The BRAFV600E mutation analysis and risk stratification in papillary thyroid carcinoma

ABSTRACT Objective: Although the prognostic role of BRAFV600E mutation in papillary thyroid carcinoma (PTC) is controversial, the American Thyroid Association (ATA) includes the mutational status in their risk stratification system. To evaluate the impact of the BRAFV600E mutation status on PTC risk stratification. Subjects and methods: PTC patients attending a university-based hospital who had the analysis of the BRAFV600E mutation were included. Persistent disease was defined as the presence of biochemical or structural disease. The performance of the ATA risk stratification system on predicting persistent disease with or without the BRAFV600E mutation status information was evaluated. Results: Of the 134 patients evaluated, 44 (32.8%) carried BRAFV600E mutation. The median tumor size was 1.7 cm (P25-75 1.0-3.0), 64 (47.8%) patients had lymph node, and 11 (8.2%) distant metastases. According to the ATA risk stratification system, patients were classified as low, intermediate, and high risk in 55 (41%), 59 (44%), and 20 (14%) patients, respectively. The data on BRAFV600E mutation reclassified 12 (8.9%) patients from low to intermediate risk. After a median follow-up of 8.5 years, the prevalence of persistent disease was similar in patients with and without BRAFV600E mutation (P = 0.42). Multivariate analysis failed to demonstrate an association between the BRAFV600E mutation and persistent disease status (RR 0.96; 95%CI 0.47-1.94). Notably, none of the patients reclassified from low to intermediate risk showed persistent disease on follow-up. Conclusion: Inclusion of BRAFV600E mutational status has a limited impact on risk stratification and does not add to the prediction of outcomes in PTC patients.

INTRODUCTION D ifferentiated thyroid cancer (DTC) is the most common malignancy in endocrine organs, and its incidence is increasing worldwide (1). The appropriate management of DTC patients, especially those with papillary thyroid carcinoma (PTC), has been changing in recent years and is still a matter of debate. Accurate prognostic factors, which could lead to a more individualized approach, are required and molecular markers have been proposed as potential key elements on the decision-making process (2).
BRAF is one of the most studied genes in PTC pathogenesis. BRAF mutations activate the mitogenactivated protein kinase (MAPK) pathway, resulting in increases in cell proliferation, dedifferentiation, and apoptosis (3). The BRAF point mutation T1799A is the most common, resulting in the exchange of valine (V) to glutamic acid (E) at residue 600 (BRAF V600E ). The first studies describing the role of BRAF V600E mutation on the pathogenesis of PTC were published in 2003 (4).
From the clinical perspective view, the prognostic role of BRAF V600E mutation is still controversial (5).
Several studies were focused on searching for an association between BRAF V600E mutation and features related to unfavorable course of the disease and showed conflicting results (6)(7)(8). Several studies have shown an association of mutated BRAF V600E with lymph node metastases, extrathyroidal extension, tumor size, and advanced disease stage (9). However, the impact of the BRAF V600E mutation on the clinical outcomes (persistent disease, recurrence, survival) is not so clear. In a pivotal multicenter cohort, Xing and cols. showed an association of mutated BRAF V600E with increased PTC disease-specific mortality on univariate analysis but failed to demonstrate independent association in a multivariable model adjusting for patient sex, age at diagnosis, medical center, and several conventional pathologic factors (10). A second study, with a similar design but including a larger number of patients (n = 2099), showed an independent association between BRAF V600E mutation and recurrent disease both in the overall PTC population and after stratification for histological subtypes (classic and follicular variant) (11). Similar results were reached in a recent systematic review (7). Nevertheless, as stated by the authors, the findings should be interpreted with caution due to heterogeneity of the data, which include differences in the patient demographics and ethnicities, as well as in the therapeutic approach and follow-up (7).
Despite these limitations, the 2015 American Thyroid Association (ATA) guidelines had included the BRAF V600E mutation status in the stratification system for persistent/recurrent disease according to the 'continuum of risk' model. In this proposal, those patients classified as low risk-disease with intrathyroidal PTC lesions of 1-4 cm should be reclassified as intermediate risk if positive BRAF V600E mutation is known (12). Here, we aimed to evaluate the impact of the BRAF V600E mutation status on risk stratification as well as on disease outcomes in a cohort of patients with PTC in a tertiary, university-based referral center.

Patients and study design
The patients included were followed in a cohort of DTC patients from the Thyroid Outpatient Clinic of the Thyroid Unit, of Hospital de Clínicas de Porto Alegre (HCPA), a tertiary care, university teaching hospital in southern Brazil. From 2009 to 2015, all consecutive patients with a histological diagnosis of PTC who had paraffin-embedded tumor tissue available for analysis of the BRAF V600E mutation were included in this study.

Treatment protocol and follow-up
Our DTC treatment protocol consists of performing total thyroidectomy (TT), followed or not by administration of radioactive iodine (RAI) as indicated, and the use of suppressive levothyroxine therapy according to current guidelines (12). The same surgical team operated all patients and decisions regarding cervical lymph node dissection were taken at the discretion of the surgical team. The iodine administration protocol used RAI activities prescribed at the attending physician's discretion. RAI was administered in a stimulated thyrotropin (TSH) condition of endogenous hypothyroidism (TSH > 30 mUI/L), after withdrawing levothyroxine (at least 3-4 weeks) (13). A post-therapy whole-body scan (WBS) was performed seven to ten days after the RAI administration. Of note, the information on BRAF V600E mutation status was not used to decide RAI indication in these patients.
In the first evaluation, the following data were recorded for each patient: demographics, tumor characteristics (e.g., histological features, extension, and lymph node involvement) and treatment (e.g., surgery, RAI, and other interventions). Each patient was classified using the 8th edition of the TNM/AJCC (TNM8) staging system (I, II, III, or IV) (14,15). N0 status was determined by clinical examination of the neck or preoperative and postoperative neck ultrasound (US) imaging or macroscopic examination during surgery and pathological examination of patients with lymph node resection.
The follow-up protocol called for an initial assessment at 3 to 6 months post-surgery, which included physical examination of the neck and measurements of serum thyroglobulin (Tg) levels under TSH suppression (Tg-T4) and antithyroglobulin antibody (TgAb). In a second evaluation, 6 to 12 months after the initial treatment, serum Tg was measured under conditions of a stimulated TSH (sTg) with endogenous hypothyroidism (TSH > 30 mIU/L) as indicated by Tg-T4 levels. Neck ultrasound (US) was also performed during the first year of follow-up. Patients classified as excellent response (see below) were scheduled for annual visits that included physical BRAF mutation in papillary thyroid carcinoma Arch Endocrinol Metab. 2020;64/6 examination of the neck and measurements of Tg-T4 and TgAb. Patients with indeterminate or persistent disease were scheduled for medical visits twice a year and evaluated for additional therapy as needed, and based on disease extension, cumulative RAI activity, and previous response to therapy. Additional imaging studies [e.g., neck US, diagnostic I-131 whole-body scan (WBS) and computed tomography (CT)] as indicated when clinical or laboratory findings raised suspicion of persistent or recurrent disease. The duration of follow-up was defined as the time between the TT and the last medical visit to the clinic.

Mutation analysis
BRAF V600E mutation analysis was performed in genomic DNA extracted from 10-µm slides of paraffin-embedded tissue blocks using the Magnesil Genomic Fixed Tissue System (Promega Corporation, Madison, USA) according to the manufacturer's instructions. BRAF exon 15 was amplified by PCR using specific oligonucleotides: 5'-ACCTAAACTCTTCATAATGCTTGCT-3' and 3'-CTGATTTTTGTGAATACTGGGAACT-5'. To PCR amplification 100-300 ng/µL of DNA were used in a reaction mix (25 µL) containing 20 mM Tris-HCl, 50 mM KCl, 2 mM MgCl2, 0.2 mM dNTPs, 0.2mM of each primer, and 1.25 U Platinum Taq DNA Polymerase (Invitrogen, Thermo Scientific, São Paulo, Brazil). For sequencing, PCR products were purified using the GFX PCR DNA purification kit (GE Healthcare, Buckinghamshire, UK) and submitted to direct sequencing using the Big Dye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems, Foster City, CA, USA).

Outcomes
Disease status was defined based on clinical examination, Tg-T4 and/or sTg levels, neck US, post-RAI WBS (when available), and additional imaging exams when indicated. Patients were classified into four categories: excellent response, indeterminate response, biochemical incomplete response, or structural incomplete response.
In patients submitted to RAI, excellent response was defined as negative imaging and Tg-T4 < 0.2 ng/mL or sTg < 1.0 ng/mL. Biochemical incomplete response was defined as negative imaging and Tg-T4 > 1.0 ng/mL or sTg > 10.0 ng/mL or rising TgAb levels. Structural incomplete response was defined as structural or functional evidence of disease with any Tg or TgAb level. Indeterminate response was defined as non-specific findings on imaging studies or Tg-T4 between 0.2 to 1.0 ng/mL or sTg between 1.0 and 10.0 ng/mL or TgAb stable or declining.
In patients not submitted to RAI, excellent response was defined as negative imaging and Tg-T4 < 0.2 ng/mL or sTg < 2.0 ng/mL. Biochemical incomplete response was defined as negative imaging and Tg-T4 > 5.0 ng/mL or sTg > 10.0 ng/mL or increasing Tg levels over time or rising TgAb levels. Structural incomplete response was defined as structural or functional evidence of disease with any Tg or TgAb level. Indeterminate response was defined as nonspecific findings on imaging studies or Tg-T4 between 0.2 to 5.0 ng/mL or sTg between 2.0 and 10.0 ng/mL or TgAb stable or declining.

Statistical analysis
The clinical and laboratory data are reported as the mean ± standard deviation (SD) values or as the median and percentiles 25 and 75 (P25-75) for continuous variables or as absolute numbers and percentages for categorical variables. Comparative analyses of frequencies were performed using Pearson Chi-Square or Fisher's Exact Test, as appropriate. These analyses were performed using the Statistical Package for Social Science Professional software version 20.0 (IBM Corp., Armonk, NY, USA). All tests were two-tailed, and a P < 0.05 was considered statistically significant.
Univariate analysis and generalized linear models with a log link and Poisson errors were used to estimate the impact of clinical variables and B-RAF V600E mutation status as potential prognostic factors for persistent disease.

Impact of BRAF mutation status on clinicopathological features and ATA risk stratification system
The prevalence of BRAF V600E mutation was 32.8% (44/134 patients, 95% CI 25-42). To investigate whether the presence of BRAF V600E mutation would impact on clinicopathological features, the patients were grouped according to their BRAF V600E mutation status. Positive BRAF V600E patients had larger tumors (median size 2.2 vs. 1.6 cm; P = 0.03) and a trend towards more gross extrathyroidal extension of the tumor (20.5 vs. 7.8%; P = 0.06). However, no differences were observed between the groups for sex, age, cervical or distant metastasis, TNM AJCC stage, or ATA risk classification (Table 2).
According to the 2009 ATA risk stratification system, the risk level was low in 55 (41%) patients, intermediate in 59 (44%) patients, and high in 20 (14%) patients. In the group of intermediate risk, the more common determinant to classify patients in this risk category was the presence of lymph nodes metastasis (n = 48, 81.3%), followed by microscopic extrathyroidal invasion (n = 23, 38.9%). In the high-risk group, the main determinant was the presence of distant metastasis (n = 11; 55%), followed by macroscopic invasion (n = 9, 45%). When we included the BRAF V600E mutation status to the ATA risk stratification system, we reclassified 12 (8.9%) patients from low to intermediate risk. All cases consisted of intrathyroidal PTC tumors (size < 4 cm, restricted to the thyroid) ( Table 3).
Notably, none of the patients who had been reclassified from low to intermediate risk due to their mutational status information showed persistent disease on follow-up (Table 3). We were unable to evaluate the impact of BRAF mutation in papillary thyroid carcinoma Arch Endocrinol Metab. 2020;64/6 BRAF V600E mutation on RAI response, due the low number of positive BRAF V600E patients who did not receive RAI dosing (n = 7).
Additional analyses using a multivariate model including disease status as the dependent variable and sex, tumor size, lymph node and distant metastasis, and BRAF V600E mutation as independent variables are shown in Table 4. Tumor size, lymph node and distant metastasis were independent risk factors for persistent disease. Positive BRAF V600E mutation was not associated with persistent disease (RR 0.96; 95% CI 0.47-1.94). BRAF V600E as a prognostic marker is hampered by its low specificity. Indeed, when we look to the performance of the mutation analysis as a diagnostic test for the prediction of recurrent/persistent disease, it has an acceptable sensitivity (60-70%), but poor specificity (34-57%) with a positive predictive positive value of only 7-25% (7,9). Our data showed similar results: sensitivity of 45%, a specificity of 69%, and a positive predictive value of 24%.
We observed a prevalence of positive BRAF V600E mutation of 32.8% in our study. These numbers are comparable to those reported in previous studies carried out in the Southeastern region of Brazil (28.1 to 48.3%) (17,18), but lower than results obtained from the central (63.8%) and Northeast (65.1%) regions (19,20). Indeed, a systematic review of 63 studies (including 20,764 patients) has shown that the prevalence of BRAF V600E mutation varied from 25 to 82.3%. Moreover, when the authors divided the study subjects into subgroups regarding the ethnicities, they observed a decrease in the heterogeneity of the data (7). These results illustrate the heterogeneity of the prevalence of BRAF V600E mutation among the different populations, even in the same genetic background, which could explain (at least in part) the conflicting results of the studies regarding the association of BRAF V600E mutation and outcomes.
Our study has several strengths. First, the fact that all patients included in this analysis were treated and followed at the same institution ensures similar therapeutic and follow-up approaches. Moreover, we analyzed the impact of the BRAF V600E mutation not only on oncologic features but also on the clinical outcomes after a median of 8.5 years of follow-up. On the other hand, we are aware that the relatively small sample size could hamper some analysis, especially those regarding the clinical outcomes; therefore, these data should be interpreted with caution. Additionally, our study could not determine a causal relationship between BRAF V600E mutation and outcomes, and BRAF V600E mutation may correlate with some other mutations or confounders, which may turn out to be even more useful prognostic indicators.
In conclusion, we have demonstrated that the current role of the BRAF V600E mutation on the risk stratification of PTC is limited. The inclusion of BRAF V600E mutational analysis in clinical practice increases the costs, whereas the impact of this strategy on patient risk classification or disease outcome prognostication remains uncertain.

DISCUSSION
The use of specific molecular classifiers as prognostic markers is a research trend in thyroid cancer, and a potential role of BRAF V600E mutation has been long advocated. While the majority of studies demonstrated a relationship between positive BRAF V600E mutation status and poorer clinicopathological features (tumor size, extrathyroidal extension and lymph nodal metastasis), its role as an independent prognostic marker in PTC patients is still a matter of debate (9). Here, we showed that including the information about the BRAF V600E status to the ATA risk stratification system resulted in increasing the risk for some patients (8.9% of our sample population), although without any impact on disease outcomes. Of note, none of the patients who were reclassified from low to intermediate risk based on BRAF V600E status showed persistent disease on followup. The inclusion of the BRAF V600E mutation analysis in the risk classification system, as suggested by ATA [12], increases the complexity of care and costs without a clear benefit. BRAF V600E mutation does not seem to be an independent prognostic marker and should be analyzed in association with other prognostic factors. Even in case of a real association of BRAF V600E mutation with disease recurrence, the clinical application of