Keywords thyroglobulin - dynamic risk stratification - prognostic
Introduction
Papillary thyroid carcinoma (PTC) represents 85–90% of differentiated thyroid
carcinoma (DTC), and it is considered a neoplasm of indolent behavior with a 5-year
survival rate of 98.1% [1 ]. However,
persistence or recurrence is a condition related to this tumor, requiring long-term
follow-up [1 ]. The estimated risk of
persistent and/or recurrent disease varies among the cohorts studied, being between
0–14% in patients with excellent response to initial treatment [2 ]
[3 ]
[4 ].
Postoperative thyroglobulin (Tg) has an important role in indicating the need of
radioactive iodine therapy (RAI), to evaluate initial response to treatment and
define response on long-term follow-up [1 ]
[5 ]
[6 ]. The evaluation of stimulated Tg (s-Tg)
immediately before RAI (pre-RAI Tg) is used as an early marker to predict clinical
recurrence, to detect distant metastasis, and also to predict the results of
radioiodine whole-body scintigraphy (WBS) after RAI [7 ]
[8 ]. Several studies have demonstrated a wide range of pre-RAI Tg
cut-offs, between 2.8 and 38.1 ng/ml to predict persistent and/or recurrence on
follow-up, with variable sensitivity and specificity, and variable thresholds [6 ]
[8 ]
[9 ]
[10 ]. Recently, it has been proposed that
values of pre-RAI Tg≥10 ng/ml have shown better sensitivity and specificity to
predict persistent/recurrent disease and survival related to DTC [11 ]
[12 ]
[13 ].
Identifying patients with excellent response to the initial therapy enables having
less follow-ups and consequently decreasing the psychological and economic impact on
the patients and their families [1 ]
[14 ]
[15 ]. Indeed, parameters in favor of excellent response or long-term
remission are necessary to individualize the management of each patient. The first
evaluation of initial treatment response allows to guide thyrotropin (TSH) goals and
to better accommodate the subsequent long-term follow-up [1 ]
[4 ]
[14 ]
[16 ]
[17 ]. Even though several studies have tried to identify the Tg cut-off in
different points during follow-up, there is no agreement on what would be the best
value of Tg to predict initial response to therapy and/or disease progression over
time [9 ]. Considering a possible role of
pre-RAI Tg to predict initial excellent response, we expected to achieve a pre-RAI
Tg cut-off value that allows to guide a possible de-escalation in the follow-up of
DTC patients.
Therefore, the objective of this study was to determine the correlation between the
pre-RAI Tg level and the first evaluation of response to initial therapy.
Furthermore, we aimed to define a cut-off level for pre-RAI Tg as a predictor of
excellent initial response therapy, as well as assessing long-term disease status in
patients with DTC.
Patients and Methods
Study population
A retrospective cohort of 353 DTC patients who underwent total thyroidectomy
(TT), with or without lymph node dissection, followed between 2011 and 2022,
were first assessed. Data for this study were collected from the Endocrine
Division of Irmandade da Santa Casa de Misericordia de Porto Alegre, a tertiary
care, university teaching hospital in Southern Brazil. Patients were treated
under the Brazilian public health system. All patients were older than 18 years,
both sexes, and their AJCC/UICC TNM staging for metastasis were M0/Mx at
diagnosis. We excluded 24 patients who had positive antithyroglobulin antibodies
(TgAb), 9 patients that had M1, 43 patients with incomplete record files, 71
patients that were not submitted to radioiodine therapy and 40 patients who had
no stimulated thyroglobulin assessed before radioiodine therapy (Fig.
S1 ). The study was approved by the ethics committee of the institution.
Laboratory analysis
During the study period, serum Tg were measured by chemiluminescence (Immulite
2000, Siemens) and electrochemiluminescence (ElecsysTgII, COBAS Roche).
Functional sensitivities varied from 0.2 ng/ml to 0.1 ng/ml. Anti-thyroglobulin
antibodies (TgAb) were measured using the electrochemiluminescence immunoassay
and by chemiluminescence. Serum TSH levels were measured with a
chemiluminescence assay (normal range 0.55–4.78 μUI/ml).
Follow-up protocol, risk stratification, response to treatment and
outcomes
Patients submitted to TT, with or without lymphadenectomy, and RAI therapy were
included. For the evaluation of the initial risk of recurrence, all patients
were classified into low, intermediate, and high risk [1 ]. They were also staged as AJCC/UICC
TNM staging 8th edition [18 ]. The
status of the disease was defined by clinical examination, serum Tg, TgAb, and
imaging evaluation (Neck US/WBS/other) when indicated. All s-Tg measurements
were obtained after thyroid hormone withdrawal (THW) for a period of at least 4
weeks, with TSH levels>30 μIU/ml. TSH, Tg and TgAb serum measurements were
performed no more than 7 days before RAI therapy.
The response to initial therapy was evaluated between 6 to 24 months after
completing the initial therapy in agreement with 2015 ATA guidelines [1 ]; then, patients were classified into
three groups. As (i) excellent response (ER): no abnormal findings on imaging,
stimulated Tg<1 ng/ml or basal Tg<0.2 ng/ml, (ii) Indeterminate response
(IndR) – nonspecific finding on imaging and stimulated Tg≥1.0 to<10 ng/ml, or
basal Tg≥0.2 to<1.0 ng/ml, or stable/declining TgAb over time. (iii)
Incomplete response (IncR): biochemical incomplete response – negative imaging
and stimulated Tg≥10 ng/ml, or basal Tg≥1 ng/ml, or rising TbAb over time, and
structural incomplete response – abnormal finding in imaging. These last two
responses were analyzed as a single group together, despite knowing that
biochemical incomplete patients perform better outcomes in the long term than
structural incomplete response patients, because both groups have a similar
pattern of follow-up.
Follow-ups with Tg, TgAb, TSH were assessed every 6 to 12 months, according to
disease status. Neck US and/or WBS were also performed to search for metastases.
Other imaging exams were performed to search for pulmonary and/or cervical
metastases in high-risk patients or those with higher Tg or ascending Tg
patterns. Structural incomplete disease was defined as positive image results,
cytology or histology, and/or unequivocal ectopic uptake on post-therapy WBS or
FDG-PET/CT. Patients with confirmed structural disease were submitted to
additional therapy according to clinical indication. The duration of follow-up
was defined as initiating at the date of surgery up to the last medical
visit.
The primary outcome of this study was excellent response after initial therapy.
Secondary outcomes were to assess the response to the therapy re-staging system
at the final follow-up.
Statistical analysis
The characteristics of our population are described as mean±standard deviation
(SD) or median and percentiles 25 and 75 (P25–75) for continuous variables and
absolute numbers and percentages for categorical variables. Categorical
variables were compared between groups by chi-square (χ2 ) using
adjusted standardized residuals. Continuous variables were compared between
groups by Kruskal–Wallis, Mann–Whitney, and ANOVA tests. Bonferroni adjustments
were used for multiple comparisons. Receiver operating characteristic (ROC)
curve (area under the curve and corresponding 95% CI) was used to obtain a
cut-off for s-Tg before RAI to predict excellent response to initial therapy.
The Youden’s index was used to choose the cut-off. For the evaluation of
excellent response, the variables with p-values<0.20 and age group (clinical
relevance) were selected, in order to calculate uni and multivariate risks. For
uni- and multivariate analysis, the sample was divided in two groups: excellent
response and incomplete response/indeterminate response. Data analysis was
performed using Statistical Package for Social Sciences (SPSS) software, version
25.0 (IBM Corp., Armonk, NY). Statistic tests were two-sided, and levels of
significance were 0.05.
Results
Baseline characteristics of DTC patients
We evaluated 166 patients with DTC who underwent TT and RAI. The mean age of the
patients was 47.6±13 years, 113 (68.1%) patients were<55 years of age, and
85.5% were women. Histologic analysis demonstrated mostly papillary thyroid
carcinoma (PTC) (88.6%), and most of the patients were classified as AJCC/UICC
TNM stage I disease (86.7%), with median tumor size of 2.0cm (1.3–3.5cm). The
ATA Risk Stratification was low in 52 patients (31.3%), intermediate in 80
patients (48.2%), and high in 34 patients (20.5%). The median radioiodine dose
was 100 mCi (30–100 mCi) and the cohort was followed for a median of 7.4 years
(3.5–10 years).
Response to initial therapy
[Table 1 ] shows the baseline and
clinical-pathological characteristics according to the initial response to
therapy assessment. From the 166 patients followed, on the first evaluation of
response to therapy the mean time was 15.8 months (±7.6 months) after initial
therapy, 103 (62%) had excellent response(ER) to therapy, 39 (23.4%) had
indeterminate response (IndR) and 24 (14.4%) had incomplete response (IncR) – of
these 11 (6.6%) patients had biochemical incomplete response and 13 (7.8%)
patients had structural incomplete response. The prevalence of female sex was
higher in the ER group when compared to IndR and IncR (91.3%; 79.5% and 70.8%,
respectively; p=0.018). AJCC/UICC TNM stage was also different between groups:
the percentage of patients with TNM stage I was 91.3% in ER, 84.6% in IndR and
70.8% in the IncR group (p=0.026). The ATA low-risk prevalence was higher in the
ER group compared to IndR and IncR groups (39.8%, 20.5% and 12.5%, respectively;
p<0.001). The median of pre-RAI Tg was 4.1 ng/ml (1.15–13.7 ng/ml). The
median pre-RAI Tg was 3.46 ng/ml (1.04–12.3 ng/ml) using chemiluminescence assay
(n=112) and it was not different from the other samples (n=54), median pre-RAI
Tg of 5.90 ng/ml (1.29–18 ng/ml) (p=0.288). Pre-RAI Tg levels were significantly
correlated with stimulated Tg (s-Tg) and basal Tg (b-Tg) measures in the initial
evaluation of therapy response, Spearman’s rho: 0.65 (p<0.001) and 0.56
(p<0.001), respectively. Pre-RAI Tg was significantly different between all
groups. The median value in patients with ER was 2.3 ng/ml (0.5–5.7 ng/ml); IndR
was 11.0 ng/ml (2.3–18.0 ng/ml), and in IncR was 37.0 ng/ml (13.5–86.2 ng/ml)
(p<0.001). Histological type, tumor size, radioiodine initial dosing, age at
diagnosis and follow-up time were also evaluated, but had no significant
difference between groups.
Table 1 Baseline clinical-pathological characteristics of
the cohort subdivided according to initial response to
therapy.
Initial response therapy
Excellent
Indeterminate
Incomplete response
n
%
n
%
n
%
p
n
103
62
39
23.4
24
14.4
Sex
Female
94
91.3
31
79.5
17
70.8
0.018
Histology
Pappilary
92
89.3
36
92.3
19
79.2
0.410
Follicular
9
8.7
2
5.1
3
12.5
Oncocytic cell Carcinoma
2
1.9
1
2.6
2
8.3
TNM stage
I
94
91.3
33
84.6
17
70.8
0.026
II–IV
9
8.7
6
15.4
7
29.2
ATA Initial risk assessment
Low
41
39.8
8
20.5
3
12.5
<0.001
Intermediate
50
48.5
22
56.4
8
33.3
High
12
11.7
9
23.1
13
54.2
Tumor size (cm)
Median|p25–75
2.0|1.2–3.2
2.0|1.3–3.4
3.0|1.8–5.1
0.063
RAI initial dosing (mCi)
30–50
33
32
11
28.2
2
8.3
0.065
100–200
70
68
28
71.8
22
91.7
Age
Median|p25–75
50.9|38–57.8
42.8|32.6–55.9
48.8|43.9–63.5
0.147
Age
<55
67
65
29
74.4
17
70.8
0.542
>55
36
35
10
25.6
7
29.2
Pre-RAI Tg (ng/ml)
Median|p25–75
2.3|0.5–5.7
11.0|2.3–18.0#
37|13.5–86.2#$
<0.001
Follow-up (years)
Median|p25–75
7.8|3.9–10
6.8|2.8–10.2
5.6|2.9–10.8
0.978
TNM: AJCC: American Joint Committee on Cancer staging 8th edition; ATA:
2015 American Thyroid Association guideline; RAI: Radioactive iodine
therapy; pre-RAI Tg: Pre-radioiodine therapy thyroglobulin. #
Significantly different from Excellent; $ Significantly
different from Indeterminate (post-hoc Bonferroni adjustments,
p<0.05).
Predictive value of stimulated thyroglobulin levels before RAI in initial
response to therapy
We demonstrated that the levels of pre-RAI Tg are significantly different between
patients according to initial response to therapy ([Table 1 ]). When we compared the
pre-RAI Tg between each group of response to therapy, we observed that the ER
group had a significantly lower pre-RAI Tg in comparison to IndR (p<0.001)
and IncR (p<0.001). In addition, there is a significant difference in pre-RAI
Tg levels between the IndR and IncR patients (p=0.02). [Fig. 1 ] shows the difference in Tg
values of pre-RAI Tg among the groups according to first dynamic risk
stratification.
Fig. 1 Violin plot of stimulated thyroglobulin before radioiodine
therapy according to response to therapy group.
Considering the correlation of pre-RAI Tg levels with the first evaluation of
response to treatment, we decided to search what would be the best cut-off that
predicts ER. To avoid the effect of different assays to calculate a specific
serum Tg value cut-off, we evaluated 112 patients in which the pre-RAI Tg was
measured by the same methodology (chemiluminescence). ROC curve analysis
demonstrated that pre-RAI Tg was predictive of excellent response. We found the
cut-off value of 7.55 ng/ml (area under the curve=0.832, 95% CI 0.76–0.91), with
sensitivity of 80.0% and specificity of 73.0%, positive predictive value (PPV)
of 85.7% and negative predictive value (NPV) of 64.3% ([Fig. 2 ]). This cut-off value was valid
for low-risk (PPV=91.9%), intermediate-risk (PPV=75.5%), and high-risk
(PPV=76.9%) when applied to all patients ([Table 3 ]).
Fig. 2 Receiver operating characteristic (ROC) curves for
differentiating ER (excellent response) from IndR (indeterminate
response) + InCR (incomplete response).
Table 3 Sensitivity, specificity, positive predictive
value (PPV), and negative predictive value (NPV) of pre-RAI Tg cut
off=7.55 ng/ml for low, intermediate, and high-risk
patients.
Cut-off 7.55 ng/ml
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
All (n=166)
78.6
71.4
81.8
67.2
Low Risk
82.9
72.7
91.9
53.3
Intermediate Risk
74.0
60.0
75.5
58.1
High Risk
83.3
86.3
76.9
90.5
Clinicopathological factors associated with initial excellent response
status
Applying univariate and multivariate logistic regression analysis, we identified
few factors as predictors of initial response to therapy. In the univariate
analysis, female sex (RR 1.77; 95% CI 1.04–3.0, p=0.03), ATA Initial risk
stratification low (RR 2.23, 95% CI 1.39–3.6, p=0.001) and intermediate (RR
1.77, 95% CI 1.09–2.88, p=0.021) and the Tg under the cut-off value of 7.55
ng/ml (RR 2.49, 95% CI 1.75–3.55, p<0.001) were all predictors of excellent
response. In the multivariate analysis, only ATA low risk (RR 1.61, 95% CI
1.06–2.43, p=0.025) and Tg below the cut-off value of 7.55 ng/ml (RR 2.17, 95%
CI 1.52–3.10, p<0.001) were associated with an excellent response to
treatment ([Table 2 ]).
Table 2 Univariate and multivariate logistic regression
analysis for initial excellent Response.
Uni and multivariate analysis
ER
IndR/IncR
Uni
Multi
n
%
n
%
p
RR
95% CI
p
RR
95% CI
Sex
Female
94
66.2
48
33.8
0.035
1.77
1.04
3.00
0.128
1.37
0.91
2.07
Male
9
37.5
15
62.5
1
1
TNM Stage
I
94
65.3
50
34.7
0.076
1.60
0.95
2.67
0.480
1.18
0.74
1.90
II–IV
9
40.9
13
59.1
1
1
ATA Initial risk assessment
Low
41
78.8
11
21.2
0.001
2.23
1.39
3.60
0.025
1.61
1.06
2.43
Intermediate
50
62.5
30
37.5
0.021
1.77
1.09
2.88
0.134
1.38
0.91
2.09
High
12
35.3
22
64.7
1
1
Pre-RAI Tg
≤7.55
81
81.8
18
18.2
0.000
2.49
1.75
3.55
0.000
2.17
1.52
3.10
>7.55
22
32.8
45
67.2
1
1
RAI initial dosing (mCi)
30–50
33
71.7
13
28.3
0.086
1.23
0.97
1.56
0.468
1.08
0.87
1.35
100–200
70
58.3
50
41.7
1
1
Age
>55
36
67.9
17
32.1
0.267
1.15
0.90
1.46
0.290
0.90
0.75
1.09
<55
67
59.3
46
40.7
1
1
Tumor size (cm)
Median (p25–75)
2.0 (1.2–3.2)
2.3 (1.5–4.0)
0.306
0.96
0.90
1.03
TNM: AJCC: American Joint Committee on Cancer staging 8th edition; ATA:
2015 American Thyroid Association guideline; pre-RAI Tg: Pre-radioiodine
therapy thyroglobulin; RAI: Radioactive iodine therapy.
Disease status of DTC patients at last evaluation
After a median follow-up of 7.4 years (3.5–10 years), 124 (74.7%) patients were
allocated in the excellent response group, 22 (13.2%) patients in the
indeterminate response group, and 20 (12%) patients in the incomplete disease
group, where 7 (4.2%) patients in biochemical incomplete response and 13 (7.8%)
patients in structural incomplete response ([Fig. 3 ]). From the initial assessment
on response to therapy, 99 (96.1%) patients remained in the excellent response
group in follow-up, and only 4 (3.9%) patients were reclassified to the
indeterminate group. Among patients with an initial indeterminate response, 13
(33.3%) remained in the same classification at last visit, 23 (59%) patients
were reclassified to excellent response; and, of those, 15 patients had no
additional therapies and 8 patients received additional lymphadenectomy and/or
RAI therapy and 3 (7.7%) patients changed status to incomplete response. Two of
these patients presented biochemical incomplete response and 1 patient had
structural incomplete disease (cervical lymph node disease). From the 24
patients with incomplete response at the initial evaluation, 17 (70.8%) patients
remained in the same classification, 2 (8.3%) patients were reclassified to the
excellent response, both of them were patients with biochemical incomplete
response (only 1 patient received additional RAI therapy) and 5 (20.8%) patients
were reclassified to indeterminate group in the last evaluation, 3 of them
received additional therapy, lymphadenectomy and/or RAI.
Fig. 3 Sankey curves of response to therapy in the first
evaluation and in long-term follow-up.
Discussion
In this study, we demonstrated that the Tg levels immediately before radioiodine
therapy (pre-RAI Tg) had a correlation with the first assessment of response to
initial therapy in DTC patients. As expected, patients with ER have lower pre-RAI Tg
than patients with IndR and IncR. Interestingly, pre-RAI Tg levels also were
significantly different between IndR and IncR groups. In clinical practice, the main
challenge is to manage patients with indeterminate response, given that most
patients remain with no evidence of disease, while up to 20% of them develop
biochemical or structural disease [1 ]
[19 ]. In our sample of patients with IndR,
92.3% of those cases were free of disease at last evaluation. These findings
reinforce that pre-RAI Tg is an important tool when making decisions regarding the
follow-up of patients with DTC.
An excellent response (ER) to therapy with its very low risk of recurrence should
lead to re-evaluation of intensity of diagnostic surveillance procedures and
treatment [1 ]. We demonstrated that a
pre-RAI Tg cut-off≤7.55 ng/ml has a very significant accuracy to predict ER to
therapy in the initial assessment. This cut-off value was reliable for patients with
low-risk and intermediate to high-risk of recurrence. In this way, other studies
have shown that values below 10ng/mL have a better NPV for no evidence of
biochemical or structural disease on long-term follow-up [8 ]
[20 ]. Recently, Tian et. al. corroborated these results that a pre-RAI
Tg≤10.1 ng/ml is a factor to predict a disease-free status, validated to all ATA
initial risk categories [13 ]. The
determination of such a specific cut-off value of pre-RAI Tg – rather than the
consideration of a “value<10 ng/ml” – can be very useful in the initial
management of DTC patients. Similar studies that evaluated the role of pre-RAI Tg in
predicting ER response after initial therapy have determinate values of cut-off
between 3.3 to 8.55 ng/ml [12 ]
[21 ]
[22 ]
[23 ]
[24 ]. These differences between the studies
may be affected by differences in patient populations, by the timing of the serum Tg
determination and/or by preparation mode for remnant ablation (thyroid hormone
withdrawal or recombinant human TSH) and variability in serum Tg depending on the
method used. Moreover, our study not only defines an accurate cut-off value of
pre-RAI Tg for predicting an excellent response applicable across all initial ATA
risk categories but also contributes to identifying patients who may not warrant
further evaluation. Furthermore, our study provides a more defined initial outcome
compared to others where the purpose of pre-RAI Tg (diagnostic or prognostic) was
not well-defined.
Although a high NPV of pre-RAI Tg for long-term remission in low-to-intermediate-risk
DTC patients has been reported previously, we observed a high NPV in high risk
(90.5%) patients compared to low (53.3%) and intermediate (58.1%) risk patients, an
effect of the higher sensitivity and lower prevalence of ER to therapy among
high-risk patients. Of note, 20.1% of patients were classified as high risk for
recurrence indicating a cohort that included many patients with advanced disease.
Together these results follow in the same direction, supporting that lower pre-RAI
Tg level is a reliable tool on evaluating patients who will have a good response to
initial therapy, helping in decision making about patient early and long-term
follow-up.
Our study focused on other prognostic factors associated with an initial ER
identifying patients which require less intensive follow-up. In fact, levels of
pre-RAI Tg ≤ 7.55 ng/ml had the best sensitivity and specificity to predict ER to
initial therapy, having even a stronger correlation to ER than female sex and ATA
initial risk evaluation, establishing good prognostic factors in DTC. Indeed,
initial risk evaluation has an important role on the treatment plan and also in
planning for long-term follow-up, pre-RAI Tg levels can guide the next step of
evaluation of initial therapy [20 ]
[21 ]
[23 ]
[24 ]. In fact, serum Tg
measured 6 to 12 months after the first RAI can be correlated with pre-RAI Tg in
low-risk patients and can predict persistent or recurrent disease in the earliest
postoperative time [11 ]
[25 ]. We observed a significant correlation
between pre-RAI Tg with b-Tg and s-Tg measures in the initial assessment of therapy
response. In patients with intermediate risk undergoing TT and RAI therapy, ATA
recommendations are unclear regarding assessment of initial response therapy by b-Tg
or s-Tg [1 ]. In our sample, in which
practically half of patients are classified as intermediate risk, a pre-RAI Tg≤7.55
ng/ml was an independent factor associated with ER. In these patients, where initial
evaluation of therapy has the greatest impact on treatment and follow-up, an ER
decreases the estimated risk of recurrence from 20–30% to<5%, leading to less
intensive follow-up and no need for TSH suppression [1 ]
[3 ]. In intermediate risk patients, we observed a pre-RAI Tg in patients
with ER [pre-RAI Tg 1.95 ng/ml (0.6–9.5)] lower than patients with IndR [pre-RAI Tg
8.4 ng/ml (1.8–21.4)] and IncR (pre RAI Tg 27.8 ng/ml (5.1–129.5)] (data not shown).
In fact, previous studies demonstrate that in the evaluation of response to initial
therapy, with the non-stimulated Tg, 56% of intermediate-risk patients had ER to
therapy, of whom none had disease at the end of follow-up [26 ]. Like these results, 88.7% of our
intermediate risk patients were also free of disease at last evaluation. In the same
way, in low or intermediate-risk patients a s-Tg may not be necessary when b-Tg
defines the response as excellent [27 ].
Thus, these data together suggest that low or intermediate risk DTC patients with a
pre-RAI Tg≤7.55 ng/ml may be initially evaluated only based on b-Tg together with
neck US.
At first evaluation, an ER was observed in 78.8% of patients who are low, 62.5%
intermediate and 35.2% high initial risk patients. The risk of recurrence in long
term follow-up of patients with initial ER is low (1–2%) [1 ]. In addition, most of the DTC patients
who were considered to be disease free after the initial treatment generally
remained with this status at long-term follow-up [28 ]. At the last evaluation we observed
that, in our sample of DTC patients with ER, the majority of patients (96.1%)
remained in the same category. Structural persistent/recurrent disease was detected
in 7.8% of DTC patients initially and 84.6% remained in the same category at the
final follow-up. This prevalence was less frequent than other studies that showed 13
to 26% of structural persistent/recurrent disease in 10 years follow-up [3 ]
[13 ]
[17 ]. Of note, the majority
of the patients in the present study were stage I and 120 (83.3%) of them were at a
low or intermediate initial risk of recurrence, which are known to have an excellent
prognosis [29 ]. In this way, our data
reinforce the role of pre-RAI Tg levels as a predictor of ER in long-term
prognosis.
There are some limitations to this study. It is a retrospective design study at a
single tertiary referral center. According to previous recommendations many patients
at low initial risk received radioiodine ablative or adjuvant, which nowadays, would
not have been submitted to this therapy. On the other hand, the data reflect
real-life clinical practice, in which different Tg measurement methodologies, with
distinct functional sensitivities over time may have enhanced the external validity
of our findings. The fact that all assays used in this cohort had functional
sensitivities<0.5 ng/ml or less allowed us to retrospectively analyze the results
using the evidence-based cut-off value recommended by the ATA [1 ]. In addition, this study included
patients followed for a long time at the same institution, being evaluated by the
same professionals, and being submitted to the same follow-up and investigation
protocol.
In conclusion, our results suggest that stimulated thyroglobulin levels immediately
before radioiodine therapy predict first response therapy evaluation. The pre-RAI Tg
was a key predictor of initial excellent response to therapy and can be a tool to
promote a patient-centered care chronogram allowing the improvement of the patient’s
quality of life and reducing the costs of follow-up, especially in the public health
system.
