Age and BRAF^V600E Mutation Stratified Patients with Cytologically Benign Thyroid Nodules

Introduction

The thyroid gland is one of the largest endocrine glands in the human body. Thyroid lesions are often found in the gland, with a prevalence of 4% to 7%. Most of them are asymptomatic, and thyroid hormone secretion remains normal.¹ Bernet et al have reported that thyroid nodules are associated with a 7–15% risk of malignancy.² Ultrasound (US) is the most sensitive model to assess thyroid nodule morphology, aiding in identifying high-risk cases for malignancy.³ Certain sonographic features like hypoechogenicity, microcalcifications, irregular margins, absent halo sign, and increased intramodular blood flow are associated with a higher risk of malignancy.¹ One of the newest radiology systems, the Chinese Thyroid Imaging Reporting and Data System (C-TIRADS), established by Chinese experts using a counting method, is suitable for Chinese clinical practices, and easy for clinical application and promotion.⁴ Nodules with higher C-TIRADS score are required to perform cytological analysis through fine-needle aspiration (FNA) to determine the risk of malignancy.⁵

Ultrasound-guided FNA (US-FNA) is a reliable and effective approach for evaluating thyroid nodules before surgery, providing cytopathological assessment with high sensitivity and specificity for malignancy.⁶ Otherwise, the false-negative rate of US-FNA diagnosis is approximately 12%. Based on the literature, approximately 70% of FNA are categorized as Bethesda II, with an anticipated 3% being false negatives.⁷ It is necessary to provide definitive diagnoses to distinguish malignant ones from benign cytological results for better management of patients.

Molecular analysis can be used to enhance the diagnostic accuracy of cytological diagnosis and as a marker for targeted therapy.⁸ The main known genetic causes of thyroid cancer include point mutations in the BRAF, RAS, TERT, RET, and TP53 genes and the fusion genes RET/PTC, PAX8/PPAR-γ, and NTRK.⁹ Fusion gene positive PTCs were associated with more aggressive behavior disease than fusion gene negative PTCs.¹⁰ A comprehensive study reported by Pekova et al recently identified that RET fusion was associated with a 100% probability of malignancy.¹¹ BRAF^V600E is the most common genetic mutation in thyroid cancer, with a prevalence ranging from 37% to 83% in papillary thyroid cancer (PTC),¹² and has shown a strong association with lymph node metastasis, extra thyroidal extension, disease recurrence, and lymph node metastasis-associated mortality.^13,14 Several studies have proved BRAF^V600E mutation is a valuable and accurate adjunctive diagnostic marker on US-FNA to screen malignancy in non-diagnostic and indeterminate thyroid nodules.^15,16 Few studies have focused on the influence of BRAF^V600E mutation in benign cytological analysis on malignancy, and the clinical characteristics associated with BRAF^V600E mutation in the Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) II remained unclear.^17–19

In this study, we evaluated the diagnostic efficacy of BRAF^V600E mutation in thyroid nodules for screening out malignancy, and the association with progression of thyroid cancer. On the other hand, in order to avoid overdiagnosis and overtreatment, we further focused on clinical characteristics and US features associated with positive BRAF^V600E mutation in patients in low-risk cytology and US diagnosis to accurately locate the risk of BRAF^V600E mutation and malignant risk.

Methods

Patients and Data

A total of 223 thyroid nodules in 210 patients underwent US and US-FNA with an additional BRAF^V600E mutation detection in the Department of Thyroid and Breast Surgery of the First Hospital of Putian City (FHPT patients cohort) in Fujian Province from January 2021 to May 2022. Six patients with 6 nodules were excluded because of an incomplete cytopathological diagnosis or insufficient sonography diagnosis. As a result, 204 patients with 217 nodules were included in the analysis. The patients included 152 females and 52 males, aged from 17 to 91 years, mean age was 47.25±12.16 years. All the patients experienced clinical follow-up after six months, and no progression occurred during this period. Among all the samples, 44 thyroid nodules were treated with thyroidectomy, and 43 thyroid nodules had been diagnosed pathologically by postoperative paraffin sections, of which 38 were PTC and 1 was medullary thyroid cancer (MTC), three cases with nodular goiter, and 1 with thyroiditis. Patients characteristics including BRAF^V600E mutation, the classification of US and FNA, sonography diagnosis information, gender, age, drinking, smoking, location, multiplicity, size, and pathological diagnosis were also recorded, the detail information was shown in Supplementary Tables 1 and 2. Both clinical data and BRAF^V600E mutation detection results of 507 patients with thyroid cancer from the Cancer Genome Atlas (TCGA) were obtained from UCSC Xena (https://xenabrowser.net/datapages/), 501 patients among them with the RNA-seq expression data and 491 patients with mutation data. The diagram of the study group is shown in Figure 1.

Figure 1 The diagram of the study. The sample used for analysis and the analysis diagram of the study.

Ultrasound Examination

US examinations were performed by an experienced US radiologist (GH) used Siemens S2000 ultrasound diagnostic instrument equipped with a 5–12–MHz linear-array transducer. The thyroid glands were scanned in the supine position to determine the location, size (at their longest diameter), number, internal echo, morphology, margin, shape, and calcification. Suspicious US features of malignancy include microcalcification, irregular margins, taller-than-wide shape, hypoechogenicity and marked hypoechogenicity, irregular thick hole, and evidence of extrathyroidal growth.²⁰ Two radiologists with more than ten years’ experience in thyroid ultrasonography classified all nodules into four category according to C-TIRADS: 3 (probably benign), 4A (low suspicion malignant), 4B (moderate suspicion malignant) and 4C (high suspicion malignant).⁴

FNA of Thyroid Nodules

FNA was performed under the guideline of US by experienced radiologists with a 0.7 mm syringe. Each lesion was aspired three or four passes in different directions to complete the sampling. Two or three of them were smeared on glass slides and fixed for cytopathological analysis. Based on TBSRTC, the results of FNA were categorized into six classifications, I–VI were non-diagnostic, benign, atypia undetermined significance/follicular lesion of undetermined significance (AUS/FLUS), suspicious for follicular neoplasm or suspicious for a Hurthle cell neoplasm, suspicious for malignancy and malignancy,²¹ respectively.

BRAF^V600E Mutation Test

The specimens for BRAF^V600E mutation analysis were from Formalin-Fixed Paraffin-Embedded (FFPE) tissues, thyroid FNA samples or serum samples. Genomic DNA extraction from FFPE tissues and thyroid FNA samples were used TargetingOne^® FFPE TNA-P1, TargetingOne^® FFPE TNA-P2 (TargetingOne^® Biotech. Beijing, China), respectively. CfDNA from serum samples were used TargetingOne^® cfDNA-B1 (TargetingOne^® Biotech. Co. Ltd. Beijing, China). The quantity of isolated DNA was assessed by Nanodrop, all samples were adequate for BRAF^V600E detection.

The BRAF^V600E mutation was detected by human BRAF^V600E mutation detection kit (TargetingOne^® Biotech. Beijing, China) according to the manufacturer's instructions. The reaction components were prepared as follows: 7.5 μL of PCR reagent A (BRAF^V600E), 7.5 μL of PCR reagent B (BRAF^V600E), and 20–50 ng of DNA, sterilized water was used to make a volume of up to 30 μL. The optimal annealing temperature is determined to be 60°C. Each experiment included a negative and positive (BRAF^V600E) control. The ddPCR platform system (TargetingOne^® Biotech. Beijing, China) was utilized for evaluation. Droplets were generated and analyzed using the Drop Maker M1 and the Chip Reader R1. The quantitative results were estimated by modeling as a Poisson distribution using chip Analyzer-V1 (TargetingOne^® Biotech. Beijing, China).

Different Expression Genes (DEGs) Identified and Immune Response Analysis

The R package “limma” was applied to calculate differentially expressed genes (DEGs) between older and younger and positive and negative BRAF^V600E mutation, the threshold was set to p < 0.05 and | logFC | > 1.²² Then, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathway analysis were performed on the DEGs within older and younger groups, and the intersection DEGs of positive and negative BRAF^V600E mutation groups and younger and older age groups, respectively. The R package “clusterProfiler” (v.4.7.1.2) was used for functional annotation of the DEGs. An immune-related gene set (immune inhibitors and immunostimulators) was collected from the online tool TISIDB.²³ The results were exhibited by bubble diagram and box plot drawn by “ggplot2” (v.3.4.1) package.

Somatic Mutation Analysis

We performed somatic mutation analysis based on thyroid carcinoma (THCA) samples. To reduce the false-positive rates, we deleted synonymous mutations. Mutational spectrum and mutational signature were depicted via “maftools” (v.2.12.0) R packages which provides a multiple of analysis modules to perform the visualization process. The mutation portrait matrix was factorized into two nonnegative matrices, “signatures” and “contributions” by the ExtractSignatures function of “maftools”, where “signatures” represented mutational processes and “contributions” reflected the associated mutational activities. This approach can automatically estimate the ideal number of retrieved mutational signatures because it is based on Bayesian variant nonnegative matrix factorization.

Statistical Analysis

The statistical analysis was performed with SPSS version 26.0 and R version 4.2.1. Fisher’s exact test or Pearson’s chi-squared test was used for categorical variables analysis. Normally distributed data were expressed as mean ± standard deviation and analyzed with a two-sample t-test, and non-normally distributed variables were presented as medians and interquartile range (IQR) using the Wilcoxon-Mann-Whitney test. The “ggalluvial” (v.0.12.3) R package was used to produce the alluvial diagrams. To measure the specificity and sensitivity of the prognostic capability of C-TIRADS, BRAF^V600E, TBSRTC category, and the combined methods, we calculated the area under the curve (AUC) using the R “pROC” (v.1.18.0) package. A p-value of <0.05 was considered statistical difference and p-value of <0.01 was considered statistically significant difference.

Results

BRAF^V600E Mutation Was Associated with Malignancy of Thyroid Nodules and Progression of Thyroid Cancer

To evaluate the effect of BRAF^V600E mutation on malignant progression of thyroid nodules, the BRAF^V600E mutation detection was performed in all 204 patients with 217 thyroid nodules, the positive BRAF^V600E mutation was detected in 89 (41.0%) thyroid nodules, and the other 128 nodules (59.0%) were negative. In TBSRTC II, the incidence of negative BRAF^V600E mutation significantly higher than that of positive BRAF^V600E mutation. Conversely, the positive BRAF^V600E mutation was statistically significantly associated with TBSRTC III–IV (p=0.007) and TBSRTC V–VI (p<0.001), which poses a higher risk of malignancy. Patients in C-TIRADS 4b-4c had more frequent positive BRAF^V600E mutation than patients with low-risk US classification (p<0.001) (Table 1). Notably, a substantial majority of thyroid nodules exhibiting positive BRAF^V600E mutation were found to be situated in high-risk FNA and US categories (Figure 2A). We further investigated the correlation of BRAF^V600E mutation with different pathological stages of thyroid cancers based on TCGA database. The BRAF^V600E mutation was increased in advanced stages (stage III–IV) compared with early stages (stage I–II) (p<0.001) (Figure 2B), similarly, the proportion of positive BRAF^V600E mutation was higher in advanced pathological T stage (T3-T4) (p<0.001) and N1 stage (p<0.001) (Figure 2C and D). With the development of M stage, there was no statistically significant difference between positive and negative BRAF^V600E mutation (Figure 2E). When separated from Asian patients, the BRAF^V600E mutation increased as the pathological stage and T stage increased (Supplementary Figure 1A and B), but the results were not statistically significant. No differences were found in M stage and N stage (Supplementary Figure 1C and D). The expression of BRAF has no statistically significant difference between positive and negative BRAF^V600E mutation (Figure 2F and Supplementary Figure 1E), indicated that the BRAF^V600E mutation has no influence in expression of BRAF, associated with malignant in thyroid nodules directly. We further analyzed the molecular features of BRAF^V600E mutation in thyroid cancer using TCGA-THCA cohort, we found that BRAF^V600E was the most commonly mutated gene. Interestingly, most patients with positive BRAF^V600E mutation did not have any other mutations in the top 30 most frequently mutated genes (Supplementary Figure 2A). And the coincident and exclusive associations across mutated genes showed that the top 30 genes with the highest mutation frequency were mutual exclusivity with positive BRAF^V600E mutation (Supplementary Figure 2B). The results indicate that positive BRAF^V600E mutation is associated with higher risk of malignancy in thyroid nodules.

Table 1 Correlation Between Clinical Characteristics and BRAF600E Mutation in All Thyroid Nodules

Figure 2 Associations of BRAFV600E mutation with thyroid nodule classifications and clinicopathological stages. (A) Alluvial diagram of the BRAFV600E mutation with different classification of US classification C-TIRADS, TBSRTC categories, and Pathological diagnosis; (B–E) Correlation between BRAFV600E mutation status and (B) tumor stages, (C) T stages, (D) N stages, (E) M stages in TCGA-THCA cohort, red: positive BRAFV600E mutation, blue: negative BRAFV600E mutation; (F) The correlation between BRAFV600E mutation status and the expression of BRAF in TCGA-THCA cohort.

BRAF^V600E Was an Adjunctive Diagnostic Marker on FNA to Screen Malignancy in Benign Cytology

Among all the thyroid nodules, 43 of them have postoperative pathological diagnosis, of which 4 cases with nodular goiter or thyroiditis were benign, and the other 39 were malignant, including 38 PTC and 1 MTC. As shown in Table 2, the percentage of positive BRAF^V600E mutation is 74.4% (29/39) in PTC cases in our study, which was significantly associated malignant of thyroid nodules (p=0.008). All the thyroid nodules belonged to TBSRTC V–VI were confirmed malignant, however, the results were not statistically significant (p=0.072). Seven (17.9%) samples classified as TBSRTC II and 18 (47.2%) samples classified as C-TIRADS 3–4a confirmed to be malignant by pathological diagnosis, indicating that the sensitivity for malignancy diagnosis of the TBSRTC and C-TIRADS classification was low in our study. No statistically significant differences were observed in gender, age, drinking, and smoking between the benign and malignant thyroid nodules. Noteworthily, the median age in malignancy is 45.92±12.55, which was younger than those in benign, but there was no statistical significance between them (p=0.664).

Table 2 Clinical Characteristics Associated with Malignant in Thyroid Nodules

We analyzed the sensitivity and specificity of TBSRTC, BRAF^V600E mutation, and C-TIRADS along with the combination of both or three methods. The results revealed that the specificity of TBSRTC was as high as 100%; however, the sensitivity was lower compared with BRAF^V600E mutation and C-TIRADS classification. The combination of TBSRTC and BRAF^V600E mutation greatly increase the sensitivity of TBSRTC, and showed the highest AUC, which was 0.923 (Figure 3 and Table 3). All samples with positive BRAF^V600E mutation were malignant by pathological diagnosis, including benign cytology diagnosis TBSRTC II (Table 4). Several studies had assessed the risk of malignancy in patients with benign cytology,^17–19 the results revealed that the majority thyroid nodules with positive BRAF^V600E mutation confirmed malignant (Table 5). The results above indicated that there was a risk of malignant in benign cytological results, which may be associated with positive BRAF^V600E mutation.

Table 3 Diagnostic Efficacy of Different Methods for Discrimination Malignant in Thyroid Nodules

Table 4 Correlations of BSRTC Categories with BRAFV600E Mutation and Pathological Results

Table 5 The Prediction of BRAFV600E Mutation on Malignancy in Thyroid Nodules with Benign Cytology

Figure 3 ROC analysis of diagnosis methods for malignancy. The ROC analysis of TBSRTC, BRAFV600E mutation status, and C-TIRADS for malignancy in thyroid nodules.

The Prediction of Clinicopathologic and US Characteristics to BRAF^V600E Mutation in Patients with Low-Risk Cytological and US Classification

Based on the results described above, the positive BRAF^V600E mutation was a risk factor of malignancy in patients with benign cytology. However, according to the 2015 American Thyroid Association (ATA) guideline, the BRAF^V600E mutation detection was un-practical and not required immediately in benign cytology,²⁴ and the clinicopathologic and US characteristics associated with BRAF^V600E mutation in patients with benign cytology were unclear. In this study, all the cases classified TBSRTC II had BRAF^V600E mutation detection, 18 of those were positive. The mean age of patients with positive BRAF^V600E mutation was 56.35±11.89 years, significantly higher than that of negative BRAF^V600E mutation, which was 47.67±9.38 years (p=0.0012), other clinicopathologic and US features had no statistically significant differences between positive and negative BRAF^V600E mutation. Meanwhile, higher US classification C-TIRADS 4b-4c was not significantly correlated with positive BRAF^V600E mutation (Table 6). Considering the majority samples with benign cytology belonged to C-TIRADS 3–4a, we further analyzed the clinical and US features associated with positive BRAF^V600E mutation in patients with C-TIRADS 3–4a in TBSRTC II results. Consisted with the result of Table 6, positive BRAF^V600E mutation was upregulated in older patients (p=0.001), multiple nodules and right location also showed difference between positive and negative BRAF^V600E mutation, with p value were 0.031 and 0.036, respectively (Table 7). Other suspicious malignant US features, such as hypoechogenicity, irregular morphology, unclear boundary and larger thyroid nodules size have no significant association with BRAF^V600E mutation.

Table 6 Correlation Between BRAFV600E Mutation and Clinical Characteristics in Thyroid Nodules with Benign Cytology

Table 7 The Relationship Between BRAFV600E Mutation and Clinical Characteristics in C-TIRADS 3–4a Patients with Benign Cytology

BRAF^V600E Mutation Was Upregulated as Age Increased in Patients with Low Risk FNA and US Diagnosis

Patients of older age exhibit a higher proportion of BRAF^V600E mutations in benign cytology TBSRTC II (p=0.0012) and in C-TIRADS 3–4a patients of benign cytology (p=0.001) (Figure 4A and B), but no significant difference was found in the entire population in our cohort (Figure 4C). We then conducted a statistical analysis using the TCGA database on thyroid cancer, there was no correlation between age and BRAF^V600E mutation in all the patients (p=0.343) and separated Asian patients (p=0.08) (Figure 4D and E). Graphing the positive probability of the three thyroid groups, including patients in benign cytology, patients in C-TIRADS 3–4a of benign cytology, and thyroid cancer in TCGA Asian patients, we found that the probability of BRAF^V600E mutations was significantly higher in thyroid cancer patients than in other two groups, regardless of age. In all three groups, the probability of positive BRAF^V600E mutations in the thyroid significantly increased with age over 50 years (Figure 4F). Therefore, greater attention should be paid to patients older than 50 years in benign cytological results, and BRAF^V600E testing should be conducted under appropriate conditions for screening high risk of malignancy, followed by surgical treatment.

Figure 4 The correlation of BRAFV600E mutation status and patient age. (A–E) Comparison of patient age between positive and negative BRAFV600E mutation (A) in benign cytology TBSRTC II, (B) in C-TIRADS 3–4a with benign cytology, (C) in all the patients of our cohort, (D) in TCGA-THCA cohort, and (E) in TCGA-THCA-Asian cohort (E). (F) Liner correlation between patient age and BRAFV600E mutation rate in different cohorts. Green: TCGA-THCA Asian cohort, red: FHPT cohort in TBSRTC II, blue: FHPT cohort in C-TIRADS 3–4a with TBSRTC II.

The Molecular Mechanisms Underlying the Increased Frequency of BRAF^V600E Mutations in Older Patients

Differential gene expression analysis was performed between patient age groups (≤50 years vs >50 years) in TCGA-THCA-Asian cohort, 277 differentially expressed genes were identified, which were significantly enriched in Ras signaling pathway (Figure 5A and B), closing interconnected with BRAF mutation. 1174 genes between the BRAF^V600E mutation groups were also identified (Figure 5C). The intersection differential genes of BRAF^V600E mutations and age were significantly enriched in immune-related signaling pathways (Figure 5D). Therefore, we hypothesized that the association between age and BRAF^V600E mutations may be immune-related. Furthermore, we investigated the correlation of immune response with patients age and BRAF^V600E mutation, we demonstrated the majority of immunostimulator were down regulated in older patients and patients with positive BRAF^V600E mutation, BTNL2, CXCR4, TNFRSF14 and TNFRSF25 were significant higher in younger patients and CXCR4 was significant higher in patients with negative BRAF^V600E mutation (Figure 5E). No statistic differences in immunoinhibitors were obtained between patients age groups (≤50 years vs >50 years) (Supplementary Figure 3).

Figure 5 The association in immune response with BRAFV600E and patients age. (A) The volcano plot of differentially expressed genes (DEGs) between older and younger patients in TCGA-THCA-Asian, younger: ≤50 years vs.older: >50 years. (B) KEGG analysis of DEGs between younger and older patients in TCGA-THCA-Asian. (C) The volcano plot of DEGs between patients with positive and negative BRAFV600E mutation in TCGA-THCA-Asian. (D) KEGG analysis intersection DEGs between BRAFV600E mutations groups and age groups. (E) The proportion of immunosimulators in younger and older patients, positive and negative BRAFV600E mutation in TCGA-THCA-Asian. Younger: ≤50 years vs.older: >50 years *p < 0.05, **p < 0.01.

Discussion

Accurately distinguishing between benign and malignant thyroid nodules is crucial to prevent patient overtreatment and tailor personalized treatment strategies. High-risk malignant thyroid nodules typically require surgical excision. Regarding surgical approaches, studies suggest no significant difference in early postoperative complications between total thyroidectomy (TT) and subtotal thyroidectomy (STT).²⁵ Similarly, LigaSure vessel (LS) or harmonic scalpel (HS) exhibit comparable safety in thyroidectomy.²⁶ For low-risk malignant thyroid nodules, repetitive ultrasound examinations and follow-ups can mitigate the discomfort associated with surgery. Ultrasound examination, widely utilized for assessing thyroid nodules, remains the most prevalent method. A study by Grimmichova et al indicates an 18% malignancy rate in thyroid nodules classified as low-risk by ultrasound.²⁷ In China, the C-TIRADS is extensively employed for thyroid ultrasound classification, utilizing a counting method based on suspicious malignant features. Nodules classified as 4b-6 are recommended for FNA testing.^4,28 FNA has traditionally been considered the gold standard diagnostic method for thyroid nodules before surgery, demonstrating high sensitivity and specificity in diagnosing thyroid malignancy. However, in cases of indeterminate FNA results, both sensitivity and specificity for malignancy significantly decrease.²⁷ In Mulita et al studies, the malignant of thyroid nodules that fall within Bethesda categories II and III–IV was 1.58% and 15–40%, respectively.^7,29 The BRAF^V600E mutation detection has proved can be used as a marker in FNA evaluation to diagnosis malignant in thyroid nodules.^30,31

In our study, 46.2% of thyroid nodules classified as C-TIRADS 3–4a and 17.9% categorized as BSRTC II were diagnosed as PTC, a proportion significantly higher than previously reported data. This discrepancy may be attributed to our comprehensive integration of ultrasound, FNA, and BRAF^V600E mutation detection results and selected high-risk thyroid nodules for surgery and pathological diagnosis, contributing to a notable increase in the overall malignancy rate. Our finding also revealed the sensitivity and specificity of C-TIRADS and TBSRTC has significant improved when combined with BRAF^V600E mutation detection, and the combination of BRAF^V600E mutation and TBSRTC categories had optimal diagnostic performance of thyroid nodules, with the AUC, sensitivity, and specificity being 0.923, 84.6%, and 100%, respectively. A large number of studies have reported that BRAF^V600E mutation was significantly associated with PTC or papillary thyroid microcarcinoma (PTMC),^13,32,33 and the specificity of BRAF^V600E mutation for PTC was nearly 100%.³⁴ Consisted with the report, in pathological diagnosed group, all the patients with positive BRAF^V600E mutation were confirmed malignant, including those belonged to low-risk cytological categories TBSRTC II and C-TIRADS 3–4a.

Positive BRAF^V600E mutation could provide prognostic information for thyroid cancer, showed significantly higher risk of mortality and recurrence.^35,36 In addition, the BRAF^V600E mutation has an additive effect with various clinical features, such as older age, male gender, and lymph node metastasis, significantly increasing the risk of mortality in PTC.^14,37,38 Actually, positive BRAF^V600E mutation was significantly associated with high-risk FNA and US classification, along with advanced pathological T stage and lymph node metastases in thyroid cancer in our study. Thus, the thyroid nodules with positive BRAF^V600E mutation should take more attention for repeating US examination to make the final decision on whether to proceed with surgical intervention and remove thyroid nodules.

Few studies have focused on BRAF^V600E mutation analysis in thyroid nodules with benign cytology, and all three literature reports we retrieved have confirmed that the majority thyroid nodules in benign cytology with positive BRAF^V600E mutation confirmed malignant.^17–19 Consisted with the reports, three patients in TBSRTC II with positive BRAF^V600E mutation confirmed malignant by pathological diagnosis, confirming that there is a malignant risk even in the low-risk category of BSRTC II, and the BRAF^V600E mutation can identify it. However, for thyroid nodules with benign cytologic evaluation, in order to avoid overdiagnosis and overtreatment, the BRAF^V600E mutation detection was un-practical and not required immediately according to the 2015 ATA guideline.²⁴ Using clinical and ultrasound features to identify positive BRAF^V600E mutation, and including the BRAF^V600E mutation status in the prognostic risk stratification can lead to more precise management of patients with low-risk cytologic category.

Recently, a study revealed that more than two suspicious US features in benign cytologic analysis should consider further treatment of BRAF^V600E mutation detection.¹⁷ Several US features, such as microcalcification, marked hypoechogenicity and irregular margins were more frequently observed in thyroid nodules harbouring positive BRAF^V600E mutation than negative BRAF^V600E mutation.^30,39 In our study, there were no significant differences in C-TIRADS classification, along with US features between positive and negative BRAF^V600E mutation in benign cytologic results. The reason may be that the majority of cases were classified as 3–4a, so the result of C-TIRADS classification was not referential. Surprisingly, the positive BRAF^V600E mutation was increased in older patients in benign cytology and in the nodules classified both C-TIRADS 3–4a and TBSRTC II.

For thyroid cancer, The American Joint Committee on Cancer (AJCC) staging system has emphasized that the age as the part of progression for malignant, treated as a dichotomous variable into risk stratification and treatment,^40,41 and the mortality risk of PTC was increased as increasing of patients age.⁴² A study has reported that in patients older ≥60 years, the papillary microcarcinomas (PMCs) were most likely to enlarge or show clinical node metastases.⁴³ Shen et al has reported that the patient age exhibits a linear correlation with mortality rates in individuals with the positive BRAF^V600E mutation, conversely, among patients with the negative BRAF^V600E mutation, mortality rates remain relatively low and stable with increasing age.³⁷ In our study, we observed a significant upregulation of the BRAF^V600E mutation in patients aged over 50, particularly in the context of low-risk FNA and US diagnoses. Specifically, when dealing with thyroid nodules exhibiting lower suspicious US features and negative FNA findings, older patients warrant heightened attention for additional BRAF^V600E testing and further surgery.

We identified the different expression genes between two age groups (≤50 years vs >50 years) were significantly enriched in Ras signaling pathway, which are closely interconnected with BRAF mutation.^44,45 The signaling pathway enriched in the different genes intersection of age groups and BRAF^V600E mutation played a crucial role in modulating immune response. Previous study has shown that in thyroid cancer, BRAF^V600E mutation exhibits a largely immunosuppressive profile and demonstrates disruptions in the immune surveillance.⁴⁶ Consisted with the study, our results showed a decrease expression of most immunostimulators in the patients with positive BRAF^V600E mutation and older age. However, immunoinhibitors had no significant difference between patients age groups. More samples are necessary for further analysis.

There are several limitations of this study. Firstly, the malignancy rate in our study was higher than other studies before, the reason may be that the nodules with the pathological diagnosis in our study were at high risk of malignant, such as higher TBSRTC category, more suspicious US features, and positive BRAF^V600E mutation, which made the proportion of malignant was high. Additionally, all the thyroid nodules were followed up by regular clinic visits for half of the year, which is shorter and inadequate for detecting the progression. On the other hand, for the correlation of BRAF^V600E mutation and patient age, the results were obtained by Asian patients in TCGA-THCA, the number of samples was limited and there was no clinical data for analysis and validation. Subsequently, more prospective data were required for a deeper analysis. Otherwise, this is a retrospective study of single-institution analysis, we need more cases and prospective studies to verify our results in future studies.

Conclusions

The BRAF^V600E mutation serves as a risk factor in screening malignancy in thyroid nodules, and its combination with TBSRTC classification can enhance diagnostic sensitivity. In cases of TBSRTC II and C-TIRADS 3–4a in benign cytology, thyroid nodules with the positive BRAF^V600E mutation exhibit an increased malignant risk. Additionally, the proportion of patients aged 50 and above with the BRAF^V600E mutation is significantly higher than those below 50 years old. The new results may help to make a better decision to screen thyroid nodules in low-risk FNA and US diagnostic results to have further BRAF^V600E mutation detection, which have higher risk of malignancy, and further surgery should be considered.