Abstract
The oncogene BRAFV600E is the most frequent genetic event in papillary thyroid carcinoma (PTC) but its prognostic impact still remains to be elucidated. We evaluated a representative series of 67 individuals with PTC who underwent total thyroidectomy. BRAF-positive tumours correlated with early recurrences (32% vs 7.6%; P=0.02) during a median postoperative follow-up period of 3 years. Interestingly, within the recurrences, a significant majority had negative radioiodine (131I) total body scans, predicting a poorer outcome as treatment with 131I is not effective. This last observation led us to investigate the role of BRAFV600E and the MEK-ERK pathway in thyroid dedifferentiation, particularly in Na+/I− symporter (NIS) impairment, as this thyroid-specific plasma membrane glycoprotein mediates active transport of I− into the thyroid follicular cells. A subset of 60 PTC samples was evaluated for NIS immunoreactivity and, accordingly, we confirmed a significant low NIS expression and impaired targeting to membranes in BRAF-positive samples (3.5% vs 30%; P=0.005). Furthermore, experiments with differentiated PCCl3 thyroid cells demonstrated that transient expression of BRAFV600E sharply impaired both NIS expression and targeting to membrane and, surprisingly, this impairment was not totally dependent on the MEK-ERK pathway. We have concluded that BRAFV600E is a new prognostic factor in PTC that correlates with a high risk of recurrences and less differentiated tumours due to the loss of NIS-mediated 131I uptake.
Introduction
Papillary thyroid carcinoma (PTC) is the most frequent malignancy of the thyroid, accounting for 1% of all human malignancies. In general terms, it has a good prognosis and high cure rates are achieved after initial treatment. However, it has been estimated that about 20–30% of patients with PTC will develop a local or distant recurrence but only about 1% will die (Schlumberger & Pacini 2003a). Identifying these high-risk patients at the time of diagnosis through well-established prognostic factors can help to ascertain the most appropriate treatment and follow-up for these patients. Several prognostic scoring systems, such as the tumour node metastasis (TNM) classification (American Joint Committee on Cancer (AJCC) 2002), have been developed for thyroid cancer and they are based on multiple regression analysis of combined prognostic factors. All of them include extrathyroidal extension and distant metastasis, and practically all are based on age at diagnosis, tumour size and histological type. Despite all the former epidemiologic data, there is still a debate on the definition and best management of low-risk versus high-risk patients, and each centre has its own protocols (Schlumberger & Pacini 2003a).
To date, no biological marker has yet been included as a prognostic factor for thyroid cancer. BRAF is a serine-threonine kinase that mediates signal transduction through the MEK-ERK pathway, and the T transversion to A at 1796 creates a constitutive active molecule (BRAFV600E) that has been proved to be an oncogene in human cancer (Davies et al. 2002, Garnett & Marais 2004). Since the oncogene BRAFV600E has become the most common genetic event associated with PTC, some studies, but not all, have recently found BRAF to be associated with aggressive clinicopathological features such as advanced clinical stages and extrathyroidal extension (Namba et al. 2003, Nikiforova et al. 2003). BRAF has also been proved to be present in poorly differentiated and anaplastic carcinomas that arise from a PTC (Nikiforova et al. 2003, Soares et al. 2004, Quiros et al. 2005). Finally, targeted expression of BRAFV600E in thyroid cells of transgenic mice resulted in invasive PTCs that underwent dedifferentiation (Knauf et al. 2005). Thus, the biological behaviour of BRAF-positive tumours seems to be more aggressive than its partnerships in thyroid tumourigenesis, RET and RAS, in which no clear evidence of aggressiveness has been found. Nevertheless, the prognostic impact of BRAF remains to be elucidated, since at least four studies have not found any association with aggressive clinicopathological features (Xu et al. 2003, Fugazzola et al. 2004, Puxeddu et al. 2004, Trovisco et al. 2005) and the risk of recurrence or mortality have not been studied. In this retrospective study, we have demonstrated that BRAFV600E is associated with a high recurrence rate during the early follow-up of a representative series of 67 patients with PTC. Interestingly, the majority of these recurrences had no avidity for radioiodine (131I), being ineffective by131I treatment and pointing to a less differentiated state.
In order to confirm this less differentiated state associated with BRAF-positive tumours, we studied the specific role of BRAFV600E in thyroid dedifferentiation, particularly Na+/I− symporter (NIS) impairment. NIS is an integral plasma membrane glycoprotein that mediates active I− transport into the thyroid follicular cells, the first step in thyroid hormone biosynthesis, and provides the basis for the effective diagnosis and therapeutic management of thyroid cancer and its metastases (Dohan et al. 2003). Assessment of NIS expression by immunohistochemistry (IHC) in human tumour samples and transfection experiments in rat thyroid cells demonstrated that BRAFV600E sharply impairs both NIS expression and trafficking to membrane. Surprisingly, this impairment is not dependent on MEK-ERK pathway activation. This last observation may have special significance as pharmacological therapies targeted to inhibit the MEK-ERK pathway will not be sufficient to re-differentiate tumours with constitutive activation of BRAF.
Materials and Methods
Subjects
A representative series of 67 patients with PTC who underwent surgical resection during the period 2000–2003 was selected from the Hospital Universitario La Paz (Madrid, Spain). The study protocol was approved by the Hospital Human Ethics Review Committee. The mean age of the patients was 42.8 ± 14 years and the female to male ratio was 3 : 1. No differences between patients existed as regards initial treatment or follow-up. Initial treatment in every patient consisted of total thyroidectomy, 131I ablation and thyrotrophin (TSH) suppression. When suspicious neck lymph nodes were noted preoperatively or at the time of surgery they were removed. Follow-up was carried out periodically with 131I total body scan (TBS) and serum thyroglobulin (Tg) (Delphia thyroglobulin kit; Perkin-Elmer, Wellesley, MA, USA), and in some selective cases cervical ultrasonography (US) was performed. We defined a low Tg as being less than 2 ng/ml when TSH was suppressed. Remission after thyroid surgery was considered when (i) no 131I uptake was found outside the thyroid bed on the postablation TBS, (ii) serum Tg levels remained undetectable following TSH stimulation and (iii) eventually no 131I uptake was seen on control 131I TBS. After thyroid ablation by surgery and 131I treatment, low uptake in the thyroid bed was not considered evidence of recurrent disease and did not warrant further treatment. Preoperative serum TSH levels were assessed in almost every patient, who were euthyroid at the time of surgery in all cases except one.
Clinical outcome
Clinical outcome was carefully reviewed, looking for any locoregional or distant recurrence. Preferentially, histology or cytology was needed to confirm recurrent disease. However, as this is not always possible, any elevation of serum Tg associated with positive 131I TBS and/or any other abnormal imaging study, including US, computerized tomography (CT) or 8F-deoxyglucose positron emission tomography (FDG-PET), were considered as showing recurrent disease. In all cases except two, Tg levels were measured during thyroxine withdrawal and significant elevations of TSH levels (>30 μUI/dl) were obtained. In the other two cases, the levels of Tg were measured during TSH suppressive therapy, but they were sufficiently high to consider them positive.
Tumour samples
Paraffin-embedded tissues were obtained from the Department of Pathology of the aforementioned hospital. Clinical and pathological staging were carried out according to the TNM classification of the AJCC (2002). After the initial review and selection, glass slides from all carcinomas were re-examined by two independent pathologists who were blinded as to BRAF status and all other patient characteristics, and were subclassified as classic papillary carcinoma or as distinct histological variants based on the histopathological typing of the World Health Organization (2004).
DNA isolation, single-strand conformational polymorphism (SSCP) and sequencing
Genomic DNA was isolated using proteinase K digestion, phenolchloroform extraction and ethanol precipitation as previously described (Nikiforov et al. 1996). BRAF exon 15 was amplified by PCR. The following exon-based PCR primers were designed to amplify exon 15: forward, CAT AAT GCT TGC TCT GAT AGG and reverse, GTA ACT CAG CAG CAT CTC A. PCR conditions were as follows: amplifications were carried out for 40 cycles with an annealing temperature of 58 °C. Fifty microlitre PCRs were performed on 200–300 ng genomic DNA, 7.5 pmol of each primer, 100 μM dNTPs, 5 μCi [α32P]dCTP, 1.5 mM MgCl2, TaqDNA polymerase high fidelity (Biotools, Madrid, Spain) and buffer. The amplified products were screened for mutation by SSCP (polyacrylamide 10%, without glycerol). Subsequently, amplified products from aberrant SSCP bands were purified using a PCR purification kit (Qiagen) and were sequenced using an automatic sequencer (ABI PRISM 3100; Applied Biosystems, Foster City, CA, USA).
Immunohistochemistry
A monoclonal antibody against the carboxy-terminal portion of human (hNIS) was used (Pohlenz et al. 2000). Thyroid tissue sections were studied using the catalyzed signal amplification protocol (DAKO Corp., Barcelona, Spain). Sections (4 μm) were mounted on charged slides. All sections were baked at 60 °C for 30 min. Slides were washed with three changes of xylene and hydrated through alcohol to distilled water. Antigen retrieval was performed using 10% citrate buffer in a steamer for 40 min, and rapid cooling was achieved with distilled water. Tissues were incubated in 3% peroxide for 15 min to quench endogenous peroxidase. Sections were blocked with serum-free protein, and endogenous biotin and avidin activity was blocked with the biotin blocking system (DAKO Corp.) All washes were performed with TBST (0.3 M NaCl, Tween 20 and 0.05 M Tris–HCl, pH 7.6) three times for 5 min each time. Slides were incubated for 30 min with human anti-NIS antibody diluted (1/60) in serum-free protein block. The strepavidin–biotin method as specified by the supplier (Dako) was followed. Peroxidase activity was detected with diamino-benzidene-hydrogen peroxide and was observed as a brown product.
Interpretation and grading of NIS staining was carried out by two independent pathologists. Immunoreactivity was characterized as negative (score=0), absent or not interpretable (score=1), weak positive (score=2) or strong positive (score=3) as described by Wapnir et al.(2003). Briefly, positive samples, either weak or strong, had to encompass at least 20% of cells to receive this overall score. When plasma membrane immunoreactivity was noticed, it was always scored as strong if 10% or more of cells demonstrated this feature either alone or in the presence of intracellular immunoreactivity. The main criterion to score NIS immunoreactivity as strong was the presence of plasma membrane immunoreactivity, as it is essential for NIS to be functional.
Cell culture
PCCl3 thyroid cells were cultured in Coon’s modified Ham’s F-12 medium supplemented with 5% donor calf serum and a six hormone mixture necessary for the growth of the thyroid cells (1 nM TSH, 10 μg/ml insulin, 10 ng/ml somatostatin, 5 μg/ml transferrin, 10 nM hydrocortisone and 10 ng/ml glycyl-l-histidyl-l-lysine actetate; complete medium). To study the effect of TSH on NIS regulation, cells were also cultured in the same medium without TSH for different periods of time, as indicated in each experiment.
Plasmids
The following promoters fused to luciferase were used: pNIS-2.8 (Garcia & Santisteban 2002), p420 (thyro-peroxidase) TPO–(luciferase) LUC (Aza-Blanc et al. 1993), minimal pTSH receptor (TSHR) (Civitareale et al. 1993) and pTg (Garcia-Jimenez et al. 2005). The expression vector pMCEF, harbouring the myc-tagged BRAFV600E and BRAF wild type (wt), is described by Marais et al.(1996). PRL-TK, which contains a cDNA encoding Renilla (Promega), was used to monitor transfection efficiency.
Transfection assays
PCCl3 cells were plated at 6×105 cells per 60 mm diameter tissue culture dish 48 h before transfection. Transfection assays were performed with calcium phosphate coprecipitation and in some cases the Fugene lipid reagent (Roche) was used. Cells were collected for a LUC and Renilla activity assay using the dual-luciferase reporter assay system (Promega). In cotransfection experiments, the amount of DNA was normalized using the corresponding insertless expression vector as the carrier. The experiments were performed in triplicate.
Immunoblot analysis
Cells were harvested with RIPA buffer (PBS, 1% Nonidet, 0.5% sodium deoxycholate, 0.1% SDS) and proteinase inhibitors. In the next step, 40 μg of whole cell lysates were separated by electrophoresis in 10% SDS-PAGE, and then blotted onto nitrocellulose membrane (Protran; Schleicher & Schuell, Dassel, Germany). To quantitate the levels of ERK and phospho (P)-ERK, the blots were incubated for 60 min with the respective antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The antigen–antibody complexes were visualized with horseradish peroxidase-conjugated anti-rabbit IgG antibody by the enhanced chemiluminescence system (Amersham Pharmacia Biotech). A polyclonal rat NIS antibody (Wapnir et al. 2003) was used for NIS detection.
Immunofluorescence
Cells grown on coverslips were fixed in paraformaldehyde at room temperature for 15 min and stained for anti-NIS and anti-myc (Santa Cruz Biotechnology). The secondary antibodies used were anti-rabbit Alexa 594, anti-rabbit Alexa 488 (Molecular Probes, Paisley, UK). The cells were mounted with medium containing DAPI (Vectashield; Vector Laboratories, Peterborough, UK) and preparations were visualized with a Leica confocal TCS SP2 microscope.
Statistical analysis
Data were stored and analyzed using the SPSS software (version 12.0). Association between recurrent disease, BRAFV600E and the histological variants or clinicopathological parameters of the thyroid tumours was determined by a χ2 test. Statistical significance was based on P<0.05.
Results
BRAFV600E is associated with some aggressive clinicopathological features
The BRAF mutation was detected in 41.7% (28 of 67) of PTCs. Relationships between BRAFV600E and clinicopathological variables are shown in Table 1. Significant association was seen between BRAFV600E and both extrathyroidal extension and advanced AJCC stages. There was no significant relationship between BRAF mutation and gender, age >45years and nodal or distant metastases at the time of diagnosis. Tumours lacking BRAF mutation were significantly associated with the follicular variant and, conversely, BRAFV600E was more frequent in the classic variant, although not statistically significant. Also, BRAF mutation was significantly associated with the tall cell variant.
BRAFV600E is associated with a high recurrence rate, particularly with negative 131I TBS
Locoregional or distant recurrences occurred in 12 PTC patients (18%) at a median postoperative follow-up period of 3 years. Table 2 shows the univariate relationship between several clinicopathological features and the recurrence rate. Extrathyroidal extension, advanced AJCC stages and lymph node metastases were associated with a high recurrence rate, whereas a lower but not significant recurrence rate was observed in the follicular variant. Interestingly, there was a significant association between BRAF mutation and cancer recurrence (32% vs 7.6%; P=0.02). Table 3 shows the main characteristics of the 12 patients who developed recurrences after a median postoperative follow-up period of 3 years. As mentioned before, patients with any elevation of Tg associated with positive 131I TBS and/or any abnormal imaging study were considered to have recurrent disease. Seven patients had persistent elevation of serum Tg but were negative for 131I TBS and therefore additional imaging studies were needed. FDG-PET was performed in four of them, finding anatomical substrate in three. Interestingly, the frequency of recurrences with negative 131I TBS was higher in the BRAF-positive group (66% vs 33%), predicting a worse outcome and indicating a less differentiated state. Statistical significance was not achieved, presumably because the number of recurrences needs to be higher.
Low NIS immunoreactivity in BRAF-positive tumours
As BRAF-positive recurrences seem to have no avidity for 131I uptake, we further studied NIS immunoreactivity in a subset of 60 tumour samples and eight correspondent lymph node metastases. Previous studies have revealed that up to 70–80% of thyroid cancer express or even overexpress NIS in IHC profiles (Wapnir et al. 2003). In our study, tumours harbouring the BRAF mutation expressed significantly less NIS immunoreactivity than those without the mutation. The NIS staining was negative or weak positive and not targeted to the membrane in a significant majority of the samples within the BRAF-positive group compared with the BRAF-negative group (2.5% vs 30%; P=0.005). In addition, strong positive samples that had membrane staining in at least 10–20% of the cells were observed predominantly in the follicular variant, yet this was not statistically significant. When concomitant node metastases at initial diagnosis were analyzed (n=8), NIS staining was concordant to its primary tumours in all cases. Figure 1 shows several tissue samples with different patterns of NIS immunoreactivity.
BRAFV600E impairs NIS promoter transcriptional activity in PCCl3 thyroid cells
We have already observed that lack of 131I uptake and low NIS immunoreactivity are associated with BRAFV600E. We therefore wanted to study if BRAFV600E affects NIS transcriptional activity in differentiated PCCl3 thyroid cells. Indeed, BRAFV600E dramatically decreased the NIS promoter transcriptional activity (>90% reduction). We also investigated the effect on other thyroid specific promoters. A smaller effect was observed on TPO promoter (60% reduction) and TSHR promoter (50% reduction), being almost undetectable on Tg promoter (10% reduction) (Fig. 2A). As BRAF mediates signal transduction through the MEK-ERK pathway, we used the MEK inhibitor U0126 to study whether this pathway was involved in NIS transcriptional impairment induced by BRAFV600E. As shown in Fig. 2B, BRAFV600E repression of NIS was partially reversed when MEK was inhibited, proving a limited role of the MEK-ERK pathway.
BRAFV600E impairs NIS trafficking to the membrane in PCCl3 thyroid cells
Active I− transport into fully differentiated PCCl3 cells depends on NIS localization in the plasma membrane. As previously described (Dohan & Carrasco 2003) and shown in Fig. 3A, this process is dependent on TSH. After TSH withdrawal, NIS targeting to membrane was impaired immediately (Fig. 3A, left panel) whereas NIS protein expression diminished progressively in 7 days and higher molecular bands became predominant (Fig. 3A, right panel), confirming the previous suggestions that postranslational mechanisms dependent on TSH are responsible for NIS targeting to the membrane (Dohan & Carrasco 2003). Interestingly, when PCCl3 cells were transiently transfected with BRAFV600E a very similar pattern was observed: NIS targeting to membrane was impaired immediately (Fig. 3B, left panel), protein expression decreased progressively and higher molecular bands became predominant (Fig. 3B, right panel), suggesting that NIS impairment is due to BRAF interference of TSH-mediated responses. To study whether the TSHR by itself or its distal signalling was affected, we stimulated PCCl3 cells expressing BRAFV600E with a cAMP inductor (forskolin). NIS expression was partially recovered but was not targeted to the membrane (Fig. 3C, left and right panels) suggesting that NIS impairment by BRAFV600E occurs, at least in part, downstream of the TSHR/cAMP signalling.
Limited role of the MEK-ERK pathway in NIS impairment induced by BRAFV600E
We have already described the limited role of the MAP kinase (MAPK) pathway on the impairment of NIS promoter activity by BRAFV600E (Fig. 2B). We further studied NIS protein expression and its subcellular localization. Although we observed partial reinduction of NIS protein expression using a MEK inhibitor, no relocalization to the membrane was observed, even when we concomitantly stimulated with forskolin (Fig. 3C, left and right panels). To further confirm these results, PCCl3 cells stably transfected with the oncogene Kirsten (K)-RAS were used. These fully transformed cells have constitutive activation of the MEK-ERK pathway and do not express NIS (Santoro et al. 1993). Again, the MEK inhibitor partially reinduced NIS protein expression but no targeting to membrane was observed (Fig. 3D, left and right panels). This indicates a very limited role of the MEK-ERK pathway in thyroid BRAF-induced dedifferentiation, especially in NIS impairment.
Discussion
The question as to whether BRAF confers a distinct biological behaviour to PTC that could have a prognostic value is still controversial. The inconsistent results described by several groups (Soares et al. 2003, Xu et al. 2003, Fugazzola et al. 2004, Trovisco et al. 2005) could be partially due to an insufficient number of patients and/or to the different combinations of various subtypes of PTC included in each study. Subtype stratification is likely to be important as BRAF is present predominantly in the classic and tall cell variants of PTC and not in the follicular variant. In our study, with classic PTC accounting for more than 50% of the total, BRAF mutation predicted an increased risk of recurrence. The univariate analysis also revealed that BRAF was associated with extra-thyroidal extension and advanced clinical stage. Indeed, larger series including mortality rate and multivariate analysis with adjustment for various confounding factors will reveal the independent prognostic role of BRAF. We not only demonstrated that BRAF-positive tumours have more risk of recurrence, but also that they are likely to be less differentiated as 131I uptake is absent in the majority of the BRAF-positive recurrences. This is challenging for thyroid cancer management as anatomical localization of recurrences cannot be assessed and treatment with ablative doses of 131I is not effective, predicting a poorer outcome. As we can see in Table 2, six out of nine (66%) BRAF-positive recurrences were negative for 131I scans and residual disease was finally located performing FDG-PET in three of them. This is interesting, as the switch of iodide uptake into high FDG uptake due to enhanced glucose metabolism has been proposed as a model of tumoral dedifferentiation in thyroid cancer (Schlumberger & Pacini 2003b). Several studies have proved the utility of FDG-PET during the postoperative management of thyroid cancer, particularly in patients with elevated serum Tg and negative 131I scans (Hooft et al. 2001). Perhaps, in the future, FDG-PET may have an important role in the initial management of patients with BRAF-positive tumours because of their high risk of less differentiated recurrences.
Concordantly, we have also observed an association between BRAF-positive tumours and low NIS immunoreactivity in a subset of 60 paraffin-embedded samples. In previous studies (Saito et al. 1998, Dohan et al. 2001, Wapnir et al. 2003), assessment of NIS expression by IHC revealed that up to 70–80% of thyroid cancers expressed or even overexpressed NIS, yet this expression was mainly cytoplasmatic and not targeted to the basolateral membrane. It has been postulated that targeting to and retention in the plasma membrane is essential for NIS to be fully functional, explaining why I− uptake is diminished in thyroid cancer (Dohan & Carrasco 2003). Additionally, positive NIS immunoreactivity in primary tumours seems to be predictive of subsequent recurrences positive in 131I scans (Castro et al. 2001, Schmitz et al. 2005), whereas low NIS expression assessed by RT-PCR has been correlated with more aggressive tumours in another study (Ward et al. 2003). Overall, NIS expression may have a role as a new biological marker in the postoperative management of patients with change to thyroid cancer (CDT). In our study, the tumours harbouring the mutation had significantly less NIS immunoreactivity and this is consistent with our previous data that suggest that BRAF-positive tumours have lost their ability to trap 131I and, thus, they are less differentiated. Additional findings, such as the association of strong NIS staining with the follicular variant, suggest that cellular polarity may play an important role for NIS to be functional. In addition, we have observed that low NIS staining is associated with advanced stages in the TNM classification (data not shown). However, the main limitation to establish NIS as a biological marker is methodological. Highly sensitive visualization systems are required and NIS antibodies are still poorly developed.
We also analyzed the specific effects of BRAFV600E in a well-differentiated rat cell line (PCCl3), which expresses functional NIS in the membrane. In a previous study (Mitsutake et al. 2005), mRNA levels of NIS were decreased by conditional expression of BRAFV600E. In our study, we observed a marked decrease of the transcriptional activity of the NIS promoter when PCCl3 cells were transfected with BRAFV600E compared with wt. By contrast, TPO and TSHR decreased more moderately, and Tg even more slightly, which may reflect the fact that thyroid cancer is a multistep process where BRAF-positive tumours are still differentiated although to a lesser extent, and more steps are required in order to evolve to a poor or anaplastic thyroid cancer, where none of the thyroid specific genes are any longer expressed. Secondly, BRAFV600E immediately impairs NIS targeting to the membrane and progressively decreases NIS protein expression in the same fashion as does TSH withdrawal. This suggests that the transcriptional and postranslational NIS modifications are due to BRAF interference of TSH-mediated responses. The fact that the impaired NIS protein has a distinct molecular weight, in this case higher, suggests that different patterns of glycosylation and especially phosphorylation are taking place. As previously described (Mitsutake et al. 2005), our study also shows that BRAF activation impairs cAMP-induced expression of NIS, although there is a partial recovery that it is not targeted to the membrane. This suggests that the effects of BRAF on NIS expression cannot be due only to a decreased abundance of TSHR, and other distal steps at a transcriptional level seem to be affected by BRAFV600E in TSH-mediated responses.
Finally, the role of the MAPK pathway in thyroid dedifferentiation induced by BRAFV600E, particularly NIS, seems to be small and limited to a transcriptional level. It is worth noting that the MAPK pathway seems to play a central role in PTC tumourigenesis as RET/PTC, RAS and BRAF are mutually exclusive genetic events, all of which activate this pathway. A recent report has demonstrated that RET/PTC induces RAS-and BRAF-dependent ERK activation, pointing out a linear oncogenic signalling cascade that governs proliferation and invasion in transformed thyroid cells (Melillo et al. 2005). Moreover, Knauf et al.(2003) have reported that MEK inhibitors increase NIS mRNA levels in thyroid cells expressing RET/PTC or RASV12 and they have suggested that inhibiting the MEK-ERK pathway may promote redifferentiation in tumours with constitutive activation of either RAS or RET/PTC (Knauf et al. 2003). In our study, although we found a partial reinduction in NIS expression with a MEK inhibitor in thyroid cells expressing BRAFV600E, this NIS reinduction was low (see Figs 2 and 3) and, most importantly, was unable to target to the membrane, and consequently was not functional. Presumably BRAF acts either directly or through pathways others than MEK-ERK in inducing thyroid dedifferentiation, as constitutive activation of MEK-ERK does not fully explain the NIS modifications at transcriptional and postransductional levels. Therefore therapies that target inhibition of the MEK-ERK pathway may not be able to redifferentiate tumours with constitutive activation of BRAF in order to express NIS in the membrane. In addition, this also underscores that, although the three oncoproteins work together along a single cascade, they are able to trigger specific signals and therefore confer distinct biological behaviour.
Overall, BRAF confers more aggressiveness to the biological behaviour of PTC, as early recurrences are more frequent and tumours are less differentiated, predicting poorer outcomes as treatment with 131I is not useful. This can help clinicians distinguish between high-risk and low-risk patients at the time of diagnosis. Although more studies, including larger series, longer follow-ups and mortality are needed, we believe that the preoperative assessment of BRAF status can improve the subsequent management of thyroid cancer, as more extensive surgery can be performed as well as a more exhaustive follow-up, including FDG-PET, can be considered (Soares et al. 2003).
Relationships between the status of BRAF mutations and clinicopathological features of papillary carcinomas
| BRAF positive (%) n=28 | BRAF negative (%) n=39 | P | |
|---|---|---|---|
| Age >45 years | 15 (53.5) | 16 (41) | 0.33 |
| Male | 4 (14.2) | 8 (20.5) | 0.74 |
| Histologic variants of PTC | |||
| Classic | 18 (64.2) | 17 (43.5) | 0.2 |
| Follicular variant | 5 (17.8) | 20 (51.2) | 0.01 |
| Tall cells | 4 (14.2) | 1 (2.5) | 0.09 |
| Columnar cells | 1 (3.5) | 1 (2.5) | |
| Oncocitic | 1 (3.5) | 0 (0) | |
| Multicentric | 4 (14.2) | 12 (30) | 0.15 |
| Extrathyroidal extension | 18 (64) | 11 (28) | 0.003 |
| Node metastases | 9 (32) | 9 (23) | 0.42 |
| Distant metastases | 1 (3.5) | 1 (2.5) | |
| Stage | |||
| I–II | 5 (17.8) | 23 (60) | 0.01 |
| III–IV | 23 (82) | 16 (41) | |
Univariate relationships with recurrence rate
| Recurrence rate (%) | P | |
|---|---|---|
| NS, not significant. | ||
| Age at time of diagnosis | ||
| >45 | 19.4 | |
| <45 | 16.7 | NS |
| Gender | ||
| Male | 25 | |
| Female | 16.4 | NS |
| PTC variants | ||
| Classic | 25 | |
| Follicular | 4 | |
| Tall cells | 40 | NS |
| Multicentric | ||
| Yes | 25 | |
| No | 15 | NS |
| Extrathyroidal extensión | ||
| Yes | 35 | |
| No | 5.3 | 0.003 |
| Lymph nodes | ||
| Yes | 44.4 | |
| No | 8.2 | 0.002 |
| Stage | ||
| I–II | 3.6 | |
| III–IV | 28.2 | 0.01 |
| BRAF | ||
| wt | 7.6 | |
| Mutant | 32.1 | 0.02 |
Main characteristics of the 12 patients with thyroid cancer recurrences
| Patient no. | Age1 | Age2 | G | Surgery | Subtype histology | Stage | ExExt | TotDo | Serum Tg (ng/ml) | TSH (μUI/dl) | TBS | Additional findings | BRAF status |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age1, age at diagnosis; Age2, age at recurrence; G, gender; T.T, total thyroidectomy; LNS, lymph node surgery; S.T., subtotal thyroidectomy; LR, local recurrence; Med, mediastinum; NEG, negative; POS, positive, ExExt, extrathyroid extension; TotDo, total cumulative dose (mCi).; TC, computerised tomography; FNAC, fine needle aspiration cytology; Mts, metastasis. | |||||||||||||
| *The histological data provided by the pathologist confirmed the presence of thyroid cancer recurrence in these lesions. | |||||||||||||
| Case 1 | 63 | 67 | F | T.T. | Tall cells | IV | Yes | 400 | 105 | 100 | LR | US: cervical nodes | |
| *Extirpation of cervical nodes | POS | ||||||||||||
| Case 2 | 36 | 37 | F | T.T. | Classic | I | Yes | 250 | 4 | 182 | LR | Partial response to I13I ablation | POS |
| Case 3 | 27 | 28 | F | T.T. | Classic | I | No | 300 | 15.5 | 98 | NEG | US: cervical nodes | POS |
| FDG-PET: negative | |||||||||||||
| Case 4 | 75 | 80 | F | S.T.+LNS | Classic | III | No | 250 | 4.6 | 1.75 | NEG | TC: cervical nodes | POS |
| *Extirpation cervical nodes | |||||||||||||
| Case 5 | 40 | 41 | F | T.T. | Classic | I | No | 250 | 8.08 | 100 | LR+Med | NEG | |
| Case 6 | 40 | 41 | T.T.+LNS | Follicular | I | Yes | 100 | 4.3 | 0.01 | NEG | FDG-PET: pulmonar nodule | POS | |
| Case 7 | 47 | 49 | M | T.T.+LNS | Classic | III | Yes | 150 | 11.6 | >100 | NEG | US: cervical node | |
| FDG-PET:cervical, clavicular and axilar | |||||||||||||
| *Extirpation cervical nodes | NEG | ||||||||||||
| Case 8 | 67 | 68 | F | T.T. | Classic | III | Yes | 200 | 14.5 | 80 | NEG | TC: cervical nodes | |
| *Extirpation cervical nodes | POS | ||||||||||||
| Case 9 | 38 | 39 | F | T.T. | Classic | I | No | 215 | 2.5 | 63 | LR | Good response to 131I ablation | POS |
| Case 10 | 38 | 39 | M | T.T.+LNS | Tall cells | IV | Yes | 200 | 2.17 | 90 | LR | FDG-PET: cervical and mediastinical nodes TC: diffuse pulmonary Bone scan: mts | NEG |
| Case 11 | 31 | 33 | F | T.T. | Classic | I | Yes | 100 | 62 | 85 | NEG | FDG-PET: cervical nodes | POS |
| Case 12 | 61 | 62 | M | T.T. | Classic | IV | Yes | 100 | 32 | 95 | NEG | TAC: cervical | POS |
| FNDA: positive | |||||||||||||
| *Extirpation cervical nodes | |||||||||||||
NIS expression in several samples of PTC. Sections (4 μm) were probed with anti-NIS antibody. Scoring system: 0=negative, 2=weakly positive and 3=strongly positive. Almost all strongly positive samples had membrane staining in at least 10% or more of the cells. (A) Normal thyroid tissue showing plasma membrane immunoreactivity in 10–20% of the follicular cells (score 3; original magnification 20×). (B) Graves’ disease showing follicular hyperplasia and predominant plasma membrane staining in more than 90% of the cells, used as a positive control (score 3; original magnification 20×). (C) Classic PTC with intracellular immunoreactivity and distinct plasma membrane in >10% of the cells (score 3; original magnification 40×). (D) Follicular variant of PTC with intracellular immunoreactivity and distinct plasma membrane in >10% of the cells (score 3; original magnification 20×). (E) Several cells with distinct plasma membrane in a follicular variant of PTC (score 3; original magnification 40×). (F) Lymphatic node with metastases of PTC showing distinct plasma membrane immunoreactivity (score 3; original magnification 20×). Scale bars=50 μM.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
NIS expression in several samples of PTC. Sections (4 μm) were probed with anti-NIS antibody. Scoring system: 0=negative, 2=weakly positive and 3=strongly positive. Almost all strongly positive samples had membrane staining in at least 10% or more of the cells. (A) Normal thyroid tissue showing plasma membrane immunoreactivity in 10–20% of the follicular cells (score 3; original magnification 20×). (B) Graves’ disease showing follicular hyperplasia and predominant plasma membrane staining in more than 90% of the cells, used as a positive control (score 3; original magnification 20×). (C) Classic PTC with intracellular immunoreactivity and distinct plasma membrane in >10% of the cells (score 3; original magnification 40×). (D) Follicular variant of PTC with intracellular immunoreactivity and distinct plasma membrane in >10% of the cells (score 3; original magnification 20×). (E) Several cells with distinct plasma membrane in a follicular variant of PTC (score 3; original magnification 40×). (F) Lymphatic node with metastases of PTC showing distinct plasma membrane immunoreactivity (score 3; original magnification 20×). Scale bars=50 μM.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
NIS expression in several samples of PTC. Sections (4 μm) were probed with anti-NIS antibody. Scoring system: 0=negative, 2=weakly positive and 3=strongly positive. Almost all strongly positive samples had membrane staining in at least 10% or more of the cells. (A) Normal thyroid tissue showing plasma membrane immunoreactivity in 10–20% of the follicular cells (score 3; original magnification 20×). (B) Graves’ disease showing follicular hyperplasia and predominant plasma membrane staining in more than 90% of the cells, used as a positive control (score 3; original magnification 20×). (C) Classic PTC with intracellular immunoreactivity and distinct plasma membrane in >10% of the cells (score 3; original magnification 40×). (D) Follicular variant of PTC with intracellular immunoreactivity and distinct plasma membrane in >10% of the cells (score 3; original magnification 20×). (E) Several cells with distinct plasma membrane in a follicular variant of PTC (score 3; original magnification 40×). (F) Lymphatic node with metastases of PTC showing distinct plasma membrane immunoreactivity (score 3; original magnification 20×). Scale bars=50 μM.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
BRAFV600E impairs NIS promoter transcriptional activity. (A) PCCl3 cells were co-transfected with expression vectors harbouring wt or BRAFV600E mutation, together with the promoter constructs of NIS, TPO, TSHR or Tg genes respectively. (B) The MEK inhibitor U0126 was used to study the MEK-ERK pathway, in this case DMSO was used as a control. Forty-eight hours after transfection the cells were harvested and luciferase and Renilla were assayed. The promoter activity is expressed as relative luciferase activity over control cells transfected with empty vector.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
BRAFV600E impairs NIS promoter transcriptional activity. (A) PCCl3 cells were co-transfected with expression vectors harbouring wt or BRAFV600E mutation, together with the promoter constructs of NIS, TPO, TSHR or Tg genes respectively. (B) The MEK inhibitor U0126 was used to study the MEK-ERK pathway, in this case DMSO was used as a control. Forty-eight hours after transfection the cells were harvested and luciferase and Renilla were assayed. The promoter activity is expressed as relative luciferase activity over control cells transfected with empty vector.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
BRAFV600E impairs NIS promoter transcriptional activity. (A) PCCl3 cells were co-transfected with expression vectors harbouring wt or BRAFV600E mutation, together with the promoter constructs of NIS, TPO, TSHR or Tg genes respectively. (B) The MEK inhibitor U0126 was used to study the MEK-ERK pathway, in this case DMSO was used as a control. Forty-eight hours after transfection the cells were harvested and luciferase and Renilla were assayed. The promoter activity is expressed as relative luciferase activity over control cells transfected with empty vector.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
BRAFV600E impairs NIS targeting to plasma membrane through pathways others than MEK-ERK. Immunofluorescence assays (left) and Western blot analysis (right) performed in differentiated PCCl3 or K-RAS-transformed cells to determine NIS subcellular distribution and protein levels. (A) Effect of TSH withdrawal, (B) effect of BRAFV600E over-expression, (C) effect of a cAMP inductor (forskolin (Forsk)) and an MEK inhibitor U0126 and (D) effect of U0126 over K-RAS-transformed PCCl3 cells. Immunofluorescence with myc antibody was performed as the control of BRAF transfection efficiency. In all cases the Western blots were also hybridized with anti-P-ERK and ERK antibodies. The arrows in the Western blot indicate the high molecular weight bands of NIS protein. d, day.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
BRAFV600E impairs NIS targeting to plasma membrane through pathways others than MEK-ERK. Immunofluorescence assays (left) and Western blot analysis (right) performed in differentiated PCCl3 or K-RAS-transformed cells to determine NIS subcellular distribution and protein levels. (A) Effect of TSH withdrawal, (B) effect of BRAFV600E over-expression, (C) effect of a cAMP inductor (forskolin (Forsk)) and an MEK inhibitor U0126 and (D) effect of U0126 over K-RAS-transformed PCCl3 cells. Immunofluorescence with myc antibody was performed as the control of BRAF transfection efficiency. In all cases the Western blots were also hybridized with anti-P-ERK and ERK antibodies. The arrows in the Western blot indicate the high molecular weight bands of NIS protein. d, day.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
BRAFV600E impairs NIS targeting to plasma membrane through pathways others than MEK-ERK. Immunofluorescence assays (left) and Western blot analysis (right) performed in differentiated PCCl3 or K-RAS-transformed cells to determine NIS subcellular distribution and protein levels. (A) Effect of TSH withdrawal, (B) effect of BRAFV600E over-expression, (C) effect of a cAMP inductor (forskolin (Forsk)) and an MEK inhibitor U0126 and (D) effect of U0126 over K-RAS-transformed PCCl3 cells. Immunofluorescence with myc antibody was performed as the control of BRAF transfection efficiency. In all cases the Western blots were also hybridized with anti-P-ERK and ERK antibodies. The arrows in the Western blot indicate the high molecular weight bands of NIS protein. d, day.
Citation: Endocrine-Related Cancer Endocr Relat Cancer 13, 1; 10.1677/erc.1.01119
We are grateful to Dr N Carrasco (Albert Einstein, College of Medicine, Bronx, NY, USA) for rat NIS antibody, Dr S Costagliola (IRIBHM, Université Libre de Bruxelles, Bruxelles, Belgium) for monoclonal human NIS antibody, Dr C Goding (Marie Curie Research Institute, UK) for BRAF wt and mutant expression vectors, Dr D Civitarreale (Molecular Pathology Laboratory, Rome, Italy) for pTSHR-LUC, Dr R Di Lauro (Stazione Zoologica A, Dohrn, Naples, Italy) for Tg and pTPO promoters construct, Dr A Fusco (Universita degli Studi, Naples, Italy) for normal and K-RAS-transformed PCCl3 thyroid cells and Dr C Cavada (Department of Anatomy, Histology and Neuroscience, Universidad Autónoma de Madrid, Spain) for her help in obtaining the microphotographs. We are also grateful to C Moya, E Costamagna, M A Zaballos, B García, P Recacha, I Cuesta, T Brandao (Instituto de Investigaciones Biomédicas Alberto Sols, Madrid, Spain), P Cejas, J A Fresno and R Madero (Hospital Universitario La Paz, Madrid, Spain) for their advice. We are also grateful to M González-Monge for her technical assistance.
Funding
G R-E is a recipient of a grant for M.D. training from FIS, Instituto de Salud Carlos III. This work was supported by grants from Dirección General de Investigación: BFBU2004-03169 (M E C); CAM GR/SAL/0773/2004; and FIS of ISCIII RCMN (C03/08), RCGC (C03/10) and PI041216. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
References
American Joint Committee on Cancer 2002 Thyroid. In AJCC Cancer Staging Handbook, pp 89–98. edn 6. Editors: FL Greene, CM Balch, ID Fleming & F April. New York: AJCC.
Aza-Blanc P, Di Lauro R & Santisteban P 1993 Identification of a cis-regulatory element and a thyroid-specific nuclear factor mediating the hormonal regulation of rat thyroid peroxidase promoter activity. Molecular Endocrinology 7 1297–1306.
Castro MR, Bergert ER, Goellner JR, Hay ID & Morris JC 2001 Immunohistochemical analysis of sodium iodide symporter expression in metastatic differentiated thyroid cancer: correlation with radioiodine uptake. Journal of Clinical Endocrinology and Metabolism 86 5627–5632.
Civitareale D, Castelli MP, Falasca P & Saiardi A 1993 Thyroid transcription factor 1 activates the promoter of the thyrotropin receptor gene. Molecular Endocrinology 7 1589–1595.
Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J, Woffendin H, Garnett MJ, Bottomley W et al.2002 Mutations of the BRAF gene in human cancer. Nature 417 949–954.
Dohan O & Carrasco N 2003 Advances in Na(+)/I(−) symporter (NIS) research in the thyroid and beyond. Molecular and Cellular Endocrinology 213 59–70.
Dohan O, Baloch Z, Banrevi Z, Livolsi V & Carrasco N 2001 Rapid communication: predominant intracellular overexpression of the Na(+)/I(−) symporter (NIS) in a large sampling of thyroid cancer cases. Journal of Clinical Endocrinology and Metabolism 86 2697–2700.
Dohan O, De la Vieja A, Paroder V, Riedel C, Artani M, Reed M, Ginter CS & Carrasco N 2003 The sodium/iodide symporter (NIS): characterization, regulation, and medical significance. Endocrine Reviews 24 48–77.
Fugazzola L, Mannavola D, Cirello V, Vannucchi G, Muzza M, Vicentini L & Beck-Peccoz P 2004 BRAF mutations in an Italian cohort of thyroid cancers. Clinical Endocrinology 61 239–243.
Garcia B & Santisteban P 2002 PI3K is involved in the IGF-I inhibition of TSH-induced sodium/iodide symporter gene expression. Molecular Endocrinology 16 342–352.
Garcia-Jimenez C, Zaballos MA & Santisteban P 2005 DARPP-32 (dopamine and 3′ , 5′-cyclic adenosine monophosphate regulated neuronal phosphoprotein) is essential for the maintenance of thyroid differentiation. Molecular Endocrinology 19 3060–3072.
Garnett MJ & Marais R 2004 Guilty as charged: B-RAF is a human oncogene. Cancer Cell 6 313–319.
Hooft L, Hoekstra OS, Deville W, Lips P, Teule GJ, Boers M & van Tulder MW 2001 Diagnostic accuracy of 18F-fluorodeoxyglucose positron emission tomography in the follow-up of papillary or follicular thyroid cancer. Journal of Clinical Endocrinology and Metabolism 86 3779–3786.
Knauf JA, Kuroda H, Basu S & Fagin JA 2003 RET/PTC-induced dedifferentiation of thyroid cells is mediated through Y1062 signaling through SHC-RAS-MAP kinase. Oncogene 22 4406–4412.
Knauf JA, Ma X, Smith EP, Zhang L, Mitsutake N, Liao XH, Refetoff S, Nikiforov YE & Fagin JA 2005 Targeted expression of BRAFV600E in thyroid cells of transgenic mice results in papillary thyroid cancers that undergo dedifferentiation. Cancer Research 65 4238–4245.
Marais R, Spooner RA, Light Y, Martin J & Springer CJ 1996 Gene-directed enzyme prodrug therapy with a mustard prodrug/carboxypeptidase G2 combination. Cancer Research 56 4735–4742.
Melillo RM, Castellone MD, Guarino V, De Falco V, Cirafici AM, Salvatore G, Caiazzo F, Basolo F, Giannini R, Kruhoffer M et al.2005 The RET/PTC-RAS-BRAF linear signaling cascade mediates the motile and mitogenic phenotype of thyroid cancer cells. Journal of Clinical Investigation 115 1068–1081.
Mitsutake N, Knauf JA, Mitsutake S, Mesa C Jr, Zhang L & Fagin JA 2005 Conditional BRAFV600E expression induces DNA synthesis, apoptosis, dedifferentiation, and chromosomal instability in thyroid PCCL3 cells. Cancer Research 65 2465–2473.
Namba H, Nakashima M, Hayashi T, Hayashida N, Maeda S, Rogounovitch TI, Ohtsuru A, Saenko VA, Kanematsu T & Yamashita S 2003 Clinical implication of hot spot BRAF mutation, V599E, in papillary thyroid cancers. Journal of Clinical Endocrinology and Metabolism 88 4393–4397.
Nikiforov YE, Nikiforova MN, Gnepp DR & Fagin JA 1996 Prevalence of mutations of ras and p53 in benign and malignant thyroid tumors from children exposed to radiation after the Chernobyl nuclear accident. Oncogene 13 687–693.
Nikiforova MN, Kimura ET, Gandhi M, Biddinger PW, Knauf JA, Basolo F, Zhu Z, Giannini R, Salvatore G, Fusco A et al.2003 BRAF mutations in thyroid tumors are restricted to papillary carcinomas and anaplastic or poorly differentiated carcinomas arising from papillary carcinomas. Journal of Clinical Endocrinology and Metabolism 88 5399–5404.
Pohlenz J, Duprez L, Weiss RE, Vassart G, Refetoff S & Costagliola S 2000 Failure of membrane targeting causes the functional defect of two mutant sodium iodide symporters. Journal of Clinical Endocrinology and Metabolism 85 2366–2369.
Puxeddu E, Moretti S, Elisei R, Romei C, Pascucci R, Martinelli M, Marino C, Avenia N, Rossi ED, Fadda G et al.2004 BRAF(V599E) mutation is the leading genetic event in adult sporadic papillary thyroid carcinomas. Journal of Clinical Endocrinology and Metabolism 89 2414–2420.
Quiros RM, Ding HG, Gattuso P, Prinz RA & Xu X 2005 Evidence that one subset of anaplastic thyroid carcinomas are derived from papillary carcinomas due to BRAF and p53 mutations. Cancer 103 2261–2268.
Saito T, Endo T, Kawaguchi A, Ikeda M, Katoh R, Kawaoi A, Muramatsu A & Onaya T 1998 Increased expression of the sodium/iodide symporter in papillary thyroid carcinomas. Journal of Clinical Investigation 101 1296–1300.
Santoro M, Melillo RM, Grieco M, Berlingieri MT, Vecchio G & Fusco A 1993 The TRK and RET tyrosine kinase oncogenes cooperate with ras in the neoplastic transformation of a rat thyroid epithelial cell line. Cell Growth Differentiation 4 77–84.
Schlumberger M & Pacini F 2003a Prognostic factors. In Thyroid Tumors, pp 111–125. Paris: Nucléon.
Schlumberger M & Pacini F 2003b Iodine metabolism and radioiodine. In Thyroid Tumors, pp 85–107. Paris: Nucléon.
Schmitz G, Fuzesi L, Struck J, Siefker U, Hamann A, Sahlmann CO, Hufner M & Meller J 2005 Expression of the sodium iodide symporter in differentiated thyroid cancer: clinical evidence. Nuklearmedizin 44 86–93.
Soares P, Trovisco V, Rocha AS, Lima J, Castro P, Preto A, Maximo V, Botelho T, Seruca R & Sobrinho-Simoes M 2003 BRAF mutations and RET/PTC rearrangements are alternative events in the etiopathogenesis of PTC. Oncogene 22 4578–4580.
Soares P, Trovisco V, Rocha AS, Feijao T, Rebocho AP, Fonseca E, Vieira de Castro I, Cameselle-Teijeiro J, Cardoso-Oliveira M & Sobrinho-Simoes M 2004 BRAF mutations typical of papillary thyroid carcinoma are more frequently detected in undifferentiated than in insular and insular-like poorly differentiated carcinomas. Virchows Archives 444 572–576.
Trovisco V, Soares P, Preto A, de Castro IV, Lima J, Castro P, Maximo V, Botelho T, Moreira S, Meireles AM et al.2005 Type and prevalence of BRAF mutations are closely associated with papillary thyroid carcinoma histotype and patients’ age but not with tumour aggressiveness. Virchows Archives 446 589–595.
Wapnir IL, van de Rijn M, Nowels K, Amenta PS, Walton K, Montgomery K, Greco RS, Dohan O & Carrasco N 2003 Immunohistochemical profile of the sodium/iodide symporter in thyroid, breast, and other carcinomas using high density tissue microarrays and conventional sections. Journal of Clinical Endocrinology and Metabolism 88 1880–1888.
Ward LS, Santarosa PL, Granja F, da Assumpcao LV, Savoldi M & Goldman GH 2003 Low expression of sodium iodide symporter identifies aggressive thyroid tumors. Cancer Letters 200 85–91.
Xu X, Quiros RM, Gattuso P, Ain KB & Prinz RA 2003 High prevalence of BRAF gene mutation in papillary thyroid carcinomas and thyroid tumor cell lines. Cancer Research 63 4561–4567.
