Introduction

Medullary thyroid carcinoma (MTC) is an uncommon neuroendocrine tumor, accounting for 1-2% of all thyroid malignancies in the United States [1]. From a clinical perspective, these tumors have a more aggressive behavior than differentiated thyroid neoplasms, and the presence of residual disease or tumor recurrence after initial surgery is a usual finding, especially when lymph node metastases are present at diagnosis [2]. Given that surgery is the only curative treatment modality at present, localization of the disease is essential in cases where persistence or recurrence is suspected by elevated levels of calcitonin (Ct) or carcinoembryonic antigen (CEA).

18-F-Fluorodihydroxyphenylalanine (18F-DOPA) is an amino acid analog used for positron emission tomography (PET). Typically, this tracer uptake is high in neuroendocrine cells because 18F-DOPA is incorporated into the cell cytoplasm through an amino acid transporter called LAT 1 (L-type amino acid transporter 1) and then decarboxylated by amino acid decarboxylase (AADC) [3, 4].

Since 2001, several studies have evaluated the sensitivity of 18F-DOPA PET in the detection of persistent or recurrent MTC compared with the morphological and functional techniques available up to date [5,6,7,8,9,10,11,12,13,14]. These studies generally agree on the superiority of 18F-DOPA PET over other diagnostic methods. In 2012, Treglia et al published a meta-analysis including the eight most relevant studies to assess the diagnostic accuracy of PET or PET/CT (computed tomography) with 18F-DOPA in recurrent MTC. This study comprised a total number of 146 patients and concluded that the detection rate per patient and per lesion was 66% and 71%, respectively. They also found an increased detection rate in patients with higher calcitonin levels and lower calcitonin doubling times (Ct-DT) [15]. However, there is a significant heterogeneity among published studies and the number of patients, ranging between 11 and 39, is quite short. Most studies suggest the superiority of PET/CT with 18F-DOPA regarding PET isolated on the location of lesions [9]. Therefore, it would be more appropriate to design studies where PET/CT with 18F-DOPA is compared to other techniques, mainly to 18F-FDG (18F-Fluorodeoxyglucose) PET/CT, as this is the reference tracer used in most hospitals and has differential characteristics. Only five papers have compared these two techniques so far [8,9,10,11,12].

The American Thyroid Association (ATA) guidelines for the management of MTC, first published in 2009 and updated in 2015 [1, 16], recommend that additional imaging procedures beyond cervical ultrasound should be done in patients with a serum calcitonin value above 150 pg/mL after the initial surgery, looking for lymphatic or distant metastases. In current literature, only two studies have tried to validate this cutoff, offering no consistent results [9, 14].

The aim of our study was to evaluate and compare the usefulness of 18F-DOPA and 18F-FDG PET/CT in the detection of persistent or recurrent MTC, as well as to analyze the correlation between their positivity and Ct, CEA, Ct-DT, and CEA-DT (CEA doubling time) values. As a secondary objective, we evaluated whether the 150 pg/mL Ct cutoff set by the ATA guidelines to perform additional imaging is appropriate.

Material and methods

Patients

This observational, prospective study was conducted between November 2011 and January 2014. All patients with suspected persistent or recurrent MTC referred from endocrinology to nuclear medicine department to perform a PET/CT, underwent a 18F-FDG and a 18F-DOPA PET/CT.

Inclusion criteria for this study were:

  • Previous total thyroidectomy associated or not to lymph node dissection, with histological result of MTC.

  • Suspected recurrent or persistent disease after initial surgery by elevated calcitonin levels above the laboratory range (> 11.5 pg/mL in women and > 18 pg/mL in men).

  • Confirmation by cytology, histological exam or other imaging tests during follow-up (at least 12 months after the PET/CT study).

Patients were referred from different endocrinology departments in Andalusia and in all cases, both PET/CT scans were performed in Virgen del Rocío University Hospital (Seville), as this is the reference center.

Biochemical analysis

The serum Ct levels were measured by chemiluminescence immunoassay (Immulite 2000, Siemens). Normal values in our laboratory are set at 5-11.5 pg/mL for women and 8.4-18 pg/mL for men. Serum CEA levels were measured by electrochemiluminescence immunoassay (Cobas 8000, Roche), normal range considered as 0-3 ng/mL. Ct-DT and CEA-DT were calculated using the web calculator provided by the ATA (http://www.thyroid.org/thyroid-physicians-professionals/calculators/thyroid-cancer-carcinoma/). We only calculated doubling times in those cases in which at least four Ct and CEA determinations over a minimum period of 2 years were available. Ct-DT and CEA-DT > 24 months were considered associated with a better long-term prognosis than those < 24 months [16].

PET/CT with 18F-FDG and 18F-DOPA

In first instance, each patient underwent a PET/CT with 18F-FDG. 18F-DOPA scan was performed at a second stage. Time between both scans ranged from 5 to 80 days (median 21.5 days, interquartile range 13-31). A blood sample to measure Ct and CEA levels was extracted before administration of the radiopharmaceutical. The study was conducted prior fasting, after appropriate hydration and furosemide administration. Women were tested for pregnancy. Both radiopharmaceuticals were administered intravenous. The activity range was 174-288 MBq for 18F-DOPA and 164-444 MBq for 18F-FDG.

Body PET/CT scan (without contrast) was performed according to the European Association of Nuclear Medicine guidelines (EANM) [17]. The images were obtained using a PET/CT hybrid tomograph model Biograph-16 (Siemens) 60 min after administration of the radiopharmaceutical, with a rate of 3 min/bed. The scan included sections from the cranial vertex up to the middle third of the femoral diaphysis. PET emission images were acquired in 3D and then normalized performing an attenuation correction with CT image (parameters 120 kV and 50 mA). Reconstructions in coronal, sagittal and transverse planes were performed through the application of five successive iterations and three subsets. Interpretation of the results was carried out by two different nuclear medicine specialists (JI.C and JL.T) with large expertise (more than six years) in PET/CT imaging. These nuclear medicine physicians were blinded to each other, but they could access to electronic health record and check out tumor markers levels and radiological tests done before 18F-FDG and 18F-DOPA PET/CT. As all the patients had elevated Ct levels, a negative result in the PET/CT scan was considered as a false negative (FN). To rule out false positive (FP), imaging results were compared with cytology, histological exams or follow-up examinations.

PET images were analyzed visually and semiquantitatively by calculating maximum standardized uptake value (SUVmax) in the region of interest (ROI). Any focal accumulation of 18F-DOPA exceeding normal regional tracer uptake was interpreted as a pathological finding. In addition, focally increased 18F-FDG that did not correspond to normal structures was recorded as positive. ROIs were placed around the regions of increased tracer uptake for SUVmax determination.

Other imaging studies

Once the study was completed, we compared 18F-FDG and 18F-DOPA PET/CT results with other previously acquired imaging studies (cervical ultrasound, CT with iodinated contrast). These morphological techniques were performed before PET/CT and reported by experienced radiologists.

Follow-up

Follow-up was carried out according to recommendations of clinical practice guidelines in-force at the time of the study [1, 16, 18,19,20], with a minimum follow-up period of 12 months after PET/CT. We identified those patients in whom the therapeutic approach was changed as a result of the PET/CT study. In patients with a positive PET/CT, findings were confirmed during follow-up using cytological analysis (with fine needle aspiration biopsy, FNAB), histological analysis (when final clinical decision was surgery), or with other imaging tests.

Those patients with an initial negative 18F-DOPA PET/CT who underwent a second PET scan because of increasing Ct and CEA level were also studied.

Statistical analysis

Data were collected using the computer system of the hospital (electronic health record). The statistical tests were performed using the SPSS Statistics software, version 21.0 (IBM). Quantitative variables were defined as mean ± standard deviation (SD) if the sample followed a normal distribution or as median [IQR, interquartile range] if not. To compare proportions between groups, Fisher’s exact test was used. Means between two groups were compared using Student’s T-test (Mann-Whitney U-test was employed for non-normal distributions). The Wilcoxon test was used to test whether there were differences between the number of lesions detected by 18F-DOPA and 18F-FDG for specific locations or not. A p value <0.05 was considered significant.

Results

Eighteen patients (13 women, five men) were included (demographics are listed in Table 1). Of the patients, 55.5% had a sporadic MTC, and the remaining 45.5% had a MEN 2A Syndrome (multiple endocrine neoplasia type 2A). The age of the patients at the time of the study was 48 ± 11.7 years, with an average disease duration of 7.3 ± 6.2 years.

Table 1 Patient characteristics at baseline
Full size table

Laboratory and imaging data for each patient are shown in Table 2. PET/CT was used for tumor restaging purposes. Although tumor markers were elevated in all cases, in 14 out of 18 patients the previous imaging tests were negative or ambiguous. The remaining four cases had other positive imaging tests before PET/CT, so it was performed to find out if there were any additional lesion. Nine out of 18 patients enrolled had undergone a cervical ultrasound (thyroid bed and lymph nodes) and a CT with iodinated contrast (neck, chest, and/or abdomen) 12 months before being referred to PET/CT. Among the remaining nine patients, five had only a CT study available and four had a cervical ultrasound.

Table 2 Biochemical markers and imaging tests
Full size table

The average level of Ct and CEA was 311.1 ± 342.6 pg/mL and 11.6 ± 9.9 ng/mL, respectively. The average Ct-DT was 62.2 ± 76 months (n = 12), while the average CEA-DT was 92.3 ± 78.3 months (n = 10). The range of follow-up time after the PET/CT study was 18 to 43 months.

Patient-based analysis

In six patients (33.3%) both 18F-DOPA and 18F-FDG PET/CTs were negative. 18F-DOPA PET/CT was positive in 12 patients (sensitivity 66.7%, 95%CI 45-88.4), while 18F-FDG PET/CT detected nine tumor-positive patients (sensitivity 50%, 95%CI 26.9-73.1; p < 0.01). In three patients with a negative 18F-FDG PET/CT result, 18F-DOPA was able to detect lesions. Moreover, 18F-FDG PET/CT did not show any pathological findings in 18F-DOPA-negative patients.

Cytohistological confirmation was obtained in eight patients. In the remaining four 18F-DOPA PET/CT-positive patients, confirmation was obtained using other imaging tests and monitoring disease progression during the follow-up. The positive predictive value (PPV) of both examinations was 100% (see Table 2).

Ct, CEA, Ct-DT and CEA-DT were similar among patients with sporadic MTC and MEN 2A. No statistical differences were found in the results of PET/CT with 18F-DOPA or 18F-FDG between patients with MEN 2A and sporadic MTC, but the sensitivity of both PET/CT tended to be greater in cases with sporadic carcinoma. 18F-FDG PET/CT had a sensitivity of 60% (95%CI 29.6-90) in patients with sporadic MTC vs. 37.5% (95%CI 4-71) in MEN 2A (p=0.32). The sensitivity of 18F-DOPA PET/CT was 80% (95%CI 55.2-100) in patients with sporadic MTC vs. 50% (95%CI 15.4-84.6) in inherit MTC (p= 0.2).

Lesion-based analysis

A total of 28 lesions were detected with 18F-DOPA compared with 16 lesions with 18F-FDG (1.56 ± 1.5 lesions per patient, lpp, with 18F-DOPA PET/CT vs. 0.89 ± 1.18 lpp with 18F-FDG PET/CT; p=0.01). Of the nine patients with a positive result in both PET/CT with 18F-DOPA and 18F-FDG, 18F-DOPA identified more lesions in 44.4% (4/9) of cases (2.56 ± 1.33 lpp with 18F-DOPA PET/CT vs. 1.78 ± 1.09 lpp with 18F-FDG PET/CT; p=0.065). 18F-FDG PET/TC did not detect more lesions than 18F-DOPA in any patient (Fig. 1).

Fig. 1
figure 1

A 44-year-old man (patient 3) with MEN 2A syndrome. He had a persistent MTC after the first surgery, with high levels of calcitonin (785 pg/mL) and CEA (24.1 ng/mL). A. 18F-DOPA PET/CT images. B. 18F-FDG PET/CT images. A1. 18F-DOPA PET/CT shows multiple areas of abnormal uptake in the neck and mediastinum. B1-B2 . 18F-FDG PET/CT only demonstrated one perivascular mediastinum lymph node (right) also seen with 18F-DOPA PET/CT (A2). A3. One of the metastatic lymph nodes detected with 18F-DOPA PET/CT located in the left cervical region (this lesion was not detected with 18F-FDG PET/CT. B3)

Full size image

With regard to region-based analysis, both types of PET/CT detected 19 cervical and mediastinum lymph nodes, six local recurrences, two bone metastases, and one liver lesion. 18F-DOPA PET/CT identified all these lesions while 18F-FDG PET/CT was only able to detect 52.6% of lymph node metastases (1.5 [0-2.8] lpp with 18F-DOPA vs. 0.5 [0-1] lpp with 18FDG; p=0.04), 83.3% of local recurrences and 50% of bone lesions. 18F-FDG PET/CT failed to detect the liver metastasis. There were no statistically significant differences when comparing the number of lesions with both methods in these last three locations (p =0.32).

No statistically significant differences were found in the maximum SUV between lesions detected with 18F-FDG and 18F-DOPA (4.2 ± 2.5 vs. 4.1 ± 2.7, respectively; p=0.9).

Calcitonin, CEA, Ct-DT, and CEA-DT values and PET/CT positivity

  1. A)

    18F-DOPA PET/CT:

    Median values of Ct, CEA, Ct-DT, and CEA-DT of 18F-DOPA PET/CT-negative and positive patients are shown in Table 3. Ct and CEA levels were significantly higher in 18F-DOPA PET/CT-positive patients than in negatives ones (Fig. 2). No statistically significant differences between medians values of Ct-DT and CEA-DT in patients with positive and negative 18F-DOPA PET/CT were found, despite being the DT shorter in patients with a positive test (see Table 3).

    Table 3 Tumor markers, duplication times, and 18F-DOPA PET/CT result
    Full size table
    Fig. 2
    figure 2

    18F-DOPA PET/CT results and calcitonin level. Box and whisker plot representing the median and interquartile range of calcitonin in patients with positive and negative 18F-DOPA PET / CT results

    Full size image

    Using a Ct threshold of 150 pg/mL, as indicated by ATA guidelines [1], 10 out of 11 patients with serum Ct levels equal to or greater than this value had positive 18F-DOPA PET scans. However, only two out of seven patients with Ct levels below 150 pg/mL were 18F-DOPA PET/CT-positive (sensitivity of 90.9% vs. 28.6%, respectively; p=0.013). The detection rate of the technique was higher in patients with CEA levels over 5 ng/mL, compared to those with serum CEA below 5 ng/mL, but no significant differences were found (sensitivity of 81.1% vs. 42.9%, respectively; p=0.11). This CEA value cutoff was set arbitrarily. No statistical differences were found when comparing 18F-DOPA PET/CT results in patients with different Ct-DTs and CEA-DTs, using a DT threshold of 24 months, perhaps due to the small number of patients (sensitivity 75% with Ct-DT ≤ 24 months vs. 62.5% with Ct-DT> 24 months, p= 0.59; sensitivity 100% with CEA-DT ≤ 24 months vs. 66.6% with CEA-DT> 24 months, p= 0.7).

  2. B)

    18F-FDG PET/CT:

    Table 4 summarizes the median values of Ct, CEA and the doubling times based on the result obtained in 18F-FDG PET scan. Ct and CEA levels were significantly higher in 18F-FDG PET/CT-positive patients compared with negative ones. No differences were found in Ct and CEA-DTs regarding PET positivity.

    Table 4 Tumor markers, duplication times, and 18F-FDG PET/CT result
    Full size table

    Eight out of 11 patients with serum Ct levels over 150 pg/mL had positive 18F-FDG PET scans, but only one out of seven patients with Ct levels below 150 pg/mL were 18F-FDG PET/CT-positive (sensitivity of 72.7% vs. 14.3%, respectively; p=0.025). Similarly, a significant increase in the detection rate of PET/CT with 18F-FDG was observed using a CEA threshold of 5 ng/mL (sensitivity of 72.7% with CEA ≥ 5 ng/mL vs. 14.3% with CEA <5 ng/mL; p= 0.025).

    In our study, as it happened with 18F-DOPA PET/CT, the detection rate of 18F-FDG PET/CT was higher in patients with Ct and CEA-DTs less than 24 months, but no statistical differences were found (sensitivity of 50% with Ct-DT ≤ 24 months vs. 37.5% with Ct-DT> 24 months, p=0.57; sensitivity of 100% with CEA-DT ≤ 24 months vs. 44.4% with CEA-DT> 24 months, p= 0.5).

Comparative results with other imaging techniques

All imaging tests performed before to PET/CT and the corresponding findings are detailed in Table 2. Cervical ultrasound imaging was available in 13 patients. Seven out of these 13 patients had a negative result, and in six patients findings were nonspecific or ambiguous (not clearly pathological cervical lymph nodes). 18F-DOPA and 18F-FDG PET/CT were able to identify cervical lymph nodes metastases in two out of the seven patients with a negative cervical ultrasound. In the six patients with ambiguous sonographic findings, 18F-DOPA PET/CT was positive for lymph node metastasis or local tumor in five cases, while the study 18F-FDG PET/CT confirmed the lesions in four patients.

CT scans of neck, chest or abdomen were available in 14 patients. In four of them, CT failed to identify disease foci; in another five, the results were inconclusive, and in the remaining five there was evidence of disease. 18F-FDG PET/CT detected metastatic disease in two patients with a previous nonspecific CT result, and in four out of five patients with an apparently positive CT. 18F-FDOPA PET/CT seems to be superior to 18F-FDG PET/CT or CT, allowing disease detection in two patients with a previous negative CT, in three patients with previous equivocal findings and in four out of five patients with a positive CT scan.

Follow-up

In all patients except one a change in clinical management was possible (see Table 2, case number 18, in which a wait-and-see attitude was decided by a multidisciplinary team, because of the small size of the detected lesion). Seven patients were surgically treated; four patients received external beam radiotherapy; and in four patients, treatment with tyrosine kinase inhibitors (TKI) was started.

18F-DOPA PET/CT was repeated throughout the clinical follow-up in two out of the six patients with an initial negative result in both PET/CT with 18F-FDG and 18F-DOPA, due to the progressive rise of tumor markers (patients number 12 and 13 of Table 2). In the case number 12 a new 18F-DOPA PET/CT was performed 26 months after the first one, with a Ct value of 307 pg/mL (previous level of 148 pg/mL), obtaining a second negative result. In the patient number 13, which had a Ct value of 93 pg/mL at the beginning of the study, 18F-DOPA PET/CT was repeated 20 months later, when Ct levels had risen to 427 pg/mL. The second PET/CT with 18F-DOPA showed a paratracheal lesion suggestive of local tumor. The patient was operated, proving the findings on histological examination of the removed node.

Discussion

Our study shows the increased diagnostic sensitivity of 18F-DOPA compared to 18F-FDG PET/CT in the residual or recurrent MTC, in both the analysis per patient and per lesion. These results are in consonance with those of other researches published to date [8,9,10,11,12]. In our study, 18F-DOPA PET/CT had a sensitivity of 66.7% in the patient-based analysis, while the sensitivity of 18F-FDG PET/CT was 50%. In addition, 18F-FDG PET/CT detected only 57.2% of the lesions identified by 18F-DOPA. These data are similar to those from other studies, because integrating all published so far, the overall sensitivity of 18F-DOPA PET or PET/CT is 70%, while for 18F-FDG PET or PET/CT is 44% per patient and 42% per lesion [3, 21, 22]. We have not found statistical differences when comparing sensitivity of PET/CT with 18F-FDG or 18F-DOPA between patients with sporadic and familial MTC. However, according to our results, the sensitivity of both scans may be higher in patients with sporadic MTC than in MEN 2A cases. No data have been reported in this respect, so additional research is required.

Regarding region-based analysis, 18F-DOPA PET/CT was more suitable to identify metastatic lymph nodes, even those small in size (6 mm). These data are consistent with those published by Hoegerle, Treglia, and Archier previously [5, 12, 22]. In particular, in our study, 18F-FDOPA detected local tumor or metastatic lymph nodes in 53.8% of patients with a negative or inconclusive cervical ultrasound (7/13). Moreover, PET/CT with 18F-DOPA has also proven to be a sensitive technique to identify metastases of MTC in unusual locations; for example, two cases have been recently published in which peritoneal carcinomatosis and orbital metastasis were diagnosed using this method [23, 24]. In our series, 18F-DOPA PET/CT identified a liver metastasis not detected by 18F-FDG.

18F-DOPA PET/CT is a powerful tool that allows changes in patient management, and thus has an important clinical weight. In our series, eight out of 12 patients with a positive 18F-DOPA PET/CT underwent surgery with a curative intent, In the remaining three patients with pathological findings on this exam, TKIs were offered (progressive disease with distant metastases).

Although the ATA guidelines recommend expanding the morphological or functional study in addition to cervical ultrasound in patients with MTC and calcitonin levels above 150 pg/mL [1, 16], there is only one study so far that supports this recommendation [9]. Luster et al. in a retrospective study that included 28 patients found a 100% sensitivity and specificity of 18F-DOPA PET/CT with calcitonin levels over this cutoff [9]. However, these outcomes were not confirmed in subsequent studies [14, 22]. Sesti et al. tried to validate the calcitonin cutoff of 150 pg/mL, finding a lower sensitivity and specificity of the technique (79% and 80%, respectively) than Luster et al [14]. In a recent paper published by Archier et al. including 86 patients with MTC, 18F-DOPA PET/CT sensitivity was not affected by serum calcitonin values when using calcitonin as a continuous variable or in a percentile group analysis [22]. In the meta-analysis from Treglia et al., the sensitivity of 18F-DOPA PET or PET/CT was 73% if calcitonin values were above 150 pg/mL, but reached 86% when the Ct value rose over 1000 pg/mL or Ct-DT was below 24 months [15]. In our series we found that only one out of all the patients with Ct ≥ 150 pg/mL had a negative result with 18F-DOPA PET/CT, being the sensitivity of this test of 90.9% using this cutoff value. We achieved a sensitivity of 100% with Ct or CEA above 200 pg/mL and 7.2 ng/mL, respectively.

Concerning CEA values, Treglia et al. in their meta-analysis also found higher detection rates of 18F-DOPA PET or PET/CT when setting the cutoff of CEA at 5 ng/mL (detection rate of 64% with CEA > 5 ng/mL vs 48% with CEA ≤ 5 ng/mL) [15]. In this study, it seems that from this threshold of CEA, the diagnostic accuracy of 18F-DOPA and 18F-FDG PET/CT increase significantly (sensitivity of 81.1% and 72.7%, respectively, with CEA ≥ 5 ng/mL vs. 42.9% and 14.3% CEA < 5 ng/mL). Nevertheless, more studies to evaluate this cutoff value are needed.

Although we have not found a significant association between Ct or CEA doubling times and PET/CT positivity, we have seen a tendency to positivity with lower values of Ct-DT. Other studies have proven that the utility of 18F-FDG PET is higher when Ct-DT is less than 24 months, or CEA levels are rapidly rising [11, 13, 25], providing a prognostic value to 18F-FDG PET, with decreased survival or faster disease progression in patients showing uptakes. In our study, only one out of 11 patients with a negative 18F-FDG PET/CT result (patient number 6 of Table 2) developed a progressive disease with distant metastases during the follow-up. The rest of the 18F-FDG-negative cases had a persistent disease with stable or decreasing tumor markers. Four out of nine patients with pathological findings on 18F-FDG PET/CT had progressive disease (cases No. 3, 9, 11, and 16 in Table 2), one patient has died (No. 8), and the other remained stable or improved after surgery. Despite the limited number of patients in our study, it also seems that those with a positive 18F-FDG PET/CT result had a worse outcome.

Gallium-68-labeled somatostatin analogues represent a promising tool for evaluation of the expression of somatostatin receptors in neuroendocrine tumors [26]. The experience with 68Ga-labelled DOTA peptides in MTC is very limited [12, 27,28,29]. Conry et al. compared 68Ga-DOTATATE PET/CT and 18F-FDG PET/CT in 18 patients with recurrent MTC and found that the sensitivity of both tracers was similar (72% vs 78%, respectively) [27]. Ozkan et al. retrospectively evaluated the role of 68Ga-DOTATATE and 18F-FDG PET-CT in 22 patients with residual MTC [28]. They found a better performance of 68Ga-DOTATATE PET/CT than 18F-FDG PET/CT (sensitivity of 68.2% vs 44.4%, respectively) and conclude, as in other papers, that 68Ga-DOTATATE PET-CT is an efficient imaging modality in MTC patients with very increased calcitonin and CEA levels (especially >500-1000 pg/mL and 50 ng/mL, respectively) [28, 29]. However, the only study comparing 18F-FDG, 18F-DOPA and 68Ga-somatostatin analogues PET/CT imaging in patients with residual MTC (n=18) showed, as in our series, that 18F-DOPA PET/CT was the most useful imaging method for detecting recurrent MTC lesions (sensitivity of 72% for 18F-DOPA, 33% for 68Ga-somatostatin analogue and 17% for 18F-FDG PET/CT) [12]. The above-mentioned studies showed that 68Ga-somatostatin analogue PET/CT missed all the liver lesions detected by the other tracers, a finding that may be explained by the low lesion-to-background ratio due the low hepatic expression of somatostatin receptors and the physiological uptake of this tracer in the liver [12, 27]. Finally, 68Ga-DOTATATE PET/CT could be a useful complementary imaging method that could identify patients suitable for targeted radionuclide somatostatin analogue therapy (177Lutetium or 90Yttrium-labelled DOTATATE) [26].

Our study has several limitations. In first instance, the number of patients recruited for the study is small. Medullary thyroid cancer is a very rare neuroendocrine tumor and it is difficult to gather long series of cases. Second, it has not been possible to compare the results between PET/CT with 18F-DOPA and 18F-FDG and other imaging modalities in all patients. This is because this objective was retrospectively included and only some of the patients had cervical ultrasound and CT before PET/CT. Finally, the doubling times of calcitonin and CEA were not available in all cases. This fact justifies that when comparing the results of 18F-FDG and 18F-DOPA PET/CT depending on calcitonin and CEA doubling times no statistically significant differences were obtained.

In this and other studies [7,8,9,10], the 18F-DOPA PET/CT sensitivity to locate the disease seems to be superior to other morphological studies (CT). Taking into account these data, it would be reasonable to position 18F-DOPA PET/CT in the second step of diagnosis (after cervical ultrasound) when trying to locate the disease in patients with recurrent or persistent MTC.

In conclusion, 18F-DOPA PET/CT seems to be superior to 18F-FDG PET/CT in demonstrating recurrent or persistent MTC. The highest diagnostic performance of the technique would be obtained in patients with Ct levels ≥ 150 pg/mL and negative or ambiguous morphological exams. The sensitivity of 18F-FDG and 18F-DOPA PET/CT may be increased with a CEA cutoff of ≥ 5 ng/mL.