollinearity and small sample size. Independent risk factors for non-excellent response of the first radioiodine treatment were male sex, intermediate and high risk of recurrence, stage II and sTg of more than or equal to 9.4 ng/mL.

Introduction

Differentiated thyroid cancer (DTC) is one of the common cancers encountered in clinical practice. According to the Global Cancer Statistics (GLOBOCAN) in 2020, thyroid cancer ranked 9th among cancers worldwide and is the most frequently encountered endocrine gland cancer [1]. The majority of cases are papillary thyroid carcinoma (PTC), accounting for over 90%, followed by follicular carcinoma at about 4%, and oncocytic thyroid carcinoma at approximately 2%. [2,3]

Although thyroid cancer currently has well-defined treatment guidelines, up to the endpoint of this study, following the American Thyroid Association (ATA) guidelines in 2015 [4]. There remains heterogeneity within each risk of recurrence group which results in different responses to treatment and disease course among patients classified in the same risk category.

Radioactive iodine (RAI) treatment is one of the important treatment modalities in DTC. Factors affecting the success of RAI treatment are diverse, but one well-known factor is the level of thyroglobulin (Tg), a glycoprotein produced specifically by the thyroid gland. However, thyroglobulin value can be affected by changes in thyroid-stimulating hormone (TSH) levels, which are regulated by the hypothalamic-pituitary-thyroid axis. To mitigate the impact of TSH on Tg measurement, the ratio of stimulated thyroglobulin to TSH (sTg/TSH ratio) has been proposed as a more accurate indicator [5–13].

Another important factor, often associated with unsuccessful RAI therapy and less studied, is lymph node metastasis in the lateral cervical region (N1b) [14–21].

This study aims to investigate the influence of sTg/TSH ratio and lateral cervical lymph node metastasis (N1b) on the treatment response of patients who have initial RAI therapy for thyroid cancer. Additional receiver operating characteristic (ROC) analysis for the optimal cut-off of sTg and sTg/ratio is also performed.

Methods

Patients

Approved by the Khon Kaen University Ethics Committee for Human Research (protocol number HE661275), this single-center, retrospective study had its informed consent requirement waived due to its design. The hospital director authorized access to patient data.

Studied population were histological-proven differentiated thyroid cancer (DTC) patients aged 18 years and older who received the first radioactive iodine (RAI) therapy at Division of Nuclear Medicine, Department of Radiology, Srinagarind Hospital, Faculty of Medicine, Khon Kaen University.

Initial ethics approval was granted on 13 June 2023 which allowed access to retrospective data from 1 January 2018–31 December 2022. After preliminary analysis of data in this period, it was found that data older than 31 December 2019 were not appropriate for analysis since the laboratory technique used at the time was radioimmunoassay which was later replaced by chemiluminescence immunoassay. The reference range of these two methods was different and could not be directly compared to each other. Therefore, an amendment request was made to the ethics committee to change the data range from 1 January 2020–31 December 2023. The amendment was approved on 7 April 2024. The researchers had access to the patient’s identifying information during the data collection process. Information which would lead to the identification of the patient, hospital number, in this case, was recorded in a separate file which could be only assessed by the researchers and used only for verification purposes. No information which would lead to identification of the patient was included in the final data set for analysis and discussion. The dataset in the supporting information was also completely anonymized.

Patients were excluded if no reported sTg, TSH and TgAb, no SPECT/CT of the neck with the post-treatment whole body scan (WBS), unclear staging or risk of recurrence, missing data of treatment response evaluation, receiving RAI more than 6 months post-thyroidectomy, having a TSH level less than 25 μIU/mL before treatment, using recombinant human thyrotropin (rhTSH) for preparation, initial TgAb level >30 IU/mL by electrochemiluminescence immunoassay (ECLIA), receiving other treatments before, or using radioimmunoassay (RIA) which was different from the method currently used in our center.

Factors and outcome

Patients’ relevant clinical information was extracted from medical records including age at diagnosis, sex, surgical history, pathology report, staging according to AJCC/IUCC (8th edition), sTg, TSH and TgAb level, radioactive iodine dosage, risk of recurrence and treatment response according to ATA 2015 guideline. Determination of nodal and distant metastasis status is based on tissue diagnosis from surgery or radioiodine-avid lesion from the post-treatment WBS.

Laboratory assay for sTg, TgAb and TSH were electrochemiluminescence immunoassay (Roche, Rotkreuz, Switzerland). Radioactive iodine treatment follows an institutional protocol. The dosage of radioactive iodine corresponds with ATA risk of recurrence category and tumor staging. A dosage of 30 mCi is prescribed for patients with low-risk of recurrence and T1 or T2 disease. A dosage of 100 mCi is prescribed for low-risk of recurrence and T3 disease. A dosage of 150 mCi is prescribe for patients with intermediate and high-risk patients as well as those with distant lung metastases and those with T4 disease. A dosage of 200 mCi is reserved for patients with bone metastases.

After radioactive iodine administration for 3–14 days, post-treatment WBS and SPECT/CT of the neck region, visualized from the base of the skull down to the upper chest, was performed using Discovery NM/CT 670 (GE Healthcare, IL, USA) or Discovery NM 830 (GE Healthcare, IL, USA). Image analysis was performed using the Xeleris program (GE Healthcare, IL, USA). Image interpretation was performed by an experienced nuclear medicine physician.

Between 6 and 18 months following radioactive iodine treatment, the treatment response was evaluated as the primary outcome of the study. The outcome of this study was a binary outcome, i.e., excellent and non-excellent response. The excellent response group includes patients who had no clinical, biochemical, or structural evidence of disease, as determined by negative imaging results and a sTg level of less than 1 ng/mL. While non-excellent response group included indeterminate response, biochemical incomplete response, and structural incomplete response according to the ATA guideline 2015.

Statistical analysis

Patient characteristics were summarized using descriptive statistics. Continuous variables were reported as mean with standard deviation, if normally distributed, or median with interquartile range, if non-normally distributed. Categorical data was expressed as counts and percentages. Maximum and minimum value of the variables was also reported. Logistic regression was used to determine predictive factors of radioactive iodine treatment outcome. First, univariable logistic regression was performed on all predictors to screen for candidate variables. At this stage, predictors with a p-value of <0.2 was considered as a potential candidate and proceeded to be analyzed in the multivariable analysis. This liberal p-value threshold was chosen to avoid excluding potential important predictors which could turn out to be significant in the multivariable model. Subsequently, the selected variables were analyzed using multivariable logistic regression to determine the independent factors with a significant level of p-value <0.05 at this stage. Receiver Operating Characteristic (ROC) analysis was also performed to determine the optimal sTg and sTg/TSH ratio cut-offs in differentiating between excellent and non-excellent response groups.

In cases where extreme values were present, for example variables reported as exceeding the upper limit or falling below the lower limit in laboratory tests, such as sTg < 0.04 ng/mL, sTg > 500 ng/mL, or TSH > 100 µIU/mL, it was impossible to calculate the true numerical value of the sTg/TSH ratio. This was not uncommon in real-world clinical practice. The analysis was therefore performed in two ways: first, by excluding data from patients whose variable values exceeded the upper limit or fell below the lower limit of detection, i.e., sTg < 0.04 ng/mL, sTg > 500 ng/mL, or TSH > 100 µIU/mL; and second, by dividing the population into subgroups to cover all potential combinations of sTg and TSH results.

A total of seven subgroups were defined based on different combinations of sTg, TSH, and the sTg/TSH ratio, as follows. Subgroup 1 comprised patients with both sTg and TSH within reportable ranges and sTg/TSH ratio < mean or median. Subgroup 2 comprised patients with both sTg and TSH within reportable ranges and sTg/TSH ratio ≥ mean or median. Subgroup 3 included patients with TSH within a reportable range and sTg < 0.04 ng/mL. Subgroup 4 included patients with TSH within a reportable range and sTg > 500 ng/mL. Subgroup 5 included patients with TSH > 100 µIU/mL and sTg < 0.04 ng/mL. Subgroup 6 included patients with TSH > 100 µIU/mL and sTg > 500 ng/mL. Subgroup 7 included patients with TSH > 100 µIU/mL and sTg within a reportable range.

Each subgroup was analyzed independently using logistic regression. The subgroups were also further combined into two broader groups based on clinical perspective: a potential low Tg and low sTg/TSH ratio group, comprising Subgroups 1, 3, 5, and 7, and a potential high Tg and high sTg/TSH ratio group, comprising Subgroups 2, 4, and 6. These two broader groups were also analyzed by logistic regression.

Descriptive statistics and logistic regression were performed using STATA version 18 (StataCorp LLC, TX, USA). ROC analysis, which was analyzed using R Statistical Software (version 4.2.1; R Core Team), RStudio version 2023.12.1 + 402 “Ocean Storm” (RStudio, Boston, MA) and the R package, pROC. Power calculation was done using the R package, WebPower.

Sample size considerations

All patient records that are available in the research data collection period which met the aforementioned inclusion and exclusion criteria were included in the analysis. To determine the adequacy of the final sample size, power calculation was done using the R package, WebPower. The statistical power was determined based on the difference of the prevalence of non-excellent response in cases with a positive versus negative key predictor of interest, and a significance level of 0.05.

Results

The population included in the analysis was 227 subjects, among which 153 (67.4%) had non-excellent response outcome. Among patients with N1b node status the non-excellent outcome was 82%, as compared with 60% in those without N1b lymph nodes. This resulted in a statistical power of approximately 89%. Detailed characteristics of the cohort were presented in S1 Table. Since the variable exhibited a non-normal distribution, the median was used instead of the mean. The majority of the population was female (80%), with a median age of 51 years. The predominant histological type was papillary thyroid carcinoma (79%), followed by follicular thyroid carcinoma (10%). Approximately half of the patients had lymph node metastasis. Among those with lymph node spread, more than half (63%) involved the lateral cervical lymph node group (N1b). Only a minority of patients (6.6%) had distant metastasis. Most of the patients had stage I cancer (68%). About one-third (33%) of patients achieved an excellent response to treatment. Regarding the laboratory values, the median sTg, median TSH and median sTg/TSH ratio was 9.4 ng/mL, 71.5 µIU/mL and 0.2, respectively. These values were used as a cut-off for dividing aforementioned sTg/TSH ratio and TSH subgroups.

Multivariable logistic regression analysis was performed on 227 participants. Model performance was evaluated using the Akaike Information Criterion (AIC = 223.6) and Hosmer-Lemeshow goodness-of-fit test (χ²(8) = 8.62, p = 0.376), confirming adequate model fit. Logistic regression analysis results of sTg/TSH ratio, lateral cervical lymph node metastasis and other significant variables were shown in Table I, the detailed results were shown in S2 Table.

Since a total of 5 variables were included in the multivariable logistic regression analysis, at least 50 events were needed according to the 10-events per variable rule. In this study, there are a total of 153 events, which is adequate for a reliable regression model (Table 1).

Table 1. Logistic regression analysis results.

Variable Total cases
(N = 227)		 Univariable Multivariable
OR
(95% CI)		 P-value Adjusted OR
(95% CI)		 P-value
sTg/TSH ratio (IQR) 0.2
(0.06,0.68)		 17.97
(2.68,120.6)		 0.003
sTg (IQR) 9.37
(3.77,34.5)		 1.04
(1.02,1.06)		 <0.001
sTg < 9.4 107 Ref.
sTg ≥ 9.4
(including sTg > 500)		 120 6.91
(3.65,13.08)		 <0.001 5.29
(2.6,10.75)		 <0.001
Cervical node metastasis location
N1a		 36 Ref.
N1b 74 0.59
(0.18,1.95)		 0.383
N1x 8 0.38
(0.06,2.53)		 0.314
Male 46 3.27
(1.39,7.73)		 0.007 4.78
(1.76,12.96)		 0.002
Stage II (compared with stage I) 62 4.19
(1.93,9.1)		 <0.001 2.63
(1.09,6.35)		 0.032
Intermediate risk of recurrence
(compared with low risk)		 97 3.4
(1.78,6.51)		 <0.001 2.92
(1.39,6.12)		 0.005
High risk of recurrence
(compared with low risk)		 61 12.64
(4.8,33.31)		 <0.001 5.86
(2.01, 17.03)		 0.001
age (IQR) 51
(38,61)		 1.02
(1.00,1.04)		 0.018
Multifocality 94 1.97
(1.09,3.54)		 0.024
Gross extrathyroidal extension (compared with no ETE) 18 5.42
(1.18,24.81)		 0.029
Minimal extrathyroidal extension (compared with no ETE) 34 1.88
(0.82, 4.63)		 0.148
Unknown extrathyroidal extension (compared with no ETE) 71 1.61
(0.85, 3.10)		 0.145
Positive surgical margin 41 4.45
(1.66,11.93)		 0.003
Presence of cervical lymph node metastasis 118 5.31
(2.86,9.85)		 <0.001
I-131 150 mCi 143 3.71
(2.05,6.71)		 <0.001
sTg/TSH ratio subgroup
Subgroup 1		 62 Ref.
Subgroup 2 64 4.71
(1.98,11.2)		 <0.001
Subgroup 4 7 NA
Subgroup 6 4 NA
Subgroup 7 90 1.06
(0.55,2.03)		 0.871
sTg/TSH ratio subgroup 1, 3, 5, 7 152 Ref.
sTg/TSH ratio subgroup 2, 4, 6 75 5.48
(2.54,11.8)		 <0.001
Subgroup 7: TSH > 100 µIU/mL and sTg have a reportable value
Tg < 9.4		 52 Ref.
Tg ≥ 9.4 38 5.11
(1.97,13.28)		 0.001
Open in a new tab

N1x; nodal metastasis that could not be clearly identified whether it was stage N1a or N1b, NA; not applicable, Ref.; reference.

Comparing the ROC curves of sTg and sTg/TSH ratio, ROC curves of both variables were very similar, and their respective area under the curve (AUC) values were also very close (Fig 1). The AUC and 95% confidence interval (95%CI) for sTg and the sTg/TSH ratio were 0.7731 (95%CI:0.6899–0.8564) and 0.7725 (95%CI:0.6894–0.8556), respectively. DeLong’s test yielded a P-value of 0.729 which indicates no significant difference between the area under the curves of the two ROCs.

Fig 1. The ROC analysis of sTg and the sTg/TSH ratio.

Fig 1

Open in a new tab

This demonstrated the ability of both variables to differentiate between the excellent and non-excellent response groups. The two graphs were found to be very similar, with the yellow line representing the sTg value and the green line representing the sTg/TSH ratio, respectively.

For the assessment of the optimal cut-off values for sTg and sTg/TSH ratio, two methods were used: First, the cut-off was derived from the median of the study, and second, it was derived from the value with the highest Youden index. The location of these cut-off points on the ROC curve was shown in Figs 2 and 3. The numerical cut-off values, along with the corresponding sensitivity and specificity, were presented in Table 2.

Fig 2. A comparison of the positions of different sTg cut-off values on the ROC curve.

Fig 2

Open in a new tab

The blue dot represented the cut-off value derived from the study’s median, and the red dot represented the cut-off value derived from the Youden index.

Fig 3. A comparison of the positions of different sTg/TSH ratio cut-off values on the ROC curve.

Fig 3

Open in a new tab

The blue dot represented the cut-off value derived from the study’s median, and the red dot represented the cut-off value derived from the Youden index.

Table 2. Cut-off values of sTg and sTg/TSH ratio from the median and Youden index.

Variables and cut-offs Sensitivity Specificity
sTg 9.4 ng/mL (median) 0.65 0.77
sTg 10.1 ng/mL (Youden index) 0.81 0.67
sTg/TSH ratio 0.2 (median) 0.61 0.75
sTg/TSH ratio 0.23 (Youden index) 0.86 0.60
Open in a new tab

The cut-off values by two methods were very close. However, there might be some differences in the diagnostic performance due to a trade-off between sensitivity and specificity resulting from the selection of different cut-off values.

Discussion

The primary objective of this study was to determine whether the sTg/TSH ratio and lateral cervical lymph node metastasis (N1b) were significant factors affecting the outcome of the first radioactive iodine treatment in patients with differentiated thyroid cancer. The optimal cut-off for sTg and the sTg/TSH ratio to differentiate between the excellent and non-excellent response groups were also analyzed using ROC analysis.

From the logistic regression analysis, the independent risk factors for developing a non-excellent response to treatment were: male sex, ATA intermediate and high risk of recurrence, stage II cancer, and an sTg value greater than or equal to 9.4 ng/mL.

Regarding the male sex and ATA intermediate and high risk of recurrence factors, these findings were consistent with previous studies [9,15,20]. The ATA guideline 2015 [4] also states that a higher risk of recurrence was associated with an increased risk of structural persistent/recurrent disease.

Generally, as the cancer stage increased, a poor treatment outcome was more likely to be observed. This was consistent with the results of this study, which found that stage II cancer significantly affected treatment outcome negatively compared to stage I. Stage III cancer, however, was not found to be a factor causing an increased poor treatment outcome. This might be due to the low number of patients in stage III, resulting in insufficient statistical power. For stage IV cancer, the odds ratio could not be calculated because the rate of the event of interest, a poor treatment outcome, was 100%.

An sTg ≥ 9.4 ng/mL, including a group with sTg > 500 ng/mL, was associated with a non-excellent response, serving as an independent risk factor (adjusted OR=5.29; 95%CI:2.6–10.75; p < 0.001). This finding was true even within the subgroup 7 (TSH > 100 µIU/mL with reportable sTg value) with crude OR of 5.11 (95%CI:1.97–13.28, p-value 0.001). The finding that a high sTg value led to a poor treatment outcome was consistent with several previous literature.

Regarding the sTg/TSH ratio, a high ratio was found to have a crude OR for a non-excellent response of 17.97 (95%CI:2.68–120.58, p-value = 0.003). However, it lost significance in the multivariable logistic regression analysis. The loss of significance and the wide confidence interval might be due to the multicollinearity among the studied variables. Nevertheless, it was found that subgroup 2 (sTg/TSH ratio ≥0.2) and other subgroups where the sTg/TSH ratio was likely to be high, such as subgroup 4 (TSH reportable and Tg > 500 ng/mL) and subgroup 6 (TSH > 100 µIU/mL and Tg > 500 ng/mL), all showed significance for a poor treatment outcome in univariable logistic regression analysis.

Other variables which were found to be significant in the univariable logistic regression, but not significant in the multivariable logistic regression analysis might be partially explained by the multicollinearity. For example, the correlation of gross ETE with having a positive margin and being classified into the high risk of recurrence group, cervical node metastasis with being classified into the intermediate risk of recurrence group. For the radioactive iodine dosage, patients receiving higher radioiodine dosage often had a poorer prognosis; this is likely due to their baseline disease severity rather than the treatment itself. The presence or absence of cervical node metastasis was not the only factor that might affect treatment outcome. The characteristics of the lymph node itself, such as size and number, were also important. For age, although it was generally believed that increasing age was associated with worse disease progression, some studies had found that age was not an independent risk factor for a poor treatment outcome. Thamnirat et al. in 2015 [22] in 256 patients found that age ≤ 45 years was not associated with disease-free status in multivariate analysis. Additionally, a study by Ganly et al. in 2015 [23] involving 3,664 thyroid cancer patients found that when age was considered as a continuous variable affecting disease-specific survival, the adjusted hazard ratio (aHR) was 1.076 (95%CI:1.06–1.1, p-value<0.001). The confidence interval was close to 1, the null hypothesis, thus studies with smaller populations, such as this study and Thamnirat et al., might not have sufficient power enough to find age as a significant factor.

Lateral cervical lymph node metastasis (N1b) was found to be non-significant for the development of a non-excellent response in this study. This lack of significance was possibly due to the insufficient sample size of the population with this specific factor. For other lymph node characteristics such as numbers, size, and extranodal extension, these factors were analyzed in the univariate analysis, as shown in S2 Table. None of the factors showed significance in predicting treatment outcome. Due to the fact that some patients were referred from other hospitals, completed details of lymph node characteristics could not be obtained for some cases.

The ability of sTg and the sTg/TSH ratio to differentiate between patients with excellent and non-excellent response outcomes, as shown by the ROC analysis, was found to be rather similar based on the shape of the graph, AUC values and also the cut-off values from the median and Youden index were rather close, at 9.4 and 10.1 ng/dL, respectively. However, it is notable that this minimal shift in the cutoff point causes the sensitivities and specificities to be “flipped”, with the cutoff of 9.4 ng/mL having a sensitivity of 65% and a specificity of 77%, while moving the cutoff to 10.1 ng/dL results in a higher sensitivity of 81% and a lower specificity of 67%. This can be explained by “datapoint clumping” with many of the patients with excellent response, and non-excellent response having the sTg around these threshold boundaries. That being said, in our opinion, a cutoff of 10.1 ng/dL based on the Youden index would be more appropriate as it gives a higher sensitivity and an acceptable specificity. This cutoff is also similar to the findings of a large study by Webb et al. in 2012 [24], which was a meta-analysis of 3,947 patients. That study proposed that a pre-ablation Tg value less than 10 ng/mL was a predictor with a high negative predictive value for determining the disease-free status of patients. As for the sTg/TSH ratio, a similar phenomenon was found when analyzing the cutoff values. Moreover, the cutoff of 0.23 based on the Youden index yielded a sensitivity of 86% at the expense of a lower specificity of 60% compared with sTg. Therefore, the sTg/TSH ratio does not reveal superiority to sTg based on these current findings.

There were several limitations of this study. First, this study has been conducted before the publication of new ATA guideline for differentiated thyroid cancer in 2025. Patients treated at our center were managed based on the now superseded ATA 2015 guidelines, thus there might be some difference in risk stratification and management of the patient. Second, due to retrospective single-center design of the study, this might lead to incomplete data for some variables, limitation in control of confounders and the data should be considered before applying to other population. Third, the nature of lymph node metastasis in thyroid cancer could be difficult to diagnose in some situations because nodal dissection is not routinely done in thyroid cancer and nodal metastasis might be small and had minimal RAI uptake, resulting in limited sample size and potential false negative. Fourth, certain factors with insufficient numbers of patients and presence of a relatively high degree of multicollinearity among the variables meant that some variables might be found to be non-significant in the statistical analysis. Lastly, due to the rather indolent disease course of thyroid cancer, limited follow-up time might not truly reflect the long-term outcome of the patients, thus longer follow-up time was encouraged in further studies.

Conclusion

The sTg/TSH ratio and the presence of lateral cervical lymph node metastasis (N1b) were not independent risk factors for a non-excellent response in this study, which could be due to a relatively small sample size in the context of the presence of multicollinearity with other well-established predictors such as sTg.

The independent risk factors for developing a non-excellent response to radioactive iodine treatment in patients with differentiated thyroid cancer were: male sex, intermediate and high risk of recurrence according to ATA guideline 2015, Stage II cancer, and an sTg value greater than or equal to 9.4 ng/mL.

Supporting information

S1 Table. Detailed clinical characteristics of the cohort.

(DOCX)

pone.0349850.s001.docx (23.1KB, docx)
S2 Table. Detailed results from logistic regression analysis for factors associated with non-excellent response outcome after I-131 treatment.

NA; not applicable, Ref.; reference.

(DOCX)

pone.0349850.s002.docx (41.2KB, docx)
S3 Dataset. Anonymized dataset.

(XLSX)

pone.0349850.s003.xlsx (38.8KB, xlsx)

Acknowledgments

I would like to express my gratitude to Pongkhun Pantaneeya, for his assistance and valuable advice in part of statistical analysis

Data Availability

All relevant data are within the paper and its Supporting Information files.

Funding Statement

The author(s) received no specific funding for this work.

References

  • 1.Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–49. [DOI] [PubMed] [Google Scholar]
  • 2.Rossi ED, Pantanowitz L, Hornick JL. A worldwide journey of thyroid cancer incidence centred on tumour histology. Lancet Diabetes Endocrinol. 2021;9(4):193–4. doi: 10.1016/S2213-8587(21)00049-8 [DOI] [PubMed] [Google Scholar]
  • 3.Kitahara CM, Schneider AB. Epidemiology of Thyroid Cancer. Cancer Epidemiology, Biomarkers & Prevention. 2022;31(7):1284–97. doi: 10.1158/1055-9965.epi-21-1440 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, 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):1–133. doi: 10.1089/thy.2015.0020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Lin Y, Li T, Liang J, Li X, Qiu L, Wang S, et al. Predictive value of preablation stimulated thyroglobulin and thyroglobulin/thyroid-stimulating hormone ratio in differentiated thyroid cancer. Clin Nucl Med. 2011;36(12):1102–5. doi: 10.1097/RLU.0b013e3182291c65 [DOI] [PubMed] [Google Scholar]
  • 6.Trevizam PGC, Tagliarini JV, Castilho EC, de Alencar Marques M, Kiy Y, Mazeto GMF da S. Thyroglobulin levels and thyroglobulin/thyrotropin ratio could predict the success of the ablative/therapeutic 131I in the differentiated thyroid cancers. Endocr Res. 2017;42(1):42–8. doi: 10.3109/07435800.2016.1173056 [DOI] [PubMed] [Google Scholar]
  • 7.Prpic M, Kust D, Kruljac I, Kirigin LS, Jukic T, Dabelic N, et al. Prediction of radioactive iodine remnant ablation failure in patients with differentiated thyroid cancer: A cohort study of 740 patients. Head Neck. 2017;39(1):109–15. doi: 10.1002/hed.24550 [DOI] [PubMed] [Google Scholar]
  • 8.De Marchi AF Junior, de Macedo ABT, Soares CSP, Bolfi F, Iessi MRG, de Oliveira CC, et al. Thyroglobulin/thyrotropin ratio for predicting long-term response in differentiated thyroid carcinoma: a retrospective study. Arch Endocrinol Metab. 2021;65(4):428–35. doi: 10.20945/2359-3997000000387 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Ju Y, Wang L, Cheng F, Huang F, Chen X, Song Q, et al. Comparing the efficacy of thyroglobulin and thyroglobulin/ thyroid-stimulating hormone ratio models in predicting a successful response to radioactive iodine therapy. BMC Endocr Disord. 2023;23(1):19. doi: 10.1186/s12902-022-01261-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Zubair Hussain S, Zaman MU, Malik S, Ram N, Asghar A, Rabbani U, et al. Preablation Stimulated Thyroglobulin/TSH Ratio as a Predictor of Successful I(131)Remnant Ablation in Patients with Differentiated Thyroid Cancer following Total Thyroidectomy. J Thyroid Res. 2014;2014:610273. doi: 10.1155/2014/610273 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Wang C, Diao H, Ren P, Wang X, Wang Y, Zhao W. Efficacy and affecting factors of 131I thyroid remnant ablation after surgical treatment of differentiated thyroid carcinoma. Front Oncol. 2018;8:640. doi: 10.3389/fonc.2018.00640 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Wen R, Zhao M, Chen C, Yang Y, Zhang B. A novel nomogram integrated with preablation stimulated thyroglobulin and thyroglobulin/thyroid-stimulating hormone ratio to predict the therapeutic response of intermediate‑ and high‑risk differentiated thyroid cancer patients: a bi-center retrospective study. Endocrine. 2024;84(3):989–98. doi: 10.1007/s12020-023-03625-y [DOI] [PubMed] [Google Scholar]
  • 13.Zheng W, Rui Z, Wang X, Li N, Tan J, Liu W. The Influences of TSH Stimulation Level, Stimulated Tg Level and Tg/TSH Ratio on the Therapeutic Effect of 131I Treatment in DTC Patients. Front Endocrinol (Lausanne). 2021;12:601960. doi: 10.3389/fendo.2021.601960 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Qu Y, Huang R, Li L. Low- and high-dose radioiodine therapy for low-/intermediate-risk differentiated thyroid cancer: a preliminary clinical trial. Ann Nucl Med. 2017;31(1):71–83. doi: 10.1007/s12149-016-1133-4 [DOI] [PubMed] [Google Scholar]
  • 15.Liu N, Meng Z, Jia Q, Tan J, Zhang G, Zheng W. Multiple-factor analysis of the first radioactive iodine therapy in post-operative patients with differentiated thyroid cancer for achieving a disease-free status. Sci Rep. 2016;6:34915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Sapuppo G, Palermo F, Russo M, Tavarelli M, Masucci R, Squatrito S. Latero-cervical lymph node metastases (N1b) represent an additional risk factor for papillary thyroid cancer outcome. J Endocrinol Invest. 2017;40(12). [DOI] [PubMed] [Google Scholar]
  • 17.Barbosa MP, Momesso D, Bulzico DA, Farias T, Dias F, Lima RA, et al. Metastatic lymph node characteristics as predictors of recurrence/persistence in the neck and distant metastases in differentiated thyroid cancer. Arch Endocrinol Metab. 2017;61(6):584–9. doi: 10.1590/2359-3997000000307 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Liu Y-Q, Li H, Liu J-R, Lin Y-S. Unfavorable responses to radioiodine therapy in n1b papillary thyroid cancer: a propensity score matching study. Endocr Pract. 2019;25(12):1286–94. doi: 10.4158/EP-2019-0155 [DOI] [PubMed] [Google Scholar]
  • 19.Kim M, Jeon MJ, Oh H-S, Park S, Song DE, Sung T-Y, et al. Prognostic Implication of N1b Classification in the Eighth Edition of the Tumor-Node-Metastasis Staging System of Differentiated Thyroid Cancer. Thyroid. 2018;28(4):496–503. doi: 10.1089/thy.2017.0473 [DOI] [PubMed] [Google Scholar]
  • 20.Mendoza ES, Lopez AA, Valdez VAU, Cunanan EC, Matawaran BJ, Kho SA, et al. Predictors of incomplete response to therapy among Filipino patients with papillary thyroid cancer in a tertiary hospital. J Endocrinol Invest. 2016;39(1):55–62. doi: 10.1007/s40618-015-0319-2 [DOI] [PubMed] [Google Scholar]
  • 21.Alzahrani AS, Moria Y, Mukhtar N, Aljamei H, Mazi S, Albalawi L. Course and predictive factors of incomplete response to therapy in low- and intermediate-risk thyroid cancer. J Endocr Soc. 2020;5(1):bvaa178. doi: 10.1210/jendso/bvaa178 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Thamnirat K, Utamakul C, Chamroonrat W, Kositwattanarerk A, Anongpornjossakul Y, Sritara C. Factors affecting disease-free status of differentiated thyroid carcinoma patients. Asian Pac J Cancer Prev. 2015;16(2):737–40. doi: 10.7314/apjcp.2015.16.2.737 [DOI] [PubMed] [Google Scholar]
  • 23.Ganly I, Nixon IJ, Wang LY, Palmer FL, Migliacci JC, Aniss A, et al. Survival from Differentiated Thyroid Cancer: What Has Age Got to Do with It?. Thyroid. 2015;25(10):1106–14. doi: 10.1089/thy.2015.0104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Webb RC, Howard RS, Stojadinovic A, Gaitonde DY, Wallace MK, Ahmed J, et al. The utility of serum thyroglobulin measurement at the time of remnant ablation for predicting disease-free status in patients with differentiated thyroid cancer: a meta-analysis involving 3947 patients. J Clin Endocrinol Metab. 2012;97(8):2754–63. doi: 10.1210/jc.2012-1533 [DOI] [PubMed] [Google Scholar]

Decision Letter 0

Afagh Hassanzadeh Rad

16 Feb 2026

Dear Dr. Theerakulpisut,

Please submit your revised manuscript by Apr 02 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

  • A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Afagh Hassanzadeh Rad

Academic Editor

PLOS One

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include a caption for figure 3.

3. We note that there is identifying data in the Supporting Information file <S3_dataset.xlsx>. Due to the inclusion of these potentially identifying data, we have removed this file from your file inventory. Prior to sharing human research participant data, authors should consult with an ethics committee to ensure data are shared in accordance with participant consent and all applicable local laws.

Data sharing should never compromise participant privacy. It is therefore not appropriate to publicly share personally identifiable data on human research participants. The following are examples of data that should not be shared:

-Name, initials, physical address

-Ages more specific than whole numbers

-Internet protocol (IP) address

-Specific dates (birth dates, death dates, examination dates, etc.)

-Contact information such as phone number or email address

-Location data

-ID numbers that seem specific (long numbers, include initials, titled “Hospital ID”) rather than random (small numbers in numerical order)

Data that are not directly identifying may also be inappropriate to share, as in combination they can become identifying. For example, data collected from a small group of participants, vulnerable populations, or private groups should not be shared if they involve indirect identifiers (such as sex, ethnicity, location, etc.) that may risk the identification of study participants.

Additional guidance on preparing raw data for publication can be found in our Data Policy (https://journals.plos.org/plosone/s/data-availability#loc-human-research-participant-data-and-other-sensitive-data) and in the following article: http://www.bmj.com/content/340/bmj.c181.long.

Please remove or anonymize all personal information (<specific identifying information in file to be removed>), ensure that the data shared are in accordance with participant consent, and re-upload a fully anonymized data set. Please note that spreadsheet columns with personal information must be removed and not hidden as all hidden columns will appear in the published file.

4. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Partly

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #1: I Don't Know

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #1: Yes

Reviewer #2: Yes

**********

Reviewer #1: It would be helpful to provide more information about the radioactive iodine treatment protocol at your center. Is there a standardized treatment at your center? This information is important for the reader to know if changes in treatment regimen may have affected the results.

The use of a significance level of p < 0.2 in univariate regression may be confusing to some readers. Typically, a p-value < 0.05 is used in both steps. If there is a specific reason for choosing a p-value level of < 0.2 in this step, it is suggested that a brief explanation be provided. This will help clarify your reasoning for this choice.

The section on dividing the data into subgroups is well explained, but needs further clarification. Why were these specific combinations chosen for the subgroup division?

These can explain to readers why this particular approach was chosen and how it might affect the results.

In cases where Tg values < 0.04 or >500 ng/mL or TSH > 100 µIU/mL are reported, a clear explanation of how these values were handled in the analyses appears to be provided. However, further clarification of how these missing or incorrect data are handled (such as omitting these data or using alternative methods for analysis) would be useful.

Especially in subgroups where these specific values exist, mention whether incomplete data were removed from the analyses or how you dealt with these values.

Further explanation on how lymph node metastasis was assessed in the study (was advanced imaging used?) and how metastasis was detected at an early stage. However, it is suggested that more explanation be provided in the ROC analysis about the sensitivity and specificity at these thresholds.

In the analysis section, age is discussed in detail, but it seems that further explanation of why age ≤45 years is not effective as an independent factor in some studies would be valuable.Is this due to a lack of evidence in small samples, or are different analysis methods in different studies causing different results?

Ultimately, your article is very valid and useful, and by making these corrections, its scientific credibility will increase

Reviewer #2: Dear Editor

Thank you for the opportunity to review this manuscript.

My comments are as follows:

This manuscript addresses an important and clinically relevant topic, namely the prediction of response to radioactive iodine (RAI) therapy in patients with differentiated thyroid cancer (DTC). The finding regarding sTg ≥ 9.4 ng/mL is noteworthy, and the use of internationally recognized sources such as GLOBOCAN and ATA guidelines reflects an appropriate scientific foundation. However, the study has several important limitations that require substantial revision prior to consideration for publication.

First, the manuscript relies on the ATA 2015 guidelines, suggesting that risk stratification and treatment response assessment were based on an outdated framework. Since then, significant updates have been introduced, including the 2022 update and the more comprehensive 2025 ATA guidelines. Continued reliance on ATA 2015 may limit alignment with current standards of care, particularly regarding recurrence risk stratification, RAI dosing strategies, and management of patients with N1b disease. The authors should either update their framework in accordance with ATA 2025 or explicitly acknowledge these updates in the Discussion and justify the use of the earlier version.

Second, the study’s retrospective, single-center design inherently limits control of confounding variables and reduces external validity. As the research was conducted in a single institution in Thailand, the generalizability of the findings to other populations and healthcare systems may be restricted.

Third, the change in laboratory assay methodology (from RIA to ECLIA) and the exclusion of earlier data potentially introduced selection bias and reduced the effective sample size. While the timeline, laboratory methods, and exclusion criteria are clearly described, inclusion of certain operational details (e.g., exact dates of data access approval) appears unnecessary and may detract from the methodological clarity of the manuscript.

Fourth, the limited sample size within certain subgroups and the presence of multicollinearity among sTg, the sTg/TSH ratio, and other covariates likely contributed to the loss of statistical significance of key variables (notably the sTg/TSH ratio and N1b status) in multivariable regression analysis. Despite this, the Discussion emphasizes the potential usefulness of the sTg/TSH ratio, which is not supported by the adjusted model. This creates inconsistency between statistical findings and interpretation.

Fifth, the follow-up period of 6–18 months primarily reflects short-term treatment response rather than long-term prognosis. Given the typically indolent course of DTC, longer follow-up would be necessary to assess recurrence and durable outcomes.

Sixth, important lymph node characteristics (such as size and number of metastatic nodes), which may influence prognosis, were not fully analyzed.

Seventh, the analysis of RAI dosage may be affected by confounding by indication. Higher RAI doses (150–200 mCi) were likely administered to patients with more advanced disease or higher ATA risk categories. Therefore, the observed association between higher RAI dose and poorer treatment response may reflect baseline disease severity rather than an independent causal effect of the administered dose. The manuscript does not clearly demonstrate adequate statistical adjustment for this potential bias. Stratified analyses within risk categories or the use of propensity score methods would strengthen internal validity.

Finally, there appears to be a discrepancy between the primary objective and the final conclusions. The study aimed to evaluate the impact of the sTg/TSH ratio and N1b status on treatment response; however, neither variable remained significant in multivariable analysis. Ultimately, only sTg was identified as an independent predictor — a finding already well established in prior literature. Consequently, the degree of novelty is limited.

In light of these considerations, the manuscript requires major revision, including updating the guideline framework, refining the statistical modeling, simplifying subgroup analyses, and providing a clearer and more balanced discussion of limitations and clinical implications, before it can be considered for publication in a reputable journal.

Sincerely

**********

what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review?  For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #1: Yes: Sara Nikpour

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures

You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation.

NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications.

PLoS One. 2026 Aug 17;21(8):e0349850. doi: 10.1371/journal.pone.0349850.r002

Author response to Decision Letter 1

26 Feb 2026

Response to reviewers can be found in the attached "Response to reviewers.docx" file.

Attachment

Submitted filename: Response to reviewers.docx

pone.0349850.s005.docx (2.7MB, docx)

Decision Letter 1

Alejandro Torrado Pacheco

11 Jun 2026

Dear Dr. Theerakulpisut,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The manuscript has been evaluated by two reviewers, and their comments are available below.

Could you please revise the manuscript to carefully address the concerns raised?

Please submit your revised manuscript by Jul 26 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

  • A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

As the corresponding author, your ORCID iD is verified in the submission system and will appear in the published article. PLOS supports the use of ORCID, and we encourage all coauthors to register for an ORCID iD and use it as well. Please encourage your coauthors to verify their ORCID iD within the submission system before final acceptance, as unverified ORCID iDs will not appear in the published article. Only  the individual author can complete the verification step; PLOS staff cannot  verify ORCID iDs on behalf of authors.

We look forward to receiving your revised manuscript.

Kind regards,

Alejandro Torrado Pacheco, PhD

Associate Editor

PLOS One

Journal Requirements:

If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #2: All comments have been addressed

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions??>

Reviewer #2: Yes

Reviewer #3: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously? -->?>

Reviewer #2: I Don't Know

Reviewer #3: No

**********

4. Have the authors made all data underlying the findings in their manuscript fully available??>

The PLOS Data policy

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English??>

Reviewer #2: Yes

Reviewer #3: Yes

**********

Reviewer #2: (No Response)

Reviewer #3: This retrospective study of 227 patients with differentiated thyroid cancer assessed predictors of response to first radioactive iodine treatment. The sTg/TSH ratio and N1b lymph node status were not independent predictors, while poorer outcomes were associated with male sex, higher recurrence risk, stage II disease, and elevated stimulated thyroglobulin (≥9.4 ng/mL). The sTg/TSH ratio performed similarly to sTg alone, offering no added benefit, indicating that sTg remains the more useful predictor.

Minor Suggested Revisions (New Concerns):

1- Add a formal sample size/power justification: The manuscript does not describe how the sample size was determined or whether the study was adequately powered to detect meaningful effects, which limits confidence in negative findings.

2- Cite the statistical software: Specify the statistical software (e.g., SAS, R, SPSS) used for the analysis to improve reproducibility and transparency.

3- Include model diagnostics: Additional details on model evaluation, such as assessment of multicollinearity (e.g., variance inflation factors) and goodness-of-fit, should be provided to support the validity of the regression results.

4- Clarify handling of missing data: The manuscript should explicitly state how missing data were handled (e.g., exclusion vs imputation), as this can influence both estimates and generalizability.

5- Provide events-per-variable justification: Report the number of outcome events and comment on the events-per-variable ratio to ensure the multivariable model is not overfit and results are stable.

6- Table 1: For the cervical node metastasis location and sTg/TSH ratio subgroups, first report an overall p-value testing for differences across all groups, followed by appropriate stepdown (post hoc) comparisons to identify which specific group pairs differ.

Prior Reviewer Comments Not Fully Addressed:

1- Subgroup analysis rationale remains weak: The explanation for subgroup construction is still ad hoc and lacks a clear statistical or prespecified justification.

2- Handling of extreme/censored values is unclear: Although some exclusions are described, the analytical approach for values below/above assay limits is not fully specified and may introduce bias.

3- ROC analysis lacks formal comparison: The added discussion is descriptive; no statistical comparison of AUCs or model performance is provided.

4- Confounding by indication not addressed analytically: The issue is acknowledged, but no statistical adjustment (e.g., stratification or propensity methods) was performed.

5- Limited novelty only reframed: The conclusions were revised for alignment, but the underlying concern about limited incremental contribution remains.

**********

what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review?  For information about this choice, including consent withdrawal, please see our Privacy Policy

Reviewer #2: Yes: Matin mojaveri samak

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures

You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation.

NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications.

PLoS One. 2026 Aug 17;21(8):e0349850. doi: 10.1371/journal.pone.0349850.r004

Author response to Decision Letter 2

18 Jul 2026

Reviewer #3

--------------------------------------------------------------------------------------

#Add a formal sample size/power justification: The manuscript does not describe how the sample size was determined or whether the study was adequately powered to detect meaningful effects, which limits confidence in negative findings.

Response: Revised as recommended.

Action:

1. Added a dedicated “Sample size considerations” subsection in the Methods section.

2. Power calculation was performed.

Initial MS: The population included in the analysis was 227 subjects.

Revised MS:

[Methods]

Sample size considerations

All patient records that are available in the research data collection period which met the aforementioned inclusion and exclusion criteria were included in the analysis. To determine the adequacy of the final sample size, power calculation was done using the R package, WebPower. The statistical power was determined based on the difference of the prevalence of non-excellent response in cases with a positive versus negative predictor of interest, and a significance level of 0.05.

[Results]

The population included in the analysis was 227 subjects, among which 153 (67.4%) had non-excellent response outcome. Among patients with N1b node status the non-excellent outcome was 82%, as compared with 60% in those without N1b lymph nodes. This resulted in a statistical power of approximately 89%.

Location of revision: Lines 165-171, and 172-175

--------------------------------------------------------------------------------------

#Cite the statistical software: Specify the statistical software (e.g., SAS, R, SPSS) used for the analysis to improve reproducibility and transparency.

Response: Revised as recommended.

Action: Add statistical software in the method section.

Initial MS: -

Revised MS:

Descriptive statistics and logistic regression were performed using STATA version 18 (StataCorp LLC, TX, USA). ROC analysis, which was analyzed using R Statistical Software (version 4.2.1; R Core Team), RStudio version 2023.12.1+402 "Ocean Storm" (RStudio, Boston, MA) and the R package, pROC. Power calculation was done using the R package, WebPower.

Location of revision: Lines 161-164

--------------------------------------------------------------------------------------

#Additional details on model evaluation, such as assessment of multicollinearity (e.g., variance inflation factors) and goodness-of-fit, should be provided to support the validity of the regression results.

Response: Revised as recommended.

Action: Add parameters regarding the logistic regression analysis in the result section

Initial MS: -

Revised MS:

Multivariable logistic regression analysis was performed on 227 participants. Model performance was evaluated using the Akaike Information Criterion (AIC = 223.6) and Hosmer-Lemeshow goodness-of-fit test (χ²(8) = 8.62, p = 0.376), confirming adequate model fit.

Location of revision: Lines 187-189

--------------------------------------------------------------------------------------

#The manuscript should explicitly state how missing data were handled (e.g., exclusion vs imputation), as this can influence both estimates and generalizability.

Response:

1. If the missing data was important primary investigated factors such as sTg, TSH and TgAb, no SPECT/CT of the neck with the post-treatment whole body scan (WBS), unclear staging or risk of recurrence, missing data of treatment response evaluation), The patient would be excluded.

2. Regarding missing of other secondary variables of interest, overall missing data rare since almost all cases had complete medical records regarding the important clinical factors. Referring to table 1 (logistic regression results), the only variable with missing data is the ETE status, due to a lack of specific reporting the pathology results. For this variable, we labeled the cases where ETE status was not specifically mentioned in the pathology report as ‘unknown’ when performing logistic regression. The results have been added to table 1.

Initial MS: In the Methods section, exclusion of cases with missing outcome variable or main predictor variable has already been mentioned in lines 91-97.

Revised MS: [Added analysis of results of cases where the ETE status is ‘unknown’]

Location of revision: Table 1

--------------------------------------------------------------------------------------

#Provide events-per-variable justification: Report the number of outcome events and comment on the events-per-variable ratio to ensure the multivariable model is not overfit and results are stable.

Response: A total of 5 variables went into the multivariable logistic regression, using the Peduzzi's Rule of 10 events per variable, at least 50 events were needed for reliable results. In our study, there are a total of 153 events, therefore adequate for the analysis.

Initial MS: -

Revised MS:

[Results]

Since a total of 5 variables were included in the multivariable logistic regression analysis, at least 50 events were needed according to the 10-events per variable rule. In this study, there are a total of 153 events, which is adequate for a reliable regression model.

Location of revision: 192-194

--------------------------------------------------------------------------------------

#Table 1: For the cervical node metastasis location and sTg/TSH ratio subgroups, first report an overall p-value testing for differences across all groups, followed by appropriate stepdown (post hoc) comparisons to identify which specific group pairs differ.\

Response: We thank the reviewer for this comment. This aspect was omitted as the relevant parameters did not reach statistical significance on logistic regression analysis which was primary analysis.

--------------------------------------------------------------------------------------

#Subgroup analysis rationale remains weak: The explanation for subgroup construction is still ad hoc and lacks a clear statistical or prespecified justification.

And

#Handling of extreme/censored values is unclear: Although some exclusions are described, the analytical approach for values below/above assay limits is not fully specified and may introduce bias.

Response: Revised as recommended.

Action: We added a more detailed explanation of this concern in the Methods section.

Initial MS:

In cases where a variable was reported as exceeding the upper limit or falling below the lower limit of reporting in laboratory tests, such as sTg <0.04, >500 ng/mL or TSH >100 µIU/mL, it was impossible to calculate the true numerical value of sTg/TSH ratio. The analysis would be performed in two ways; by excluding the data from the population group whose variable values exceeding the upper limit or falling below the lower limit i.e. sTg <0.04, >500 ng/mL or TSH >100 µIU/mL and by dividing into various subgroups. There were total seven subgroups with different combinations of sTg/TSH ratio and TSH as follows: subgroup 1, sTg/TSH ratio < mean or median; subgroup 2, sTg/TSH ratio ≥ mean or median; subgroup 3, TSH having a reportable value and sTg <0.04 ng/mL; subgroup 4, TSH having a reportable value and sTg >500 ng/mL; subgroup 5, TSH >100 µIU/mL and sTg <0.04 ng/mL; subgroup 6, TSH >100 µIU/mL and sTg >500 ng/mL; and subgroup 7, TSH >100 µIU/mL and sTg having a reportable value. There were combinations of these subgroups in the analysis; combination of subgroups 1, 3, 5 and 7 and combination of subgroups 2, 4 and 6 based on the rationale that sTg/TSH ratio might tend to be lower and higher in each combination, respectively.

Revised MS:

In cases where extreme values were present, for example variables reported as exceeding the upper limit or falling below the lower limit in laboratory tests, such as sTg <0.04 ng/mL, sTg >500 ng/mL, or TSH >100 µIU/mL, it was impossible to calculate the true numerical value of the sTg/TSH ratio. This was not uncommon in real-world clinical practice. The analysis was therefore performed in two ways: first, by excluding data from patients whose variable values exceeded the upper limit or fell below the lower limit of detection, i.e., sTg <0.04 ng/mL, sTg >500 ng/mL, or TSH >100 µIU/mL; and second, by dividing the population into subgroups to cover all potential combinations of sTg and TSH results.

A total of seven subgroups were defined based on different combinations of sTg, TSH, and the sTg/TSH ratio, as follows. Subgroup 1 comprised patients with both sTg and TSH within reportable ranges and sTg/TSH ratio < mean or median. Subgroup 2 comprised patients with both sTg and TSH within reportable ranges and sTg/TSH ratio ≥ mean or median. Subgroup 3 included patients with TSH within a reportable range and sTg <0.04 ng/mL. Subgroup 4 included patients with TSH within a reportable range and sTg >500 ng/mL. Subgroup 5 included patients with TSH >100 µIU/mL and sTg <0.04 ng/mL. Subgroup 6 included patients with TSH >100 µIU/mL and sTg >500 ng/mL. Subgroup 7 included patients with TSH >100 µIU/mL and sTg within a reportable range.

Each subgroup was analyzed independently using logistic regression. The subgroups were also further combined into two broader groups based on clinical perspective: a potential low Tg and low sTg/TSH ratio group, comprising Subgroups 1, 3, 5, and 7, and a potential high Tg and high sTg/TSH ratio group, comprising Subgroups 2, 4, and 6. These two broader groups were also analyzed by logistic regression.

Location of revision: Line 140-160

--------------------------------------------------------------------------------------

#ROC analysis lacks formal comparison: The added discussion is descriptive; no statistical comparison of AUCs or model performance is provided

Response: We used DeLong’s test to compare the AUCs of the ROC curves which yielded a p-value of 0.729 indicating no significant difference.

Initial MS: -

Revised MS:

DeLong’s test yielded a p-value of 0.729 which indicates no significant difference between the area under the curves of the two ROCs.

Location of revision: Line 203-204

--------------------------------------------------------------------------------------

#Confounding by indication not addressed analytically: The issue is acknowledged, but no statistical adjustment (e.g., stratification or propensity methods) was performed.

#Limited novelty only reframed:

Response: We carefully considered the issues raised in this comment and we believe that these issues do not apply to this specific study, because they are biases specific to observational studies of effect of an exposure (or treatment) on an outcome. “Confounding by indication” is when a confounding variable affects both the exposure and the outcome of interest, cf. https://catalogofbias.org/biases/confounding-by-indication/. Our study simply studies the prognostic implications of clinical factors on the successfulness of I-131 treatment prescribed according to standard medical care, and therefore is not affected by this bias. Propensity methods such as propensity score matching are used to reduce effects of confounding factors in observational studies of treatment effects by adjusting for the probability of treatment assignment. It’s a method used to ‘emulate’ a randomized controlled trial using observational data, therefore this method is not relevant to this current study. Regarding dealing with confounders, the multivariable regression which was already done should be sufficient.

--------------------------------------------------------------------------------------

Attachment

Submitted filename: DT_R2_Authors_response_to_reviewer_comments.docx

pone.0349850.s006.docx (2.7MB, docx)

Decision Letter 2

Antimo Migliaccio

5 Aug 2026

The effects of stimulated thyroglobulin/ thyroid-stimulating hormone ratio and N1b status on the efficacy of the first I-131 treatment in patients with differentiated thyroid cancer

PONE-D-25-59175R2

Dear Dr. Theerakulpisut,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Antimo Migliaccio, M.D.

Academic Editor

PLOS One

**********

Acceptance letter

Antimo Migliaccio

PONE-D-25-59175R2

PLOS One

Dear Dr. Theerakulpisut,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Antimo Migliaccio

Academic Editor

PLOS One

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Table. Detailed clinical characteristics of the cohort.

    (DOCX)

    pone.0349850.s001.docx (23.1KB, docx)
    S2 Table. Detailed results from logistic regression analysis for factors associated with non-excellent response outcome after I-131 treatment.

    NA; not applicable, Ref.; reference.

    (DOCX)

    pone.0349850.s002.docx (41.2KB, docx)
    S3 Dataset. Anonymized dataset.

    (XLSX)

    pone.0349850.s003.xlsx (38.8KB, xlsx)
    Attachment

    Submitted filename: Response to reviewers.docx

    pone.0349850.s005.docx (2.7MB, docx)
    Attachment

    Submitted filename: DT_R2_Authors_response_to_reviewer_comments.docx

    pone.0349850.s006.docx (2.7MB, docx)

    Data Availability Statement

    All relevant data are within the paper and its Supporting Information files.

    Articles from PLOS One are provided here courtesy of PLOS

    RESOURCES