Key words
dynamic risk stratification - initial response to therapy - recurrence
Introduction
Although initial risk stratification provides important prognostic information that
can allow individualized management recommendations, clinicians have long recognized
that dynamic risk-stratification scheme is recommended over time as a function of
response to initial therapy and biological behavior of the papillary thyroid cancer
(PTC) [1]
[2]
[3]
[4]. An indeterminate response is nonspecific
biochemical or structural findings that cannot be confidently classified as either
benign or malignant [2]
[3]. The clinical outcome of patients with an
indeterminate response is reported to be intermediate, between incomplete and
excellent responses, and the structural persistence/recurrence rate around
13–20% [3]
[5].
Identifying patients at greater risk of adverse outcomes is an important step in the
management of PTC patients. Since PTC is known to be usually an indolent neoplasia,
the ‘‘over follow-up effect’’ with unnecessary
surveillance, diagnostic tests, and medical appointments is a concern [6]
[7]. The American Thyroid Association (ATA)
guidelines also outline the management implications for patients with an
indeterminate response recommending continuous observation with appropriate serial
imaging and serum thyroglobulin (Tg) and thyroglobulin antibodies (TgAb) [3]. However, these recommendations are general
and nonspecific for this group of PTC patients. Recent studies have demonstrated
that patients with indeterminate stimulated Tg at 12–24 months after initial
treatment have more risk of structural disease than patients with positive results
for TgAb or nonspecific imaging findings [8].
The objective of the present study is to evaluate the clinical outcome of patients
with an indeterminate response in a tertiary, university-based hospital and in
addition, to assess whether indeterminate clinical findings in the initial
evaluation post-therapy may affect the rates of recurrent/persistent
disease.
Patients and Methods
Study population and design
We retrospectively reviewed the medical records of Papillary Thyroid Carcinoma
(PTC) patients followed at the Endocrine Division of Irmandade da Santa Casa de
Misericórdia de Porto Alegre, a tertiary care, university teaching
hospital in Southern Brazil. The study was approved by the ethics committees of
the institutions (CAAE:1.555.870 and 1.600.048). Following total thyroidectomy
(TT) with or without radioactive iodine (RAI) therapy, PTC patients were
identified as having an indeterminate response to therapy within the first
6–24 months after initial treatment. An indeterminate response was
defined as any of the following findings: (i) basal thyroglobulin (bTg)
≥ 0.2 to < 1 ng/ml or stimulated
Tg (sTg) ≥ 1 to < 10 ng/ml (for
patients treated with RAI) and bTg ≥ 0.2 to
< 5 ng/ml or sTg ≥ 2 to
< 10 ng/ml (for patients without RAI therapy);
(ii) stable or declining TgAb or (iii) nonspecific findings on neck US or faint
uptake in the thyroid bed on diagnostic whole-body scan (DxWBS) [3]. Patients with known structural disease
after initial treatment were excluded.
Follow-up protocol
Our treatment protocol is according to the ATA guideline and consists of
performing TT, administering or not an ablative or therapeutic dose of RAI and
levothyroxine suppression therapy according to the initial risk assessment [3]. Follow-up duration was defined as the
time between the TT and the last medical visit to the clinic. Structural disease
was defined based on imaging, cytology or histology results, and/or
unequivocal ectopic uptake (excluding false-positive results) on post-therapy
DxWBS or FDG-PET/CT.
Risk stratification and outcomes
All patients were classified according to the ATA 2015, into low, intermediate,
or high risk of recurrence [3].
Additionally, they were staged according to the American Joint Committee on
Cancer and International Union Against Cancer (TNM staging 8th
edition) [9]. As patients were followed
up, response to therapy was dynamically defined until last follow-up visit as
excellent, indeterminate, biochemical incomplete and structural incomplete,
according to ATA 2015 guidelines [3]. When
evaluating the initial response therapy, PTC patients were further classified
into 3 groups in order to specify the indeterminate group according to
recurrence. Tg group: detectable Tg and negative TgAb, regardless
nonspecific imaging findings; TgAb group: positive TgAb, regardless Tg
levels and nonspecific imaging findings; Image group: nonspecific
findings on neck US or faint uptake in the thyroid bed on DxWBS, undetectable Tg
and negative TgAb. For the purposes of patient enrollment and the analysis, the
results of each TgAb assay were recorded simply as “positive” or
“negative” based on a titer above or below the cut-off of the
specific assay. We considered TgAb positive with any value different from
negative value according to the method. The primary outcome of this study was
structural disease. Disease free survival (DFS) was considered the time from
initial surgery to last follow-up or structural disease detected.
Laboratorial analysis
Serum Tg measurements were performed using immunoradiometric assays:
radioimmunoassay (until 2000); electrochemiluminescence (from 2000 to 2011), and
chemiluminescence (from 2011 until the present), with functional sensitivities
from 0.5 to the current 0.1 ng/ml. Serum TgAb were measured
using the agglutination method until 2001, electrochemiluminescence until 2003
and by chemiluminescence from 2004 until the present. Serum TSH levels were
measured with a chemiluminescence assay.
Statistical analysis
Clinical and laboratory data are described as the mean±standard deviation
(SD) or median and percentiles 25 and 75 (P25–75) for continuous
variables and absolute numbers and percentages for categorical variables. To
determine the association between categorical variables, the chi-square test was
used. The DFS curve was plotted using Kaplan–Meier method and the
log-rank test was used to determine their significance. Data analysis was
performed using Statistical Package for Social Sciences (SPSS) software, version
25.0 (IBM Corp., Armonk, NY). A p-value of <0.05 was considered
statistically significant.
Results
Baseline characteristics of PTC patients with indeterminate response to
therapy
Sixty-six PTC patients were included in this study. In the general evaluation of
the sample, the mean age at diagnosis was 44.1±12.1 years, 58 patients
(87.9%) were women and 50 patients (75.8%) were <55
years of age. All patients underwent TT±lymph node dissection and 52
patients (78.8%) received ablative or therapeutic dose of RAI. Clinical
and oncological characteristics of these patients according to the indeterminate
group are described in [Table 1]. There
were no significant differences in prognostic clinical features between
groups.
Table 1 Characteristics of patients with papillary thyroid
carcinoma and indeterminate response after initial therapy according
to the indeterminate group.
|
Tg
|
TgAb
|
Image
|
p-Value
|
|
Female sex – n (%)
|
32 (84.2)
|
15 (93.8)
|
11 (91.7)
|
0.560
|
|
Age at diagnosis in years – mean (±SD)
|
42.8 (± 12.8)
|
47.4 (± 14.5)
|
44.3 (± 10.3)
|
0.491
|
|
<55 years at diagnosis – n (%)
|
30 (78.9)
|
10 (62.5)
|
10 (83.3)
|
0.347
|
|
Primary tumor size, cm – median (P25–75)
|
1.75 (0.68–2.58)
|
2.40 (1.08–3.95)
|
1.25 (0.73–2.08)
|
0.495
|
|
Microcarcinoma – n (%)
|
13 (34.2)
|
5 (31.3)
|
4 (33.3)
|
0.978
|
|
AJCC 8th Clinical Stage – n
(%)
|
|
|
|
|
|
Stage I
|
34 (89.5)
|
15 (93.8)
|
12 (100)
|
0.473
|
|
Stage II
|
4 (10.5)
|
1 (6.3)
|
0
|
|
|
ATA 2015 Risk Stratification – n (%)
|
|
|
|
|
|
Low
|
11 (28.9)
|
7 (43.8)
|
7 (58.3)
|
0.295
|
|
Intermediate
|
19 (50)
|
8 (50)
|
4 (33.3)
|
|
|
High
|
8 (21.1)
|
1 (6.3)
|
1 (8.3)
|
|
|
Total thyroidectomy – n (%)
|
20 (52.6)
|
10 (62.5)
|
7 (58.3)
|
0.788
|
|
Total thyroidectomy with neck dissection – n
(%)
|
18 (47.4)
|
6 (37.5)
|
5 (41.7)
|
|
|
Patients who received RAI – n (%)
|
30 (78.9)
|
13 (81.2)
|
9 (75)
|
0.789
|
|
Total RAI dose, mCi – median (P25–75)
|
114 (100–162)
|
100 (30–275)
|
100 (40–125)
|
0.922
|
|
Follow-up time, years – median (P25–75)
|
5.65 (2.78–8.88)
|
7 (3.23–10.25)
|
3.90 (2.23–7.78)
|
0.348
|
AJCC: American Joint Committee on Cancer; ATA: American Thyroid
Association; RAI: Radioactive iodine.
Disease status of PTC patients with indeterminate response at last
evaluation
[Figure 1] shows the outcomes of PTC
patients according to the dynamic risk stratification. After a median follow-up
of 5.7 years (2.6–9.75 years), 32 out of 66 (48.5%) of
indeterminate patients were reclassified as excellent response. Among them,
71.9% were without additional therapy. In fact, the majority of patients
(59/66; 89.4%) with an indeterminate response to therapy remain
disease free during prolonged follow-up. It was observed that among patients
with excellent or indeterminate response at final evaluation, only 2 patients
(6.3%) and 4 patients (14.8%), respectively, were high
recurrence risk. No low-risk patient showed evidence of disease.
Fig. 1 Follow-up outcomes, according to the dynamic risk
stratification (DRS). PTC: Papillary thyroid carcinoma.
Clinicopathologic features of patients with structural disease
Structural persistent/recurrent disease was detected in 7 (10.6%)
patients during follow-up. The median time to detection of structural disease
from the initial thyroid surgery was 7.2 years (min 4.1 – max 11.8
years). All patients had cervical disease, and one patient had concomitant lung
metastasis. Five patients (71.4%) were classified as intermediate risk
of recurrence and 2 patients (28.6%) as high risk. Interestingly, 5 of 7
patients (71.4%) had positive thyroglobulin antibodies at first 24
months after initial therapy, among them, 2 patients also had nonspecific
findings on neck US.
Structural disease and disease-free survival (DFS) of PTC patients according
to indeterminate response group
The persistence/recurrence rate in the TgAb group was significantly higher
than in the Tg and image groups, as follows: Tg group
(1/38), 2.63%; TgAb group (5/16), 31.25%;
Image group (1/12), 8.3% (p=0.007). To further
explore a possible role of TgAb positivity on disease outcome, we compared the
DFS rates in patients with positive TgAb (TgAb+) (n=16) and
negative TgAb (TgAb–) (n=50), regardless Tg levels and
nonspecific imaging findings. The 10-year DFS rate in group of patients with
increase of TgAb levels was significantly reduced (p=0.022) ([Fig. 2]). The 10-year DFS rates of
patients in the TgAb+ and TgAb– group were, 70.7% and
93.3%, respectively.
Fig. 2 Disease-free survival rates in patients with papillary
thyroid carcinoma who have indeterminate response according to positive
antithyroglobulin antibody (TgAb+) and negative
antithyroglobulin antibody (TgAb–). Overall log-rank test,
p=0.022.
Discussion
This study demonstrated that most PTC patients with an indeterminate response after
initial therapy remain disease-free after prolonged follow-up. Also, it was shown
that the presence of TgAb, regardless of nonspecific imaging findings is a potential
prognostic factor to predict disease status in these patients.
In our sample, the majority of patients (89.4%) were reclassified as having
an excellent response or remained in the indeterminate response therapy after
follow-up. In fact, in a cohort of 588 DTC patients treated with TT and RAI therapy,
the risk of having persistent or recurrent disease was significantly higher in
patients having an incomplete response to therapy (96%) than in patients
having an excellent (4%) or indeterminate response to therapy (13%)
[4]. In our study, structural
persistent/recurrent disease was detected in 10.6% of patients, less
frequent than other studies that demonstrate rates of 13–20% in a
longer follow-up [4]
[5]
[8]. Of note, the majority of the patients in
the present study were stage I and 85% of them were low or intermediate
initial risk of recurrence.
We observed that PTC patients with positive TgAb, at first 24 months after initial
therapy have more recurrence/persistent structural disease. Between patients
with structural disease, 71.4% presented positive TgAb at initial treatment
response. In addition, the frequency of ATA initial risk and TNM stage among the
groups seem not to contribute to this outcome, since there were no differences
between groups.
Thyroglobulin antibodies are present in about 14% of patients with PTC in
preoperative, but fall in the first year after effective initial therapy, especially
with RAI therapy [10]
[11]. Previous studies have demonstrated that
PTC patients with positive serum TgAb titer during the first year after treatment
were more likely to have persistent/recurrent disease than those who were
consistently TgAb-negative. Negative titers at 1 year may be associated with more
favorable outcomes [12]
[13]. In our study, most patients received RAI
therapy, and these findings suggest that patients with persistent elevated TgAb
levels in the first 24 months after primary therapy may have more chance of disease.
In this way, monitoring TgAb titers can provide surrogate information on disease
recurrence, and declining titers are usually associated with remission [12]
[13].
The clinical importance of low but detectable basal or stimulated levels of Tg is
unclear. In contrast of our results, a recent study has shown more prevalence of
structural disease in patients with indeterminate stimulated Tg levels [8]. However, in the study cited above, the
proportion of stage I disease or low-risk patients was significantly smaller in the
Tg group than in the TgAb or image groups. Furthermore, Tg levels were evaluated
regardless the presence of TgAb, that can interfere with measurements of serum Tg.
In fact, not all patients in our study were initially evaluated by stimulated Tg
(sTg); 22 of 38 patients (57.9%) had sTg, with median of
4.06 ng/ml (data not shown). The availability of ultrasensitive Tg
assays seems to remove the need for routine sTg determinations [3]
[14]. The role of basal and/or
stimulated Tg in evaluation of therapy in patients with indeterminate response is
not well established. Based on sTg, a change in category occurred in half the
patients with an initial indeterminate response, 44.4% to excellent response
and 5.5% to biochemical incomplete response and others remained in same
category [15]. Interestingly, some studies
have demonstrated that if sTg is done 5 years after the initial treatment, the vast
majority of responses originally classified as indeterminate (up to 98%) can
be reclassified as excellent response [16]. In
addition, patients with indeterminate response, rising values of unstimulated Tg,
independently from the basal levels, may be useful to identify patients with
progressive disease [17]. Therefore, following
the curve of basal Tg seems to be a suitable laboratory tool in selected cases of
indeterminate response [7].
In this study, 8.3% of patients in the image group had cervical structural
disease during follow-up. Nonspecific lesions can be difficult to distinguish
between recurrent thyroid cancer and non-recurrent benign lesions with neck US alone
[18]
[21]
[22]
[23]
[22]. Recent studies have proposed to restrict
US to patients with basal Tg ≥1 ng/ml after treatment with
RAI because this imaging method revealed disease in a small number of patients with
basal Tg <1 ng/ml [21]
[22]. In fact, it is very important the
routine neck US evaluation after thyroid surgery in the setting of increased Tg
levels or presence of increasing TgAb levels [3]
[22]
[23]. Of note, the European Thyroid
Association guidelines identify US characteristics that can be used to distinguish
indeterminate from suspicious findings in neck US, being useful in evaluation of PTC
patients with indeterminate Tg and/or presence of TgAb [24].
There are some limitations to this study. It is a retrospective design study in a
small sample, especially TgAb positive patients. Because the cohort was composed of
patients diagnosed over a period of 20 years, baseline Tg levels were measured with
different assays. The fact that all assays used in this cohort had functional
sensitivities of 1 ng/ml or less allowed us to retrospectively
analyze the results using the cut-off value recommended by the ATA [3]. Another possible limitation is that there
was no standardization of US evaluation to address lymph node disease. However, this
limitation may be partially mitigated by the fact that all ultrasound reports were
reviewed. Despite these limitations, this study included patients seen at the same
institution, were evaluated by the same follow-up and investigation protocol.
In conclusion, the majority of patients with an indeterminate response to therapy
remain disease-free during prolonged follow-up. Our results suggest that PTC
patients with indeterminate response after initial therapy by the presence of
positive serum TgAb have more risk of development of structural disease in the long
term. Prospective studies with a larger number of patients and long-term follow-up
are needed to clarify the potential role of positive serum TgAb in the prediction
of
PTC.
