Background

Thyroid cancer is a common head and neck malignancy and also it is the most common endocrine tumor in the body [1,2,3]. The incidence of thyroid cancer has rapidly increased in the USA and other developed countries over the past 30 years [4,5,6,7]. From 1975 to 2009 there was a three-fold increase in incidence rates, from 4.9 to 14.3 per 100,000 individuals [6, 8]. Papillary thyroid carcinoma (PTC) and follicular thyroid carcinoma (FTC) are collectively classified as DTC and account for at least 94% of thyroid carcinoma [9]. Both have an excellent prognosis, with a 20 years survival of 90%–95% and 75%, respectively [10]. The postoperative recurrence rate of thyroid cancer is up to 23–30% [11,12,13,14,15,16]. The most effective therapeutic strategies for differentiated thyroid carcinoma (DTC) are thyroid surgery (total or near-total thyroidectomy) followed by iodine-131 (131I) ablation [17]. 131I therapy is not only appropriate for primary tumors but also for lymph node (LN) and distant metastases [18, 19]. However, therapeutic strategies should be more cautiously refined [20].

The role of 18F-fluorodeoxyglucose-positron emission tomography/computerized tomography (18FDG-PET/CT) starts with the development of metastatic diseases, which are not responsive to radioiodine therapy anymore. FDG accumulates in tumor lesions that are missed by iodine scintigraphy [21,22,23]. Loboulleux et al. in their study concluded that in patients with elevated serum Tg level after a total thyroidectomy and a normal post-ablation whole body scan (WBS), 18FDG-PET/CT rather than a post-empiric 131I WBS is preferred for detection of recurrence/metastases [24]. Over the last years, several studies have shown high sensitivity (80–90%) of 18FDG-PET/CT in detecting local or distant recurrences in patients with high thyroglobulin level and negative 131I WBS [25,26,27,28] in addition to recent studies that discuss the role of 18FDG-PET/MR in those patients [29, 30]. However, Piccardo et al. [31] in their study declared impact of 18FDG-PET/CT in changing the management plan in stage IV DTC subgroup with positive 131I WBS. To the best of our knowledge, there is a paucity of studies discussed this issue.

The current study aims to evaluate additive value of 18FDG-PET/CT to patients who were 131I post-therapy whole-body scan (TxWBS) positive for recurrent DTC and to assess how this can influence the treatment strategy.

Methods

The present study enrolled 43 patients selected from 1002 DTC patients referred to nuclear medicine and clinical oncology department from January 2011 to May 2021 by reviewing their medical records, including clinical evaluation, serum Tg and anti-Tg antibodies, neck US, diagnostic 131I WBS, 131I TxWBS, and 18FDG-PET/CT. 18FDG-PET/CT was done at the national cancer institute during the follow-up.

The study design was approved by institutional review board of clinical oncology and nuclear medicine department. Patient consent was not possible because of the retrospective design of the study.

Inclusion criteria: Patients age above 18 years old, with histopathologically proven DTC treated by thyroidectomy and 131I treatment. We selected the patients who done18FDG-PET/CT within 3 months prior to 131I therapy in whom 131I WBS was positive for suspicious locoregional or distant recurrence either clinically or biochemically by elevated Tg serum level in hypothyroid status. The locoregional and distant recurrences were proven pathologically (fine needle aspiration [FNA] cytology or excisional biopsy), or by conventional imaging (neck ultrasound [US], Tc99m MDP bone scan, chest CT and triple phase CT) or by follow-up for at least one year post 18FDG-PET/CT scan to evaluate lesion behavior overtime; stationary or regressive course of the lesion after 131I therapy or progression with time confirms its malignant nature.

Exclusion criteria: Patients with elevated thyroglobulin (Tg) serum level and negative 131ITxWBS or with history of second primary malignancy.

Serum Tg and Tg antibodies measurement: Blood samples were obtained to measure both Tg and Tg antibodies serum levels in all patients in hypothyroid state (thyroid stimulating hormone (TSH) level ≥ 30 mIU/L). Serum Tg level was considered abnormal when its value was higher than 1 ng/ml.

Neck ultrasound (U/S): Neck U/S was performed for all the patients to evaluate the thyroid bed as well as central and lateral cervical nodal compartments. Sonographically suspicious lymph nodes were biopsied for cytology or excised for histopathology. Suspicious criteria for metastases include round shape, hyperechogenicity, microcalcification, hypervascularity, cystic aspect and loss of hilar fat.

CT chest and triple phase CT were done for suspected lung and liver metastases, respectively. CT imaging was acquired by 64 multi-detectors CT scanner. All CT images were interpreted by expert radiologists.

Technetium methylene diphosphonate (Tc99m MDP) scan: 3 h post-injection of 740 Mbq (20 mCi) of Tc99m MDP anterior and posterior whole body planner images were acquired. Abnormal focal tracer uptakes are suspicious of osseous deposits.

131I scans: Both diagnostic and Tx WBS were done in all the patients after hormonal withdrawal for 4 weeks with the TSH serum level ≥ 30 mIU/L. Patients were also instructed to follow a low iodine diet 2 weeks prior to iodine intake. Diagnostic 131I WBS was performed after oral intake of 185 MBq (5 mCi), while 131I TxWBS was done after large dose (100-200 mCi) 5–7 days later. Imaging was done with a dual-headed gamma camera using high-energy collimators. The energy window was set at 15% centered on 364 keV with a 256 × 1024 size with a scan speed of 15 cm/min. The images were evaluated for iodine avid locoregional or distant recurrences. Images were interpreted by two experienced nuclear medicine physicians. Any abnormal focal iodine activity in operative thyroid bed, anatomical sites of lymph nodes or the rest of the body was interpreted as positive result.

18FDG-PET/CT: The scan was done using (PET/CT 710, Discovery; general electric (GE)). All patients were instructed to fast for 4–6 h before injection of 5.2 MBq (0.19 mCi)/kg body weight 18FDG. Blood glucose level was not exceeding 150 mg/dL, 40–60 min post-tracer injection CT was acquired first followed by PET scanning (5–7 bed positions; acquisition time, 2–3 min/bed position). CT without contrast agent was performed with the following parameters: 40 mAs, 130 kv, slice thickness: 2.5 mm, and pitch: 1.5. The CT scans were acquired during breath holding. Patients are scanned in the supine position from skull base to mid-thigh. The CT-data were used for attenuation correction and anatomical localization. Images were reconstructed applying a standard iterative algorithm (ordered-subset expectation maximization). Images were interpreted at a workstation equipped with fusion software that provides multi-planar reformatted images and enables display of the PET, CT and fused PET/CT images. Image interpretation was accomplished by two experienced nuclear medicine physicians. Abnormal FDG uptake was defined as a focal increased uptake higher than that of liver activity at anatomic localizations.

Data analysis was based on patients, organs and lesions. Organs are classified into: thyroid bed, LN (cervical), lung, liver, and bone. Lesions in one lung regardless of their numbers were counted as one lesion.

The collected data were computerized and statistically analyzed using Statistical Package for Social Science (SPSS) program version 25.0. Quantitative data were expressed as median and range. Qualitative data were represented as frequencies and relative percentages. Chi square test was used to calculate difference between qualitative variables. Cohen's kappa test was used to estimate the agreement between different methods of diagnosis. Validity data were calculated using Sensitivity, Specificity, Positive predictive value, negative predictive value and accuracy. P value of < 0.05 indicates significant results and P value of < 0.001 indicates highly significant results.

Results

This retrospective cohort study enrolled 43 patients with female predominance 34 (79%) and 9 males (21%) with median age 55 years and range 34–59 years. Regarding the histopathological analysis, papillary type represented 33(76.7%) out of 43 patients, while follicular type was less than 10 (23.3%). According to the American Joint Cancer Committee (AJCC) TNM staging system 8th edition, the majority of the patients (76.7%) were at stage IV, while the remainder of the patients were represented almost equally at stages I, II, III (Table 1).

Table 1 Clinico-pathological data of the studied group
Full size table

Two patients out of 43, in whom diagnostic 131I WBS and US revealed only small recurrence in thyroid bed with significantly high serum Tg.18FDG PET/CT was done which revealed 3 cervical LN metastases, the dimensions of the smallest LN were 1.1 × 0.8 cm. Those patients were referred to surgery before receiving 131I therapy so they were excluded from the comparison between 18FDG PET/CT and 131I Tx WBS.

Performance of 18 FDG-PET/CT

The 18FDG-PET/CT detected tumor recurrent lesions in 35 (81.3%) patients with statistically significant relation between positivity of 18FDG-PET/CT and old age, female sex, high tumor stage and progressive course as 94.5% of patients equal or older than 55 years, 91.1% of females,75 and 97% of patients in stage III and IV and 97%showed progressive course. compared to 0% of patients with positive 18FDG-PET/CT < 55 years old, 44.4% of males, 0% of patients in stage I and II and 30% with regressive course (P value = 0.000, 0.001, < 0.001 and 0.000, respectively) (Table 2).

Table 2 Relation between positivity of 18FDG-PET/CT and age, sex, DTC staging and outcome
Full size table

Comparison between 18 FDG-PET/CT and 131 I TxWBS

In one patient 131ITxWBS falsely revealed skull osseous deposit that was attributed to surgical intervention and also 18FDG-PET/CT falsely reported cervical LN metastasis (1 lesion) in one patient, which was confirmed to be sarcoidosis by FNA biopsy.

Based on lesion analysis (Table 3) 18FDG-PET/CT could detect more lesions compared to 131ITxWBS [98 (79%), 87 (70.2%) out of 124] lesions, respectively. Regarding the different organs, PET/CT detected 52 out of 61 (85.2%) versus 48(78.6%) detected by TxWBS. Kappa value of agreement was poor between each technique and the confirmatory different tools; however, statistically significant good agreement between 18FDG-PET/CT and 131I TxWBS was noted (Kappa 0.78 and p < 0.001) (Table 3).

Table 3 Organ- and lesion-based analysis by 18FDG-PET/CT and 131I TxWBS in the context of histopathology, conventional radiological tools and lesion behavior over time
Full size table

18FDG-PET/CT could detect additive lesions compared to 131I TxWBS in 18 patients, while 131I TxWBS could detect additive lesions in only 6 patients.

18FDG-PET/CT has higher sensitivity than that of 131ITxWBS either based on lesion or organ analysis (78.2% vs. 69.4 and 85.2% vs. 77%, respectively) (Tables 4, 5).

Table 4 Diagnostic performance of 18FDG-PET/CT and 131I WBS based on lesion analysis in relation to histopathological results, conventional radiological tools and lesion behavior overtime
Full size table
Table 5 Diagnostic performance of 18FDG-PET/CT and 131IWBS based on organ analysis in relation to histopathological results, conventional radiological tools and lesion behavior overtime
Full size table

Impact of 18 FDG-PET/CT on the treatment plan

Twenty-five patients (58.2%) out of the 43 patients revealed no change in management (8 patients with negative 18FDG-PET/CT and 17 with positive 18FDG-PET/CT), while 18FDG-PET/CT could change the treatment plan in 18 (41.8%) patients out of 43; in two patients, WBS and US revealed only small recurrence in thyroid bed with significantly high serum Tg, so 18FDG PET/CT was done which revealed cervical lymph nodes (LN) metastases; therefore, the treatment plan was changed from only RAI therapy to pre-RAI therapy neck dissection (Fig. 1). Besides, 18FDG-PET/CT detected symptomatic osseous deposits that were missed by diagnostic and post-therapeutic WBS in seven patients that necessitated external radiation after receiving 131I therapeutic dose instead of only radioactive iodine (RAI) (Fig. 2). Also large hepatic metastasis was detected by18FDG-PET/CT in one patient with maximum standard uptake value (SUVmax) 30 in addition to recurrence in thyroid bed, while the diagnostic 131I WBS and post-therapeutic scans revealed only recurrence in thyroid bed, the decision was taken to undergo surgical excision of the hepatic metastasis then patient will receive RAI dose. Unfortunately, the surgery was postponed as the patient was surgically unfit, so he received his 131I therapeutic dose only. Moreover, adjustment of the RAI dose from 5.55 GBq (150 mCi) to 7.4 GBq (200 mCi) in six patients in consideration with their lung and bone recurrences determined by 18FDG-PET/CT, which were not detected by the conventional radiological tools. In two patients 18FDG-PET/CT demonstrated thyroid recurrence in operative thyroid bed encroaching on larynx and tracheal lumen which warranted surgical referral before RAI; however, one patient refused the surgical approach and the other was unfit for surgery so both patients received their therapeutic 131I doses only. It was noticed that TxWBS detected the lesions but unfortunately failed to characterize the lesion as accurately as 18FDG-PET/CT did (Table 6).

Fig. 1
figure 1

60 year old male patient with stage III papillary thyroid cancer who underwent total thyroidectomy then he received radioactive iodine ablative dose. During follow up he was represented by elevated level of I diagnostic WBS revealed only small sized recurrent thyroid 131 and ng/ml While both neck US 60, serum Tg-FDG18 was done for proper evaluation of the condition. (A1, A2); Axial FDG-PET/CT18, cancer. Therefore (arrows) sections reveal bilateral metastatic cervical LNs level II PET/CT (B) 131I diagnostic WBS shows only iodine avid residual thyroid tissue in the operative thyroid bed (arrow)

Full size image
Fig. 2
figure 2

Fifty-seven-year-old female patient with stage III differentiated follicular thyroid cancer underwent total thyroidectomy and 131I ablation. During follow-up, patient was represented by elevated serum Tg level; 200 ng/ml. Diagnostic WBS revealed only small recurrent thyroid tissue in thyroid operative bed (not available), so 18FDG-PET/CT was done to detect the possibility of hidden metastases. A 18FDG-PET/CT axial, sagittal and coronal sections reveal thyroid distant metastases; supraclavicular region LN, dorsal and lumbar vertebrae, paraspinal mass, pelvic bones and left femur (arrows) suggesting dedifferentiating nature of these lesions. B 131I-post-therapeutic WBS shows only bifocal iodine avid recurrent thyroid tissue at operative bed (arrow)

Full size image
Table 6 Impact of 18FDG-PET/CT on treatment management
Full size table

Discussion

Long years ago, 18FDG-PET/CT has been widely used as a tumor-seeking agent for various cancers. Nevertheless, researches on its role in thyroid cancer staging appeared only in the mid-1990s [31].

In recent years, it is obvious that 18FDG-PET/CT has a major role in thyroid cancer management but its utility in patients with recurrent DTC should be tailored according to circumstance of each patient [32].

In general, the role of 18FDG-PET/CT in initial staging and follow-up of low-risk patients with DTC is limited [33]. Conversely, in patients with more aggressive thyroid cancers, 18FDG-PET/CT is a very useful tool for determining the extent of metastatic disease [34, 35]. Also, for prognostic purposes and treatment response assessment [33, 36]. Moreover, Abraham et al. in their study stated that performing 18FDG-PET/CT in patients with recurrent/persistent DTC should take into account not only Tg levels or 131I WBS findings, but also, should be on the basis of clinical and histopathological features and their individual risk [33].

Schönberger et al. stated that, iodine-avid lesions of DTC usually have low or absent FDG uptake with a positive correlation between expression of glucose transporter 1 (GLUT1) and dedifferentiation of thyroid cancer cells [37]. Moreover, Piccardo et al. suggested the occurrence of iodine-concentrating metastases with high glucose uptake [31].

Some researchers declared that 18FDG-PET/CT has been shown to be a valuable diagnostic tool for the detection not only of 131I non-avid lesions, but also of 131I avid lesions of metastatic DTC [38].

Treglia et al. in a review article done in 2013 stated that; 18FDG-PET/CT and 131I WBS may provide complementary information useful in the restaging of DTC patients [38]. Moreover, Piccardo et al. believe that 18FDG-PET/CT may change the therapeutic approach in a stage-IV DTC subgroup of patients [31].

In the current study; 18FDG-PET/CT detected lesions in 35 (81.3%) out of 43 patients which goes with that declared by Loboulleux et al. in their study that enrolled 47 patients with DTC after thyroidectomy and 131I treatment, 18FDG-PET/CT was done for only 34 patients. The percentage of positive 18FDG-PET/CT among those with positive Tx 131I WBS was 5 out of 6 patients (83.3%) and also concordant with Piccardo et al. in their study which enrolled 20 stage-IV DTC patients with elevated Tg levels associated with positive I31I WBS. The 18FDG PET/CT was positive in 16 out of 20 patients (80%). On the other hand, the rate of 18FDG-PET/CT lesion detection in the current study is lower than that of Liu et al. and higher than that of Sandra et al. and Oh et al. studies. Liu et al. study enrolled 212 DTC patients, among 59 patients with positive TxWBS the 18FDG-PET/CT positivity was 51(86.4%). On the other hand, Sandra et al. in their study which enrolled 90 patients with DTC after first 131 I ablation 18FDG-PET/CT was positive in 13(40%) out of 32 patients with positive I131 Tx WBS. Regarding Oh et al. study which enrolled 140 only 30 patients of the total number were positive in WBS and only 16 (53.3%) of them were positive in 18FDG-PET/CT [2, 24, 39,40,41]. this variation may be attributed to difference in population in each study.

The sensitivity of 18FDG-PET/CT based on lesion analysis in our study more or less goes with that has been detected by Piccardo et al. (78.2% vs. 80%) meanwhile, obviously lower than that stated by Riemann et al. (91%) among positive131 I WBS subgroup. These differences may be attributed to the differences in study population regarding the risk stratification [31, 39, 42].

Obviously, the specificity of 18FDG-PET/CT in our study was significantly lower than that was calculated by Riemann et al. in their study (50% vs. 89%). We believe that small sample size did not enable precise assessment of the specificity [42].

In the current study, the sensitivity of 131I TxWBS based on lesions analysis goes with that measured by Piccardo et al. in their study (69.4% vs. 67%). Reduction in sensitivity may be attributed to an increased number of patients in III and IV stages on expense of I and II stages.

In our study 18FDG-PET/CT findings led to modifications in the management of 41.6% which was less than that declared by Piccardo et al. in their study 55%, this may be explained by decrease percentage of patients in stage IV compared with that in Piccardo et al. 76.7% versus 100%, respectively [31].

Our result was higher that detected by Maamoun et al. in their study which enrolled 126 patients who underwent 18F FDG PET/CT initially before 131I ablation 30.6% [43].

Limitation

One of the limitations of the study was non-availability of histopathological examination in a number of patients to confirm recurrences for both practical and ethical reasons. We admit that small sample size influenced proper assessment of specificity and also fair comparison between low and high risk groups, which may be taken in consideration in future studies. Also another limitation is comparing 18FDG-PET/CT with planer 131 I WBS and not with single photon emission tomography/computerized tomography (SPECT/CT).

Conclusion

18FDG-PET/CT may be complementary to 131ITxWBS in high-risk DTC with impact on treatment strategy (Additional files 1, 2, 3: Fig. 1).