Fax: (011) 55-11-5084-5231
High prevalence of BRAF mutation in a Brazilian cohort of patients with sporadic papillary thyroid carcinomas†
Correlation with more aggressive phenotype and decreased expression of iodide-metabolizing genes
Version of Record online: 17 JAN 2009
Copyright © 2009 American Cancer Society
Volume 115, Issue 5, pages 972–980, 1 March 2009
How to Cite
Oler, G. and Cerutti, J. M. (2009), High prevalence of BRAF mutation in a Brazilian cohort of patients with sporadic papillary thyroid carcinomas. Cancer, 115: 972–980. doi: 10.1002/cncr.24118
Dr. Oler is a scholar from São Paulo State Research Support Foundation (FAPESP) and Dr. Cerutti is an investigator of the Brazilian Research Council (CNPq).
- Issue online: 18 FEB 2009
- Version of Record online: 17 JAN 2009
- Manuscript Accepted: 22 SEP 2008
- Manuscript Revised: 26 AUG 2008
- Manuscript Received: 19 JUN 2008
- São Paulo State Research Foundation (FAPESP). Grant Number: 05/60330-8
- papillary thyroid carcinoma;
- clinicopathologic features;
- NIS, TSHR
Although several studies undoubtedly demonstrated that BRAF mutation is an important genetic event in the pathogenesis of papillary thyroid carcinoma (PTC), its prognostic significance and correlation with less differentiated states remains unclear. It has been suggested that the discrepancy may be at least partially due to the insufficient number of cases analyzed, epidemiologic factors, and a combination of different variants of PTC included in these studies.
In this context, the prevalence of the BRAF mutation in a Brazilian cohort of PTCs (n = 120) was first assessed and correlated with clinicopathologic features. The BRAF exon 15 mutation was evaluated by direct sequencing. Furthermore, using quantitative polymerase chain reaction, the issue of whether the expression level of the iodide-metabolizing genes (NIS and TSHR) was correlated with BRAF mutational status was investigated.
A high prevalence of the BRAF mutation was found in PTC cases (48%). The BRAF mutation was found to be significantly associated with the classic variant of PTC (66%; P < .0001), although it was found in the follicular variant as well (21%). Subtype stratification demonstrated that the BRAF V600E mutation was associated with tumor size, extrathyroid invasion, the presence of lymph node metastasis and risk of disease recurrence, and mortality in patients with the classic variant of PTC. Moreover, the expression levels of NIS and TSHR were remarkably lower in PTCs harboring the BRAF V600E mutation.
These findings provide further evidence that BRAF might be associated with a more aggressive phenotype and less differentiated state due to decreased expression of iodide-metabolizing genes. The search for a BRAF mutation in the current study population appears to be valuable for predicting prognosis and guiding management in patients with PTC. Cancer 2009. © 2009 American Cancer Society.
Papillary thyroid carcinoma (PTC) is the most common endocrine malignancy. To our knowledge to date, the main molecular finding is the activating mutations of RAS/BRAF or rearrangement of RET, both of which activate the mitogen-activating protein kinase (MAPK) pathway. These genomic events appear to be mutually exclusive and are the earliest mutations occurring in PTC. The activating point mutation V600E in exon 15 of BRAF gene is the most common and specific genetic event in PTC, with a worldwide prevalence reported to range from 22.9% to 83% in different series.1, 2 Unusual mutations within exon 15 of the BRAF oncogene also have been described as being associated with pathogenesis3 or progression of PTC.4 Although these findings undoubtedly demonstrated that BRAF mutation is an important genetic event in the pathogenesis of PTC, discrepancies exists regarding the presence of the BRAF V600E mutation and a more advanced stage of disease, lymph node/distant metastases at diagnosis, tumor recurrence, and treatment failure in patients with recurrent PTC.3, 5, 6 To identify the factors that may explain treatment failure and tumor recurrence, a possible association between BRAF mutational status and changes in the gene expression of iodide-metabolizing genes has been considered. It was demonstrated that BRAF V600E mutation impairs both NIS expression and trafficking to the membrane.7 Similarly, down-regulation of thyroid markers, particularly a decrease in the expression of TSHR and NIS, was observed in cells expressing BRAF V600E.8 Others have found a decrease of NIS in BRAF V600E‒positive tumors, although TSHR expression was not found to be correlated with mutational status.9 Further investigation on this issue may help to clarify whether BRAF mutation modulates the expression of iodide-metabolizing genes and therefore could predict poor outcome because BRAF-positive tumors may not demonstrate iodine uptake.
In the current series of sporadic PTCs, we sought to investigate the BRAF mutational status in a Brazilian cohort of 120 PTC patients and to correlate this status with clinicopathologic features and NIS and TSHR expression. Although we have previously investigated BRAF mutation in 13 matched primary tumor and lymph node metastases,4 to our knowledge this is the largest Brazilian study reported to date and might help to elucidate whether the BRAF mutation was associated with prognostic factors such as age, sex, tumor size, extrathyroidal extension, multifocality, the presence of metastases, and histologic subtype. In addition, it may help to determine whether the BRAF V600E mutation is correlated with a less differentiated tumor due to changes in the expression of thyroid iodide-metabolizing genes and therefore has important clinical implications.
MATERIALS AND METHODS
A total of 120 PTCs and 7 normal adjacent thyroid tissues were obtained from consecutive patients who underwent thyroid surgery from 2000 through 2007 at Hospital São Paulo, Universidade Federal de São Paulo and Hospital das Clínicas, Universidade Estadual de São Paulo. Tissue specimens were frozen immediately after surgical biopsy and stored at −80°C until use. Final histologic classification was obtained from paraffin-embedded sections that were stained with hematoxylin and eosin. The study was approved by the Ethics and Research Committees from both Universities and was conducted in accordance with the Declaration of Helsinki Principles.
Patient information and clinicopathologic parameters were analyzed retrospectively. Our study group included 73 classic variants of PTC (CVPTC; 21 of which measured <1 cm) and 47 follicular variants of PTC (FVPTC; 11 of which measured <1 cm). PTCs containing less-differentiated components within or adjacent to the tumor or other variants (such as the tall cell variant, columnar cell variant, and the diffuse sclerosing variant) were not included. Cases of tumors associated with lymphocytic infiltration or Hashimoto thyroiditis were not excluded. Of 120 patients, 101 were females and 19 were males. The mean age at the time of diagnosis was 44.7 ± 14.32 years (range, 15 years-76 years). The mean size of the tumor was 2.39 ± 1.96 cm (range, 0.2 cm-9 cm). According to the pathologic TNM (pTNM) staging system,10 79 of 114 (69%) were at stage I, 5 of 114 (4%) were stage II, 22 of 114 (19%) were stage III, and 8 of 114 (7%) were stage IV. Of 120 cases, pTNM was not available for 6 cases. In accordance with this system, the study group was classified into 3 risk categories at the time of initial treatment: very low risk: unifocal T1 (≤1 cm) N0M0 and no extension beyond the thyroid capsule; low risk: T1 (>1 cm) N0M0 or T2N0M0 or multifocal T1N0M0; and high risk: any T3 and T4 or any T, N1 or any M1.11, 12
RNA Isolation, cDNA Synthesis, Reverse Transcriptase-Polymerase Chain Reaction, and Direct Sequencing
BRAF mutation screening was performed by direct sequencing at the RNA level. Total RNA was isolated from the tumor core (in an attempt to avoid the surrounding normal tissue) using Trizol (Invitrogen Corporation, Carlsbad, Calif). RNA isolation and cDNA synthesis were performed as previously reported.13, 14 Briefly, 1 μg of total RNA was treated with DNAse (Ambion, Austin, Tex) and was reverse-transcribed to cDNA using Super-Script III Reverse Transcriptase and oligo(dT)12-18 primer and 10 U of RNase inhibitor (Invitrogen Corporation). Aliquots of 2 μL of cDNA were used in 20-μL amplification reactions as described previously.4 Polymerase chain reaction (PCR) products were resolved by electrophoresis in agarose/ethidium-bromide stained gels, purified, directly sequenced using ABI Prism BigDye Terminator Cycle Sequencing Ready Reaction Kits (version 3.1), and analyzed in an ABI 3100 Genetic Analyzer (Applied Biosystems, Foster City, Calif).15, 16
Quantitative PCR (qPCR) was used to evaluate NIS and TSH-R expression. An aliquot of cDNA synthesized as above was used in 20-μL of PCR reaction containing SYBR Green PCR Master Mix (PE Applied Biosystems, Foster City, Calif) and 200 nM each primer for the target or control gene (RS8; gene ribosomal protein 8).17-19 Primers sequences were as follows: NIS sense: 5′ CAGAACCACTCCCGGATCAA 3′ and antisense: 5′ ACCCACCACAAAAGTCCAGAA 3′; TSH-R sense: 5′ ACATGACGTCAATCCCTGTG 3′ and antisense: 5′ TGAAAGCATATCCTTGGACTG 3′; S8 sense: 5′ AACAAGAAATACCGTGCCC 3′ and antisense: 5′ GTACGAACCAGCTCGTTATTAG 3′. The PCR reaction was performed using a 4-step program: 50°C for 2 minutes, 95°C for 10 minutes, 40 cycles of 95°C for 15 seconds, and annealing/extension temperature 60°C for 1 minute. The qPCR reactions were performed in triplicate; the threshold cycle (Ct) was obtained using Applied Biosystem software and was averaged (SD ≤1). Relative expression levels were calculated according to the formula 2 ((Ce-Re)/2 (Cn-Rn)) in which Ce is the Ct cycle number observed in the experimental sample for the control gene, Re is the Ct cycle number observed in the experimental sample for the target gene, Cn is the average Ct cycle number observed in the normal thyroid tissues for the control gene, and Rn is the average Ct cycle number observed in the normal thyroid tissues for the target gene.17, 20 The results obtained from relative expression levels were log transformed and used to statistical analysis.
The correlation between BRAF V600E, histologic variants, or clinicopathologic features was determined using the chi-square test. The association between BRAF mutational status and gene expression was performed using a 1-sided Student t test for unpaired data. Data analysis was performed using StatView 4.5 software (Abacus Concepts Inc., Berkeley, Calif). A P value <.05 was considered statistically significant.
Prevalence of BRAF Mutation in Primary PTC
Patients were separated into 2 groups according to the presence (BRAF-mutated) or absence (BRAF wild–type) of the BRAF exon 15 mutation. Of 120 patients, 58 (48%) were BRAF-mutated and 62 (52%) were BRAF wild–type. The most common mutation was BRAF V600E, which was found in 57 of 58 cases (98%). The BRAF V600E + K601del mutation, which was described in our original study in 3 lymph node metastases but was found to be absent in the primary tumor,4 was found in 1 case of a metastatic FVPTC (Table 1).
|BRAF Exon 15 Mutation|
|PTC (n=73)||FVPTC (n=47)||Total (n=120)|
|No. (%)||P||No. (%)||P||No. (%)||P|
|≥45||20/31 (65%)||>.05||6/23 (26%)||>.05||26/54 (48%)||>.05|
|<45||27/40 (68%)||4/20 (20%)||31/60 (52%)|
|Female||41/62 (66%)||>.05||10/39 (26%)||>.05||51/101 (50%)||>.05|
|Male||7/11 (64%)||0/8 (0)||7/19 (37%)|
|Tumor size, cm†|
|≥1||38/50 (76%)||.0023||7/35 (20%)||>.05||45/85 (53%)||>.05|
|<1||8/21 (38%)||3/11 (27%)||11/32 (34%)|
|≥2||23/29 (79%)||.0333||5/28 (18%)||>.05||28/57 (49%)||>.05|
|<2||23/42 (55%)||5/18 (28%)||28/60 (47%)|
|Yes||25/33 (76%)||.0473||1/7 (14%)||>.05||26/40 (65%)||.0056|
|No||19/36 (53%)||9/38 (24%)||28/74 (38%)|
|Lymph node metastases|
|Yes||33/41 (80%)||.0027||2/10 (20%)||>.05||35/51 (69%)||.0001|
|No||15/32 (47%)||8/37 (22%)||23/69 (33%)|
|Yes||24/41 (59%)||>.05||4/19 (21%)||>.05||28/60 (47%)||>.05|
|No||23/31 (74%)||6/26 (23%)||29/57 (51%)|
|High||43/55 (78%)||<.0001||3/20 (15%)||>.05||46/75 (61%)||.0005|
|Low||5/18 (28%)||7/25 (28%)||12/43 (28%)|
|Total of BRAF mutated cases||48/73 (66%)||<.0001¶||10/47 (21%)#||58/120 (48%)|
BRAF Mutational Status and Clinicopathologic Correlation
We correlated the BRAF mutational status with clinicopathologic parameters such as subtype stratification, age, sex, tumor size, multifocality, extrathyroidal invasion, the presence or absence of lymph node metastases, and risk stratification in accordance with the pTNM system. No significant correlation was found between BRAF mutation with age (mean age, 44.44 ± 13.81 years [BRAF V600E] vs 44.95 ± 14.83 years BRAF wild–type), sex, and multifocality. Conversely, an association was found between BRAF mutation and extrathyroidal invasion (P = .0056), the presence of lymph node metastasis (P = .0001), risk for disease recurrence, and mortality (P = .0005) (Table 1).
BRAF Mutational Status and Clinicopathologic Correlation by Histologic Subtype
Histologic subtype stratification revealed that the BRAF mutation was significantly associated with CVPTC (66%) versus FVPTC (21%) (P < .0001) (Table 1). In CVPTC, extrathyroidal invasion (P = .0473), the presence of lymph node metastases (P = .0027), risk for tumor recurrence, and mortality (P = <0.0001) were found to be significantly correlated with the BRAF mutation. In addition, PTCs that were positive for BRAF had greater tumor size (mean, 2.3 ± 1.6 cm) compared with BRAF wild–type (mean, 1.56 ± 1.7 cm) (P = .0367). When patients were subdivided into 2 groups according tumor size (≥2 cm and <2cm; P = .0333) or (≥1 cm and <1 cm; P = .023), a statistically significant association was found (Table 1).
Although the BRAF mutation was less prevalent in thyroid papillary microcarcinoma than in tumors measuring ≥1 cm, 5 of 7 of metastatic papillary microcarcinomas were found to be positive for the BRAF V600E mutation versus 3 of 14 nonmetastatic cases. Even though these results have borderline statistical significance (P = .0555), they demonstrate a trend in the association between the BRAF mutation and lymph node metastasis that is independent of tumor size.
It is interesting to note that no significant association was found between the BRAF mutation and clinicopathologic features in FVPTC (Table 1).
Correlation Between the BRAF Mutational Status and Expression of Iiodide-metabolizing Genes NIS and TSHR
To test whether NIS and TSHR are down-regulated in the BRAF-mutated group compared with the BRAF wild–type group, qPCR was performed in the aforementioned 120 PTC cases. The relative expression value was log transformed before statistical analysis.
Although reduced in the majority of cases of CVPTCs compared with normal thyroid (data not shown), NIS expression was found to be markedly lower in the BRAF-mutated group compared with the BRAF wild–type group (P < .0001) (Fig. 1A). Similar results were obtained when TSHR mRNA expression was evaluated (P = .0019) (Fig. 1B). In FVPTCs, NIS expression was found to be significantly lower in the BRAF-mutated group (P = .0325) (Fig. 1C). Although not considered significant, TSHR expression was found to be lower in the BRAF-mutated group (P = .2401) (Fig. 1D).
When NIS and TSHR expression were analyzed in tumors measuring <1 cm, NIS expression was found to be lower in classic papillary microcarcinoma harboring the BRAF mutation (P = .0153). TSHR was not found to be differentially expressed (Figs. 2A and B).
Although PTCs are usually curable with standard surgical and adjuvant radioiodine treatment, lymph node metastases are found in 30% to 65% of cases at the time of initial diagnosis and 15% of tumors with lymph node metastases also display very aggressive behavior, characterized by local invasion, distant metastasis, treatment resistance, and increased mortality.21 In view of the finding that the incidence of this malignancy has been increasing over the last decade, it is of particular importance to identify those patients with advanced disease and those who fail to respond to radiodine ablation therapy.
With these factors in mind, several biologic markers have been investigated for their value in predicting prognosis. Among all candidate markers, BRAF mutation has been determined to be 1 of the most effective markers in predicting clinical outcome in CVPTC, although there is controversy. It has been suggested that such discrepancies may be due to the inclusion of different histologic subtypes of thyroid cancer, multifocality, epidemiologic factors, and small study cohorts.1, 3, 22 In fact, many of the cohorts studied are small, tumors were not stratified by histologic subtype, and the prevalence has shown a marked geographic discordance.
In our previous study, which had the purpose of examining any inter-relation between the BRAF mutation and presence of lymph node metastasis, 13 metastatic tumors and their corresponding lymph node metastases were investigated.4 Hence, this study was designed to determine the prevalence of the BRAF mutation in a Brazilian cohort and correlate it with clinicopathologic features and subtype stratification.
A high prevalence of the BRAF mutation (48%) was found in PTCs. When stratified by histologic subtype, the prevalence was found to be significantly higher in the CVPTC (66%) compared with FVPTC (21%). The current study cohort were within the range of that previously reported for CVPTC (38%-69%) and FVPTC (0%-32%).2, 3, 23, 24 To our knowledge, the current study not only represents 1 of the largest cohort of CVPTC patients reported in the literature to date, but also reports a prevalence that is comparable to the highest rates detected.2, 3, 23 Lastly, although some authors have suggested that geographic factors may explain the differences observed in the literature, data from the current study support the view that the more likely reason is a lack of histologic stratification.
The results of the current study indicate that, when CVPTC and FVPTC are grouped together, the presence of the BRAF mutation was associated with extrathyroid invasion, the presence of lymph node metastasis and a high risk for tumor recurrence, and mortality. Although others have found that the BRAF mutation was significantly associated with age,25 male sex,26 tumor size,26 and multifocality,3, 25 we did not find any associations.
When we examined patients with CVPTC separately, the BRAF mutation was found to be correlated with tumor size (in addition to lymph node metastases, extrathyroidal invasion, risk for tumor recurrence, and mortality). Although the majority of the studies did not find any correlation between the BRAF mutation and tumor size,2, 27 large tumor size is reported to be 1 of main determinants of poor outcome in patients with PTC.28 Although BRAF mutation was also found in tumors measuring <1 cm, the majoritiy of BRAF-mutated papillary microcarcinomas of the classic variant were metastatic.
It is important to note that no correlation was found between the BRAF mutation and clinical pathologic features in patients with FVPTC. One could argue that the results of the current study may be partially due to the finding that the BRAF mutation was infrequently found in FVPTC and that a small number of positive cases had lymph node metastasis, extrathyroidal invasion and a high risk of recurrence, and mortality and, therefore, a larger series may be needed. However, the majority of studies demonstrated that BRAF mutation is more prevalent in the classic or tall-cell variant of PTC and are rarely found in the follicular variant.1, 2 These findings suggest that a different genetic event might be associated with tumor progression in the FVPTCs and therefore a better understanding of the biology underlying this “subphenotype” is required.
Overall, data from the current study are in agreement with those reported in other series, which demonstrated that the BRAF V600E mutation plays an important role in the tumorigenesis of CVPTC. Although the prognostic value of the BRAF V600E mutation remains controversial, the current study data demonstrated that BRAF confers a more aggressive phenotype and a higher risk for tumor recurrence and mortality in CVPTC.
Beyond the likely impact of the presence of the BRAF mutation in the final preoperative decision in terms of the extent of surgery to be performed and which patients need to be monitored aggressively for disease recurrence, a key question now is whether the BRAF mutation will have an impact on therapeutic treatment.
Recently, it was reported that tumors with the BRAF V600E mutation have a high recurrence rate and that the majority of these recurrences had no avidity for radioiodine (131I) and therefore are predictive of a poor outcome because 131I treatment is not effective.7 The author also demonstrated that the high risk of disease recurrence was associated with less differentiated tumors due to the loss of NIS-mediated 131I uptake.7 When expressed in tumors, NIS was mainly cytoplasmic and not target to the membrane. In addition, it was demonstrated in vitro that BRAF decreased the promoter activity of NIS, TSHR, and TPO.7
Concordantly, Liu et al demonstrated in vitro that the expression of NIS and TSHR were most susceptible to the influence of the BRAF V600E mutation as well as most readily restorable in the rat thyroid cell line PCCL3 by the suppression of the BRAF/MEK/MAP kinase pathway either by siRNA targeting the sequence of BRAF or U0126 treatment (a specific MAPK/ERK kinase inhibitor).8 In addition, TSHR and NIS expression were found to be decreased after BRAF V600E expression in the PCCL3 cell line.29
Although they represent useful models and will help clarify the therapeutic implications of BRAF, the majority of data published to date were generated in vitro. When NIS and TSHR expression was investigated in BRAF-mutated versus BRAF wild-type in thyroid tumors, the results were controversial. Riesco-Eizaguirre et al7 and Durante et al9 found an association between BRAF mutation and loss of NIS expression. In contrast, Mian et al found no BRAF-related differences in NIS expression.30 No association was demonstrated between BRAF and TSHR expression in patients with thyroid tumors.9
To help to elucidate this significant clinical issue, we investigated the level of expression of NIS and TSHR in our cohort of PTCs. We observed a correlation between the presence of the BRAF V600E mutation and a decrease in NIS expression in both CVPTC and FVPTC. Remarkably, a decreased in NIS expression was also observed in papillary microcarcinoma of the classic variant harboring the BRAF mutation. Although TSHR was down-regulated in both the classic and follicular variants of PTC with a BRAF mutation, a significant association was found in patients with CVPTCs. It is interesting to note that NIS expression was found to be lower than TSHR expression in patients with BRAF mutations.
One possible explanation for our findings comes from in vitro analysis. It was demonstrated that NIS expression was markedly decreased after the expression of the BRAF V600E mutation, whereas TSHR decreased more moderately.7 In addition, it has been demonstrated in vitro that NIS expression is lost first followed by Tg expression, whereas TSHR expression is lost only in dedifferentiated thyroid cell lines.31
The association between the BRAF mutation and a decrease in NIS expression in papillary microcarcinoma described in the current series may be explained by the finding that the BRAF mutation is likely an early event in the tumorigenesis process. Although most likely insufficient alone for the fully aggressive phenotype, it may predispose the tumor cells to the acquisition of additional genetic alterations, which in turn activates more aggressive pathways and leads to dedifferentiation.29 With regard to TSHR expression, no difference was observed in papillary microcarcinoma with or without the BRAF mutation. As demonstrated in vitro,31 in thyroid tumors, NIS expression is more likely lost first followed by a decrease in other iodine-metabolizing genes.
Because the success of the radioiodine 131I treatment requires both adequate stimulation by thyroid–stimulating hormone and the ability of tumoral cells to incorporate 131I, the expression of iodine-metabolizing genes are required. On the basis of the finding that the expression of TSHR and NIS was restored through the suppression of the BRAF/MEK/MAP kinase pathway, preoperative assessment of BRAF status may improve patient management because more extensive surgery could be preformed and kinase inhibitors could be used in conjunction with radioiodine ablation.
Conflict of Interest Disclosures
Supported by the São Paulo State Research Foundation (FAPESP) from Grants 05/60330-8 to JMC.
- 9BRAF mutations in papillary thyroid carcinomas inhibit genes involved in iodine metabolism. J Clin Endocrinol Metab. 2007; 927: 2840-2843., , , et al.
- 18A preoperative diagnostic test that distinguishes benign from malignant thyroid carcinoma based on gene expression. J Clin Invest. 2004; 1138: 1234-1242., , , et al.
- 33The heterogeneous distribution of BRAF mutation supports the independent clonal origin of distinct tumor foci in multifocal papillary thyroid carcinoma. J Clin Endocrinol Metab. 2007; 929: 3511-3516., , , et al.