Disease status stratification individualizes radioiodine therapy for differentiated thyroid cancer: a prospective, multicenter, real-world study

Abstract

Background

Recurrence risk stratification system-guided radioiodine (¹³¹I) therapy of differentiated thyroid cancer (DTC) has fallen into dilemma due to its poor feasibility and insufficient justification.

Methods

Subjects referred for ¹³¹I therapy were consecutively labeled with no evidence of disease, unexplained hyperthyroglobulinemia, or known disease based on stimulated serum thyroglobulin, radioiodine uptake, and medical imaging, and then assigned to radioiodine remnant ablation (RRA), radioiodine adjuvant treatment (RAT), or radioiodine oncolytic treatment (ROT) group, respectively. The primary endpoints were disease-free survival (DFS) for the RRA and RAT cohorts and progression-free survival (PFS) and overall survival (OS) for the ROT cohort. Secondary endpoints included success rate in the RRA cohort, response classification in all cohorts, and biochemical and structural responses in the ROT cohort.

Results

In the RRA and RAT cohorts, the 5-year DFS rates were 93.28% and 76.02%, respectively. In the ROT cohort, the 5-year PFS and OS rates were 72.23% and 99.01%, respectively. In the RRA cohort, an 85.06% success rate was achieved, with excellent response (ER), indeterminate response (IR), biochemical incomplete response (BIR) and structural/functional incomplete response (S/FIR) rates of 77.92%, 15.58%, 3.9%, and 2.6%, respectively. In RAT cohort, ER, IR, BIR and S/FIR rates were 16.42%, 26.87%, 33.83%, and 22.89%, respectively. In the ROT cohort, ER, IR, BIR, and S/FIR rates were 2.5%, 3.33%, 8.33%, and 85.83%, respectively, with biochemical remission rate of 69.49% and structural disease control rate of 96.15%.

Conclusion

This prospective, multicenter, real-world study justified postoperative disease status stratification as a feasible guidance for individualizing ¹³¹I therapy of DTC.

Introduction

Thyroid cancer is among the most common malignancies with a steadily and rapidly increasing global incidence. In 2020, there were approximately 586,000 new cases worldwide, ranking it 9th place in cancer incidence [1]. Differentiated thyroid cancer (DTC) accounts for > 90% of all thyroid cancers, pathologically including papillary thyroid cancer, follicular thyroid cancer, and oncocytic cell cancer [2, 3]. Taking advantages of the nature of DTC and the properties of radioiodine (¹³¹I), ¹³¹I therapy combined with levothyroxine therapy following thyroidectomy has become a mainstay procedure to further reduce tumor-related recurrence and mortality in comparison with levothyroxine therapy alone [4].

As a broad term, ¹³¹I therapy for DTC encompasses three independent or cooperative primary approaches: (i) radioiodine remnant ablation (RRA) to facilitate response classification post initial treatment and dynamic recurrence risk stratification; (ii) radioiodine adjuvant treatment (RAT) of potential residual thyroid cancer to decrease recurrence and improve disease-free survival; (iii) radioiodine oncolytic treatment (ROT) of known disease with either curative or palliative intent to enhance progression-free survival and overall survival [2, 5–7]. In fact, ¹³¹I therapy evolves over time, moving from a standardized practice to a tailored approach. “One-size-fits-all” approach had been abandoned owing to either insufficiency or abuse of ¹³¹I therapy [8, 9]. In the year 2009, a postoperative recurrence risk stratification system was initially developed by the American Thyroid Association, which was revised with proposed modifications in the 2015 version [2, 10], striving for defining indications of ¹³¹I therapy by providing predictive and prognostic information. However, such retrospective data-derived recurrence risk stratification models, which were mainly based on static pathological findings, have been recognized subsequently as lack of theranostic value [11, 12]. This is probably due to their technical difficulties, insufficiently proven rationality and inability to explicitly assign appropriate candidates to RRA, RAT or ROT. Moreover, the systematic use of ¹³¹I therapy in patients with low-intermediate recurrence risk remains largely controversial, owing to the relatively heterogeneous population enrolled and conflicting data available [13–15]. Therefore, high-quality studies to establish a more viable stratification system to individualize ¹³¹I therapy for DTC and elucidate its efficacy are urgently needed, additionally considering the outcomes of ¹³¹I therapy based on the risk stratification systems remain unavailable to date.

Fortunately, several tools have been readily utilized to achieve reliable postoperative disease status stratification (DSS) in patients with DTC, such as stimulated serum thyroglobulin via levothyroxine withdrawal (Tg_off), neck ultrasound, and radioactive iodine uptake (RAIU) [16–18]. Besides, other medical imaging modalities, such as computed tomography (CT), magnetic resonance imaging, and positron emission tomography (PET)/CT, have made it possible to more precisely label postoperative patients with real-time disease status, paving the way to individualized ¹³¹I therapy in the era of precision medicine [7, 11, 19, 20].

We, therefore, conceived this prospective, multicenter, real-world study to assess the feasibility of a postoperative DSS system in guiding ¹³¹I therapy for DTC.

Patients and methods

Study populations

All patients with DTC aged 18–70 years, who were referred for ¹³¹I therapy after total or near-total thyroidectomy with or without lymphadenectomy, were consecutively and prospectively enrolled at five tertiary care institutions from 1 January 2012 to 1 January 2018, with follow-up until 1 January 2021. The exclusion criteria included anti-Tg antibody (TgAb) over 100 IU/mL, consent withdrawal, and concomitant malignancy [5, 17].

The study was approved by the Ethics Boards of Shanghai Jiao Tong University Affiliated Sixth People’s Hospital and Ren Ji Hospital, Tenth People’s Hospital of Tongji University, the Affiliated Suzhou Science & Technology Town Hospital of Nanjing Medical University, and Affiliated Hospital of Jiangnan University. All subjects provided written informed consent. The authors vouch for the completeness and accuracy of the data and analyses.

Study design

The primary endpoints included disease-free survival (time interval between ¹³¹I administration and structural recurrence of disease) in the RRA and RAT cohorts, progression-free survival (time interval between initial ¹³¹I administration and structural progression of disease) and disease specific survival (time interval between initial ¹³¹I administration and death from DTC) in the ROT cohort. Secondary endpoints included success rate in the RRA cohort, therapeutic response classification in all the three cohorts, and biochemical and structural responses in the ROT cohort.

Disease status stratification

Since recombinant human thyroid stimulating hormone (rh-TSH) remained unavailable in China, postoperative DSS was performed at four weeks after levothyroxine withdrawal, a classic approach to increase internal TSH level for all candidates of ¹³¹I therapy, and a low-iodine diet, with routine consideration of medical history, Tg_off, RAIU (24 h post oral administration of 185 KBq of ¹³¹I), neck ultrasound, and chest planar CT findings. Magnetic resonance imaging or 2-deoxy-2-[fluorine-18]fluoro-D-glucose (¹⁸F-FDG) PET/CT was utilized in selected subjects with suspicious metastases beyond neck and chest.

Based on the above laboratory and imaging findings, subjects were deemed to have no evidence of disease, unexplained hyperthyroglobulinemia (Tg_off > 10 ng/mL without known disease), or known disease, and then received RRA, RAT, or ROT, respectively, regardless of postoperative risk stratification [21].

¹³¹I therapy

RRA was conducted utilizing RAIU&Tg_off-guided gradient dosimetry, and patients were given 1.1, 1.85, 3.7, or 5.55 GBq of ¹³¹I based on the higher result of either the radioiodine uptake (≤ 2%, 2–5%, 5–15%, and > 5%) or stimulated Tg (≤ 2 ng/mL, 2–5 ng/mL, 5–10 ng/mL , and > 10 ng/mL ) [17]. RAT for unexplained hyperthyroglobulinemia was performed using 5.55 GBq of ¹³¹I as previously reported [5]. In ROT, 5.55 GBq of ¹³¹I was given only if local diseases were identified; when distant metastases were identified, 7.4 GBq of ¹³¹I was prescribed.

At three days after oral administration of therapeutic ¹³¹I, a planar whole-body scan (R_XWBS) was routinely performed alone or in combination with single photon emission computed tomography (SPECT)/CT if the RxWBS provided inconclusive findings [22]. ROT was terminated if radioiodine-refractory DTC (RR-DTC) was confirmed as either of the follows: (i) all the foci did not concentrate ¹³¹I per R_XWBS; (ii) despite evidence of ¹³¹I concentration, neither biochemical nor structural remission could be achieved [23, 24].

Biochemical evaluations

Serologic examinations, including serum TSH, Tg, and TgAb levels, were simultaneously performed just before and 1, 4, and 12 months after ¹³¹I administration, and then every 6–12 months, unless otherwise indicated, using electrochemiluminescence immunoassay on a Cobas analyzer (Roche Diagnostics Gmbh, Roche Ltd., Basel, Switzerland) in all centers. The lower and upper detection limits of the TSH assay were 0.005 and 100 mIU/L, respectively, and TSH levels lower or higher than those were recorded as 0.005 and 100 mIU/L, respectively. Similarly, Tg levels lower or higher than their thresholds were recorded as 0.04 ng/mL and 25,000 ng/mL (50-fold dilution), respectively; TgAb levels lower or higher than thresholds were counted as 10 and 115 IU/mL, respectively.

For the ROT cohort, biochemical responses were categorized as follows as previously described: (i) biochemical remission: decrease of at least 25% in Tg_on level; (ii) biochemical progression: increase in Tg_on level of at least 25%; (iii) biochemical stabilization: changes in Tg_on level < 25% [25].

Structural assessments

Radiographic assessments were performed by competent radiologists using CT, magnetic resonance imaging, or PET/CT. For patients with target lesions, radiographic examinations were conducted every 3–6 months. To calculate progression-free survival, structural response was defined according to RECIST 1.1 as follows: complete response, disappearance of all target lesions. Any pathologic lymph nodes (target or non-target) should demonstrate a reduction in short axis to < 10 mm; partial response, ≥ 30% decrease in the sum of diameter of target lesions; progressive disease, ≥ 20% increase in the sum of diameter of target lesions or appearance of new lesions; stable disease, neither sufficient shrinkage to qualify for partial response nor sufficient increase to be eligible for progressive disease [26].

Therapeutic response classification

At 4–12 months after the initial specified ¹³¹I therapy, success rate of RRA and response classification were assessed. A successful RRA was defined as serum Tg under levothyroxine administration at proper doses (Tg_on) below 0.2 ng/mL [2]. Therapeutic response classification in all the three cohorts included excellent response (negative imaging and Tg_on < 0.2 ng/mL), indeterminate response (negative imaging and Tg_on level 0.2-1 ng/mL), biochemical incomplete response (negative imaging and Tg_on ≥ 1 ng/mL), and structural/functional incomplete response (anatomic or metabolic evidence of disease with any Tg_on level), in terms of 2015 American Thyroid Association guidelines with minor modifications. Excellent response, indeterminate response, and biochemical incomplete response were collectively termed as non-structural/functional incomplete response [5].

Statistical analyses

All analyses were conducted using SPSS version 25.0. Categorical variables are presented as numbers and percentages, and continuous variables are presented as means and standard deviations or as medians with ranges. The Kruskal-Wallis test and variance analysis were used to compare median values between and among groups, respectively. The chi-square test and Fisher’s test were used for univariate analyses as needed. Disease-free survival, progression-free survival and disease specific survival were calculated using the Kaplan-Meier survival method. A log-rank test was used to test the differences between/among time-to-event curves. Two-tailed probabilities were reported. Values of p < 0.05 indicated statistical significance.

Results

Patient features

A total of 900 patients were initially recruited; of them 271 were excluded (235 patients with TgAb over 100 mIU/L, 35 patients withdrew consents and 1 patient concomitantly carrying breast cancer). With regard to the postoperative DSS, 302, 207, and 120 eligible patients were explicitly assigned to RRA, RAT, and ROT group, respectively (Fig. 1).

Fig. 1.

Enrollment, exclusion, disease status stratification, and radioiodine dosing strategy in patients with differentiated thyroid cancer. DTC, differentiated thyroid cancer; RAIU, radioactive iodine uptake; Tg_off, stimulated thyroglobulin via levothyroxine withdrawal

The baseline characteristics of the 629 eligible patients are summarized in Table 1. The male-to-female ratio was 1:1.80, and 104 (16.53%) patients were older than 55 years at diagnosis. Papillary thyroid cancer accounted 96.18% of all subjects, and neck lymph node dissection was conducted in 89.34% of all patients. The median time interval between operation and initial ¹³¹I therapy was 4.30 months (range, 1.03-283.67 months). At the time of the data cut-off, the overall median duration of follow-up was 37.63 months (range, 6.53-102.67 months). In the ROT group, ¹⁸F-FDG PET/CT scan was performed in 27 (22.5%) patients.

Table 1.

Baseline characteristics of enrolled patients with differentiated thyroid cancer

Patient characteristic	All patents, n (%)	Patients grouped by disease status stratification-based 131I treatment, n (%)
		RRA	RAT	ROT	P value
Age (mean ± SD, year)	43.1 ± 11.8	42.2 ± 11.5	42.2 ± 10.8	45.7 ± 13.7	0.05
Sex					0.08
Female	404 (64.23)	206 (68.21)	120 (58.97)	77 (64.17)
Male	225 (35.77)	96 (31.79)	87 (42.03)	43 (35.83)
Operation type
Total thyroidectomy	591 (93.96)	274 (90.73)	205 (99.03)	112 (93.33)	< 0.01
Near-total thyroidectomy	38 (6.04)	28 (9.27)	2 (0.97)	8 (6.67)	< 0.01
Neck lymph node dissection	562 (89.34)	264 (87.42)	188 (90.82)	110 (91.67)
Histology
PTC	605 (96.18)	281 (95.36)	204 (98.55)	113 (94.17)	0.02
FTC	24 (3.82)	14 (4.64)	3 (1.45)	7 (5.83)
T stage					< 0.01
T1	439 (69.79)	240 (79.47)	154 (74.40)	45 (37.50)
T2	85 (13.51)	27 (8.94)	29 (14.01)	29 (24.17)
T3	83 (13.20)	31 (10.26)	15 (7.24)	37 (30.83)
T4	22 (3.50)	4 (1.33)	9 (4.35)	9 (7.50)
N stage					0.01
N0	68 (10.81)	41 (13.58)	11 (5.31)	16 (13.33)
N1	544 (86.49)	254 (84.11)	187 (90.34)	103 (85.84)
Nx	17 (2.70)	7 (2.31)	9 (4.35)	1 (0.83)
M stage					< 0.01
M0	562 (89.35)	301 (99.67)	206 (99.52)	55 (45.83)
M1	67 (10.65)	1 (0.33)	1 (0.48)	65 (54.17)
Initial AJCC stage					< 0.01
I - II	608 (96.66)	299 (99.01)	199 (96.14)	101 ((84.17)
III - IV	21 (3.34)	3 (0.99)	8 (3.86)	19 (15.83)
Metastases
Lung	59 (9.38)	0 (0.00)	1 (0.48)	58 (48.33)
Bone	9 (1.43)	1 (0.33)	0 (0.00)	8 (6.67)
Lymph-node	68 (10.81)	0 (0.00)	0 (0.00)	68 (56.67)
Soft tissue	2 (0.32)	0 (0.00)	0 (0.00)	2 (1.67)
Locally advanced	1 (0.16)	0 (0.00)	0 (0.00)	1 (0.83)
Recurrent risk stratification					< 0.01
Low	86 (13.67)	60 (19.87)	26 (12.56)	0 (0.00)
Intermediate	470 (74.72)	241 (79.80)	180 (86.96)	49 (40.83)
High	73 (11.61)	1 (0.33)	1 (0.48)	71 (59.17)
Median TSH after THW (range, mIU/mL)	100.00 (3.3–100)	100.00 (4.32–100)	100.00 (8.09–100)	100.00 (3.3–100)	0.872
Median Tgoff (range, ng/mL)	14.41 (0.04–17934.00)	4.73 (0.04–56.90)	31.57 (10.70-563.10)	174.55 (1.98–17934.00)	< 0.01
Median 24-h RAIU (range, %)	4.00 (0.20–47.2)	3.60 (0.50–47.2)	4.10 ( 0.20–14.8)	5.0 (0.30–30.00)	0.01

PTC Papillary thyroid cancer, FTC Follicular thyroid cancer, TSH Thyroid stimulating hormone, THW Thyroid hormone withdrawal, Tg_off Stimulated thyroglobulin, RAIU Radioactive iodine uptake, RRA Radioiodine remnant ablation, RAT Radioiodine adjuvant treatment, ROT Radioiodine oncolytic treatment

According to the recurrence risk stratification recommended by American Thyroid Association [2], no evidence of disease and unexplained hyperthyroglobulinemia were identified in 60 (69.77%) and 26 (30.23%) of the low-risk patients, respectively. In all 470 intermediate-risk patients, no evidence of disease, unexplained hyperthyroglobulinemia, or structural disease were identified in 241 (51.28%), 180 (38.30%), and 49 (10.43%), respectively. In a total of 73 high-risk patients, 71 (97.26%) patients were assigned to ROT owing to the identification of structural disease. Notably, one high-risk patient was assigned to RRA and another one was assigned to RAT, owing to postoperative DSS. Significant differences in the distribution of recurrence risk and median Tg_off level among the RRA, RAT, and ROT groups were noted (both p values < 0.01) (Table 1).

Outcomes of RRA

According to the RAIU&Tg_off-guided gradient dosimetry, 1.1, 1.85, 3.7, and 5.55 GBq of ¹³¹I were prescribed for 68, 59, 140 and 35 patients, respectively.

Eleven (3.64%) patients had recurrence, and all were alive at the end of a median follow-up of 33.90 months (range, 6.53-102.17 months). The median disease-free survival was not reached at data cut-off (Fig. 2A). Five-year disease-free survival and disease specific survival rates were 92.97% and 100%, respectively. The subgroup analysis showed significant difference in disease-free survival among subgroups regarding therapeutic response (Fig. 2B), but no significant difference among/between subgroups based on administered ¹³¹I activity (Fig. 2C), age (Fig. 2D), and sex (not shown).

Fig. 2.

Kaplan–Meier estimates of disease-free survival (DFS) in radioiodine remnant ablation. Total group analysis (Panel A) and subgroup analyses according to therapeutic response (Panel B), RAIU&Tg_off-guided gradient activity (Panel C), and age (Panel D). RAIU, radioactive iodine uptake; Tg_off, stimulated thyroglobulin via levothyroxine withdrawal; ER, excellent response; IR, indeterminate response; BIR, biochemical incomplete response; SIR, structural incomplete response; NR, not reached

The success rate of RRA in this group was 85.76%, with median Tg_on level of 0.06 ng/mL (range, 0.04–24.21 ng/mL), which was evaluated at a median follow-up of 6.3 months (range, 4.03–12.43 months). The subgroup analysis revealed success rates of RRA of 100.00%, 93.22%, 75.71%, and 85.71% in the 1.1, 1.85, 3.7, and 5.55 GBq subgroups, respectively. Notably, a total of 15 (4.97%) patients were diagnosed with persistent disease by initial RxWBS ± SPECT/CT (Table 2).

Table 2.

Outcome analysis of subjects underwent radioiodine remnant ablation using RAIU&Tg_off-guided gradient dosimetry

Characteristic	Patients grouped by RAIU&Tgoff-guided gradient dosimetry (GBq), n (%)
	1.1	1.85	3.7	5.5	P value
Age (mean ± SD, years)	41.84 ± 11.38	44.60 ± 12.85	42.74 ± 11.38	41.53 ± 10.43	0.66
Sex					< 0.01
Female	52 (76.47)	36 (61.02)	97 (69.29)	22 (62.86)
Male	16 (23.59)	23 (38.98)	43 (30.71)	13 (37.14)
Pathology					0.84
PTC	66 (97.06)	56 (94.92)	132 (94.29)	34 (97.14)
FTC	2 (2.94)	3 (5.08)	8 (5.71)	1 (2.86)
T stage					0.06
T1-2	62 (91.18)	51 (86.44)	127 (90.71)	26 (74.29)
T3-4	6 (8.82)	8 (13.56)	13 (9.29)	9 (25.71)
N stage					0.37
N0	9 (13.24)	10 (16.95)	17 (12.14)	5 (14.29)
N1	55 (80.88)	49 (83.05)	121 (86.43)	29 (82.85)
Nx	4 (5.88)	0 (0.00)	2 (1.43)	1 (2.86)
M stage					1.00
M0	68 (100.00)	59 (100.00)	139 (99.29)	35 (100.00)
M1	0 (0.00)	0 (0.00)	1 (0.71)	0 (0.00)
Initial AJCC stage					0.51
I - II	68 (100.00)	59 (100.00)	138 (98.57)	34 (97.14)
III - IV	0 (0.00)	0 (0.00)	2 (1.43)	1 (2.86)
Recurrent risk stratification					0.60
Low	14 (20.59)	8 (13.56)	33 (23.57)	5 (14.29)
Intermediate	54 (79.41)	51 (86.44)	107 (76.43)	30 (85.71)
high	0 (0.00)	0 (0.00)	1 (0.71)	0 (0.00)
Median TSH after THW (range, mIU/mL)	100.00 (34.74–100.00)	100.00 (40.40–100.00)	100.00 (34.88–100.00)	79.06 (4.32–100.00)	< 0.01
Median Tgoff (range, ng/mL)	0.70 (0.04–2.23)	3.04 (0.04–5.12)	6.71 (0.27–14.63)	27.34 (0.17–56.90)	< 0.01
Median 24-h RAIU (range, %)	1.60 (0.50–1.30)	3.00 (1.00-5.08)	5.20 (0.50–14.90)	21.90 (9.2–47.2)	< 0.01
Findings of RxWBS ± SPECT/CT
No 131I uptake	2 (2.94)	0 (0.00)	0 (0.00)	0 (0.00)
131I uptake in thyroid bed	66 (97.06)	59 (100.00)	140 (100.00)	35 (100.00)
131I uptake in lymph node	1 (1.47)	3 (5.08)	10 (7.14)	0 (0.00)
131I uptake in lung	0 (0.00)	0 (0.00)	1 (0.71)	0 (0.00)
Successful RRA	68 (100.00)	55 (93.22)	106 (75.71)	30 (85.71)	< 0.01
Median Tgon after RRA (range, ng/mL)	0.04 (0.04–0.63)	0.04 (0.04–0.51)	0.13 (0.04–6.25)	0.15 (0.04–24.21)	< 0.01
Metastasis identified in follow-up	1 (1.47)	0 (0.00)	2 (1.43)	4 (9.76)	0.01
Retreatment with 131I	1 (1.47)	4 (6.78)	13 (9.29)	5 (14.29)	0.06

PTC Papillary thyroid cancer, FTC Follicular thyroid cancer, TSH Thyroid stimulating hormone, THW Thyroid hormone withdrawal, Tg_off Stimulated thyroglobulin, Tg_on Suppressed thyroglobulin, RAIU Radioactive iodine uptake, RxWBS Post therapeutic whole body scan, RRA Radioiodine remnant ablation

Moreover, excellent response, indeterminate response, biochemical incomplete response, and structural/functional incomplete response in the entire RRA group were achieved in 239 (79.14%), 47 (15.56%), 8 (2.65%), and 8 (2.65%) patients, respectively (Fig. 3A). In detail, an excellent response rate of 98.53% (67/68) was achieved in the 1.1 GBq subgroup, 93.22% (55/59) in the 1.85 GBq subgroup, 65.71% (92/140) in the 3.7 GBq subgroup, and 71.43% (25/35) in the 5.55 GBq subgroup. Indeterminate response rates of 1.47%, 6.78%, 26.43%, and 14.29% were obtained in the above subgroups, respectively. Biochemical incomplete response was only found in subgroups of 3.7 and 5.55 GBq, with rates of 3.57% and 14.29%, respectively. Of note, 4.29% (6/140) of patients in the 3.7 GBq subgroup and 5.71% (2/35) in the 5.55 GBq subgroup were assessed as structural/functional incomplete response (Fig. 3B).

Fig. 3.

Therapeutic response to radioiodine remnant ablation. Entire group (Panel A) and subgroups stratified by RAIU&Tg_off-guided gradient dosimetry (Panel B). RAIU, radioactive iodine uptake; Tg_off, stimulated thyroglobulin via levothyroxine withdrawal

Outcomes of RAT

During a median follow-up of 40.43 months (range, 6.93-100.53 months), 35 (16.91%) patients had disease recurrence, but none died. The median disease-free survival in RAT cohort was 85.00 months (95% CI, 77.71–92.29 months), with 5-year disease-free survival and disease specific survival rates of 77.01% and 100%, respectively (Fig. 4A). A median disease-free survival was obtained only in biochemical incomplete response subgroup (85.83 months) (Fig. 4B). Moreover, no statistically significant difference in disease-free survival was found between the subgroups in term of age (Fig. 4C) or sex (Fig. 4D).

Fig. 4.

Kaplan–Meier estimates of disease-free survival (DFS) in radioiodine adjuvant treatment. Total group analysis (Panel A) and subgroup analyses according to therapeutic response (Panel B), age (Panel C), and sex (Panel D). ER, excellent response; IR, indeterminate response; BIR, biochemical incomplete response; SIR, structural incomplete response; NR, not reached

After the initial administration of ¹³¹I in the RAT cohort, metastases in lymph nodes, bones, and lungs were revealed through R_XWBS with or without SPECT/CT in 35, 3, and 1 patients, respectively. Subsequently, excellent response, indeterminate response, biochemical incomplete response, and structural/functional incomplete response in the RAT cohort were achieved in 34 (16.43%), 55 (26.57%), 72 (34.78%), and 46 (22.22%) patients, respectively.

Outcomes of ROT

Lymph nodes accounted for the majority (56.67%) of the organs affected by metastases, followed by lungs (48.33%) and bones (6.67%) (Table 1). After a median total course number of 2 (range, 1–7) with a median cumulative activity of 11.1 GBq (range, 5.55–51.8 GBq), a total of 83 (69.17%) patients were finally defined as having RR-DTC. In detail, no ¹³¹I-avid lesion was identified in 67 (80.72%) patients, and neither biochemical nor structural remission was obtained in the remaining 16 (19.28%) patients with ¹³¹I-avid lesions. During a median follow-up of 48.60 months (range, 6.53-102.67 months) for the 83 patients with RR-DTC, five (6.02%) patients underwent additional resections for neck lesions, and seven (8.43%) patients received tyrosine kinase inhibitors for symptomatic or progressive disease (Fig. 5).

Fig. 5.

A male patient with symptomatic and progressive metastatic radioiodine-refractory papillary thyroid cancer carrying NCOA4-RET rearrangement treated with pralsetinib. ¹⁸F-FDG PET/CT images before (Panel A, maximum intensity projection; Panel B, neck fusion image; Panel C, chest fusion image; Panel D, abdomen fusion image) and 4 months after (Panel E, maximum intensity projection; Panel F, neck fusion image; Panel G, chest fusion image; Panel H, abdomen fusion image) treatment illustrated robust therapeutic response, consistent with the relief of dysphagia, hemoptysis, and pain. Repeated pathological examinations recognized high mitotic activity (Panel I, arrows) and necrosis (Panel J, circle), reclassifying the disease as high grade differentiated thyroid cancer

During a median follow-up of 45.40 months (range, 6.53-102.67 months) for ROT cohort, 27 patients had disease progression, and two patients died. The median progression-free survival was 86.00 months (Fig. 6A), with a 5-year progression-free survival rate of 72.23% and a 5-year disease specific survival rate of 99.01%. Better progression-free survival was noted in ¹³¹I-responders (Fig. 6B), younger patients (Fig. 6C) and those with loco regional disease (Fig. 6D), compared to the counterparts. Nevertheless, no difference in progression-free survival was found between the male and female subgroups (Fig. 6E) or between the structural/functional incomplete response and non-structural/functional incomplete response subgroups (Fig. 6F).

Fig. 6.

Kaplan–Meier estimates of progression-free survival (PFS) in radioiodine oncolytic treatment. Total group analysis (Panel A) and subgroup analyses according to ¹³¹I-avidity (Panel B), age (Panel C), location of metastasis (Panel D), sex (Panel E), and therapeutic response (Panel F). NR, not reached; S/FIR, structural/functional incomplete response

All patients in this group were eligible for analysis of the response to initial ROT. Excellent response, indeterminate response, biochemical incomplete response, and structural/functional incomplete response were achieved in three (2.50%), four (3.33%), 10 (8.33%), and 103 (85.83%) patients, respectively. The median serum Tg_on level decreased significantly from 20.11 ng/mL (range, 0.22-4,711.00 ng/mL) to 2.19 ng/mL (range, 0.04-2,213.00 ng/mL), yielding biochemical remission and stabilization rates of 69.49% and 20.34%, respectively. After excluding 42 patients without target lesions, the remaining 78 (65.00%) patients were eligible for structural response assessment. Post final ROT, complete response, partial response, and stable disease were reported in one (1.28%), two (2.56%), and 72 (92.31%) patients, respectively, yielding a structural disease control rate of 96.15%.

Discussion

Integrating medical history, Tg_off, RAIU, and medical imaging findings, we creatively proposed a postoperative DSS system to explicitly label DTC patients with no evidence of disease, unexplained hyperthyroglobulinemia, or structural disease, who were subsequently assigned to RRA, RAT, or ROT, respectively. Successful RRA, favorable therapeutic response, and excellent survival outcomes conjunctively demonstrated that the postoperative DSS system feasibly individualizes ¹³¹I therapy of DTC, representing a reliable complementation for the recurrence risk stratification systems with broad application in a universal setting. The strengths of the study include the large cohort of over 600 patients, multi-center enrollment, greater than 3 year median duration of follow up, and novelty of approach.

It is worth noting that the purpose of RRA is to facilitate therapeutic response classification and dynamic recurrence risk stratification, rather than to improve survival or decrease recurrence [2, 17, 27]. In the current study, a median serum Tg_on nadir as low as 0.06 ng/mL was reached post RRA, featuring an excellent response and definite response classification. Interestingly, the success rate of RRA (85.76%) seemed a little bit lower than that (94.7%) reported by our previous prospective, randomized, open-label, controlled trial [17]. This may be due to the interference by 4.97% patients who were finally revealed with persistent disease via initial RxWBS ± SPECT/CT. Of note, these patients were initially classified as low-intermediate recurrence risk, indicating a possible miss of timely treatment for occult disease if the decision-making had been merely guided by recurrence risk stratification systems [28, 29]. And only 4.97% of patients with persistent disease could potentially be partially attributed to the non-utilization of SPECT/CT in all patients [30]. Moreover, a success rate of 85.71% was achieved in patients with RAIU over 15% using a relatively high activity of 5.5 GBq (Table 2), indicating that RRA may act as an ideal alternative to completion thyroidectomy for patients with thyroid remnant if sufficient activities of ¹³¹I are prescribed [31]. Additionally, as expected, a meaningful 5-year disease-free survival rate of 92.97% and non-structural/functional incomplete response rate of 97.35% were reached in RRA group, which were in line with 95.6% of patients without recurrence at 3 years post RRA reported by Leboulleux S, et al [27]. Conversely, the remaining patients (2.65%) were revealed with structural/functional incomplete response, suggesting additional treatments and more intensive and frequent follow up, owing to a disease specific mortality as high as 11% [32, 33].

In the current study, the therapeutic response was analyzed in patients labeled as having unexplained hyperthyroglobulinemia who received RAT. The non-structural/functional incomplete response rate of 77.78% and structural/functional incomplete response rate of 22.22% seem comparable to those reported previously [5]. Further, we calculated a RAT-derived median disease-free survival of 85.00 months and a 5-year disease-free survival rate of 77.01% in this study. Thus, along with the findings of Piccado et al.. that RAT improved the progression-free survival and overall survival in DTC patients with Tg_on > 1 ng/mL after thyroidectomy and lymphadenectomy, unexplained hyperthyroglobulinemia might be adopted as an appropriate indication of RAT [34]. And either Tg_off or Tg_on should be regarded as a pivotal element of the postoperative DSS system. Notably, benefits in disease-free survival and disease specific survival strongly supported that RAT using 5.55 GBq of ¹³¹I for patients with unexplained hyperthyroglobulinemia is a reasonable scheme, bridging RRA and ROT.

ROT, the application of ¹³¹I to destroy persistent, recurrent, or metastatic DTC, has become a routine therapeutic strategy to improve progression-free survival and disease specific survival in patients with ¹³¹I-avid lesions [2, 6, 35]. As expected, after receiving a median cumulative activity of 11.1 GBq, biochemical remission rate of 69.49% and structural disease control rate of 96.15% were achieved. Meaningfully, the present study revealed a median progression-free survival of 86.00 months, a 5-year progression-free survival rate of 72.23%, and a 5-year disease specific survival rate of 99.01%, which are superior to those reported by previous studies [36, 37]. This may be partially explained by the higher proportion of patients with lymph node disease and younger age, who benefited more from ROT than the other groups of patients [36]. And 35% patients, who were labeled as having structural/functional disease, were identified without target lesions, indicating a lower tumor burden, which may be attributed to timely postoperative DSS and eventually contribute to the relatively higher 5-year disease specific survival rate.

Of note, 69.17% patients in the ROT cohort, who were initially labeled with structural disease by the DSS system, were finally defined as having RR-DTC, which seems a little bit higher than those reported by our team and others [17, 38]. This may be due to the modified definition of RR-DTC, which covered both patients with ¹³¹I-non-avid lesion and those with biochemically and structurally stable/progressive disease post ROT, considering the indolent nature of ¹³¹I-avid DTC. In addition, ¹⁸F-FDG PET/CT was extensively incorporated, manifesting an incremental value of nuclear molecular imaging in the management of DTC [7, 11, 19, 20]. Notably, the response to ROT seems a reliable predictor for progression-free survival, indicating that ¹³¹I-responders may have a significantly longer progression-free survival than ¹³¹I-nonresponders. On the contrary, high grade DTCs should be cautiously identified by pathological examinations and other assessments such as diagnostic and post-therapeutic ¹³¹I scans to avoid potentially ineffective ROT and timely switched to targeted therapy when indicated (Fig. 5) [3].

We admit that the current study has several limitations. Firstly, the incorporation of a diagnostic radioiodine scan and genetic profiling (BRAF, TERT, RET, etc.) prior to ¹³¹I therapy might be of value for improving this clinically practical DSS system, which merits further investigations [39, 40]. Secondly, considering the aggressive histology features (like hobnail variant of PTC) and extent of lymph node dissection might help develop appropriate RAI therapy activity. And neglecting pathological features also might reduce the number of patients receiving RAT [27]. Thirdly, due to the immaturity of non-immunometric assay and that an acceptable cut-off value of TgAb to alleviate the potential interference with Tg determination has not been well established, TgAb ≤ 100 IU/mL was adopted like before [5, 17, 41], needing further verification. Fourthly, since nearly a quarter of our patients were excluded from the current study for having TgAb > 100 IU/mL, the clinical outcomes in this entity are required to be investigated to reveal the whole landscape of ¹³¹I therapy.

Conclusions

This prospective, multicenter, real-world study demonstrated that the postoperative DSS system may act as a feasible guide to individualized ¹³¹I therapy for DTC, warranting head-to-head comparisons to other decision-making systems, such as TNM-stage, pathological subtype and recurrence risk stratification.

Abbreviations

DTC

Differentiated thyroid cancer

RRA

Radioiodine remnant ablation

RAT

Radioiodine adjuvant treatment

ROT

Radioiodine oncolytic treatment

DSS

Disease status stratification

Tg_off

Stimulated serum thyroglobulin via levothyroxine withdrawal

rh-TSH

Recombinant human thyroid stimulating hormone

TgAb

Anti-Tg antibody

RAIU

Radioactive iodine uptake

CT

Computed tomography

PET/CT

Positron emission tomography/CT

¹⁸F-FDG

2-deoxy-2-[fluorine-18]fluoro-D-glucose

R_XWBS

Whole-body scan

SPECT/CT

Single photon emission computed tomography/CT

RR-DTC

Radioiodine-refractory DTC

Authors' contributions

All authors participated in the study’s conceptualization; data collections were by all the authors; Xian Qiu and Lin Cheng participated in data analysis; Xian Qiu, Yuchen Jin and Libo Chen participated in writing the original draft; All authors edited and reviewed draft.

Funding

This study was sponsored by the National Natural Science Foundation of China (nos. 81671711 and 82171981).

Data availability

All data generated or analyzed during this study will be made available by the corresponding authors, upon reasonable request.

Declarations

Ethics approval and consent to participate

This research was conducted in accordance with the Declaration of Helsinki. This study was approved by the Ethics Committee of Shanghai Sixth People’ Hospital (Ethical Code is 2000-022) and Ren Ji Hospital, Tenth People’s Hospital of Tongji University, the Affiliated Suzhou Science & Technology Town Hospital of Nanjing Medical University, and Affiliated Hospital of Jiangnan University. All patients read and signed the informed consent form.

Consent for publication

Not applicable.

Competing interest

The authors declare no competing interests.